Sunday, March 29, 2009

Mackie and Kevin and Ben...Oh My! (A re-defense of moral subjectivism)

I have been recently, and pleasantly, surprised by a recent, lengthy post on my friend Ben Hayek's blog with a main subject (Is there a better subject for an extended essay, I wonder?) Specifically, Ben takes issue with several posts I've made here defending moral subjectivism - the view that moral judgments are subjective preferences rather than objective facts.

Like many objectivists, Ben is quite angered by subjectivism because, without it, we could not make objectively valid moral judgments. The first part of his post suggests a moral dilemma: Ben and I confront a person torturing a cat. Ben, the objectivist, would be able to very objectively condemn the torturer, while I would not.

On Kevin's view, when a subject declares that "torturing kittens for pleasure is immoral," the subject is not identifying a a fact about the world (i.e., not picking out a constituent of objective reality), but rather identifying only a state within the subject's consciousness. Kevin describes this state within the subject as no different to whenever a human being feels an emotion. Thus, if I encountered a sadist on the street about to torture kittens for the pleasure it gives him, if I rescued the kittens by force prior to their fate, I would not be doing so because it is a matter of objective fact that what the kittens were about to be subjected to is objectively wrong, but because I only feel that it would be wrong.

I suppose I am guilty as charged here. Ben is right that I agree with him that torturing kittens is wrong, but that I disagree with him about the objective import of that view. I would, in a sense, be powerless to objectively condemn the sadist.

I am quite at a loss, though, to think about why objectively condemning a sadist would have any more import than subjectively condemning her. Does the sadist care if my condemnation is coming from me or from the Cosmic Objective Order? I doubt it. I doubt a sadist who tortures kittens feels very bound by a moral excoriation - no matter whether it is subjective or objective in nature (whatever the hell the difference looks like.)

And this is one of the big objections I have to moral objectivism; like JL Mackie and others have noted, if morals exist as objective things, they are very strange things - things like no other things we know of. They do not have any physical embodiment or mode of detection (other than an introspective method identical to that used to ascertain subjective emotion!). And they are the only things (assuming they are things) that impose obligations on witnesses' behavior. Facts are real, but I can think of no fact that obligates us to change our behavior like morals do. So, are they facts? I cannot see any good reason to see that they are.

But apparently, Ben does. He suggests that they are facts just as 2+2=4 is a fact. He suggests that just as this mathematical axiom has no basis in observable sense data (but remains a fact nonetheless), so with moral facts. 2+2=4, killing kittens is wrong... pretty similar, eh?

Actually not. 2+2=4 actually does have basis in physical fact and is observable (verifiable, if you will). As numbers are symbols, "2" means ** and "4" means ****. We can verify, then, by using our number system that 2+2=4 is valid by adding ** to ** and seeing whether it equals **** (the representation for "4").

But Ben has upped that ante. He suggests that if I hold that morals are not facts because they are not grounded in the world of observable and testable things, then I have a problem: that very statement (that morals are not facts because not grounded in observable and testable things) is itself not grounded in the observable, testable world. Oh dear!

Of course, Ben's objection misunderstands the burden of proof. Just as he who claims that faeries do not exist does not bear the burden of proof (the affirmer of faeries does), my statement is something of a null hypothesis equivalent to a denial of faeries existence. Just as the faery denier is waiting for proof of so-called objective faeries, I am waiting for proof of so-called objective morals.

And, yes, there is reason to doubt such existence. While Ben is right to say that the fact of moral disagreement is not enough to make the case for moral subjectivism, it does not itself bode well for moral objectivism. Moral disagreement is not just existent, but quite pervasive and often intractable. When two people disagree over whether capitalism is wrong, for instance, they can argue facts (capitalism leads to x, socialism does not lead to x), but the argument often comes down to an impasse (I think x is wrong, and you don't), where one cannot think of any fact-based way to convince the other.

This is not very explainable under moral objectivism other than to say, as Ben does, that " people in general are stupid." This will not do, though, because he who makes that charges must be able to point to - POINT TO - what it is that the dummies are overlooking (and if that something is intuition, I am sure the dummies also have those, in which case, the accuser must argue as to why his is superior to theirs non-circularly).

Lastly, Ben wonders what my criteria for truth is. Unfortunately, this is an area of philosophy I have little patience for. What we mean by "true" appears to be a tautology: "truth" is "that which accurately describes how things are." How do we recognize truth? The only good answer to this I see is: by experiment. If I write out the directions to Sean's house, the best way to test this is to see if the directions actually get us there. If I say that x theory is true in science, the way to test that (unlike what ID creationists do) is to experiment the theory, and try and falsify the theory. If I say, "I love my Fiancee," the only - and imperfect - way to see if that is true is to observe me and see if my actions are consistent or inconsistent with that statement.

But how in the world do we verify "killing cats is wrong." We can verify "killing cats will cause them physical pain." We can verify "I have an aversion to seeing cats inflicted with physical pain." Those are all matters up to perfect or imperfect experiment. But what we cannot verify is the next step: "imposing physical pain is wrong." What does "wrong" refer to? We can't observe it. We can only intuit it (just as we intuit our own subjective emotions, or God talking to us.)

Ben is clearly not satisfied with the idea that to be called "truth" a thing has to be observable. He hints at logic being a falsifier of this claim. I think not. Logic is only so good as the results it leads to, and the only reason we call the law of non-contradiction true is because we test it both empirically and mentally. After all, logic is seen as a tool, and the only truth a tool has is that it does what it says it will do. When we use logic, it leads us to results that work as they say they will (as theories are tested by seeing if they work the way they say they will.)

And as for 2+2=4, it is the same. Its truth comes not from a heavenly mystical Platonic realm that humans cannot see (but can intuit) but from the fact that we can observe ** plus ** equalling ****. So, I am afraid, I hold to the horrible idea that "truth" gets its name from its ability to work in practice.

In the meantime, I'm going to keep on believing concepts that I believe exist in the world, albeit in a transcendental sense of course, such as mathematical truth, logical truth, and moral truth, just like I believe in the thoroughly objective nature of truth in general, along with the believe that they are - all of them - constituents of the world.

Many men intuit God and argue very fervently that He "exist[s] in the world, albeit in a transcendental sense," and that, despite his objective nature, we can't sense him with the five senses. We can only intuit him. Ben has made the case that ethics are like this. But is he persuaded by such religious arguments. (Hint: Ben is not a Christian because, despite God having all the properties Ben argues that morals have, he does not believe that there is evidence for God's existence.)


1 comment:

  1. As usual, my friend (and scholar) Kevin Currie makes some very interesting points in response to my post arguing against the plausibility of the notion known as “moral subjectivism,” the view that when a subject utters a moral proposition regarding act x, all that one is really doing is identifying one’s subjective preference towards act x. Instead, to use Kevin’s terminology, “we may FEEL or THINK certain things to be right and good, but we are in error if we thing there is any objective reason compelling others to see it the same way.” In other words, Kevin argues, the purely phenomenological aspects of morality deceive us in some sort of fundamental way into (mis)believing in “objective moral values,” but, in fact, “moral objectivism” is false because “there really is no evidence” of it. There are a few assumptions here at work that are worth spending some time on.

    Consider the “matter” we are discussing here, namely, whether moral objectivism is true or whether moral subjectivism is true. Kevin’s first assumption is that there is a truth of the matter, that is, one of these view is true and the other false. So Kevin presupposes some kind of theory of truth in his argument. Second, Kevin acknowledges, at some level, the phenomenological fact of the matter, namely, that moral objectivism at least appears or feels like it is true (but isn’t). Third, Kevin acknowledges the work that evidence has in his moral theory, and whatever it is, the phenomenology involved either isn’t sufficient evidence or isn’t evidence at all. And this leads to Kevin’s fourth assumption, which is the position that the evidence one has for any given proposition must be in some way scientifically verifiable, that is, based on some empirical facts from which we may infer conclusions.

    We can therefore work Kevin’s argument backwards to understand where he is coming from. Without any empirical facts of any given matter, there is no way to verify the “evidence” one offers for believing any given proposition. Appeals to phenomenology (e.g., “ice is cold”) are not appeals to empirical facts because phenomenological properties aren’t “out there, in the world, but exist only in minds.” Moral claims are no different from phenomenological claims, that is, moral claims are not empirical claims. Therefore, the global claim that “objective morality is true” is false. With these preliminaries in mind, we can now turn to Kevin’s objections.

    The Kitten Torturing Sadist scenario is designed to put Kevin on the defensive about what FEELS like a clear-cut case of immorality in the strong, objective sense. Kevin’s first objection is, I think, that there is no difference between objectively condemning the sadist for its acts and subjectively condemning the sadist for its acts. In other words, Kevin appears to be arguing here that the entire objective vs. subjective controversy in meta-ethics is over a distinction without a difference. If this is true, then I’m uncertain as to why either one of us would be taking time out of our busy schedules arguing for one position over the other. The second objection is that the sadist probably wouldn’t care too much about the controversy anyway – it is, after all, a sadist we’re talking about. To be candid, I’m not certain where this objection goes either, as it is the conduct on the sadist we’re focusing on here, not what the sadist thinks about his conduct, or what the sadist thinks about us thinking about his conduct.

    In the next paragraph, however, we get a little bit more from Kevin regarding why the notion of objective moral truths are such exotic, “very strange things – things like no other things we know of. They do not have any physical embodiment or mode of detection (other than an introspective method identical to that use to ascertain subjective emotion!).” There are three objections are work here.

    The first is that moral values are unique, sui generis concepts. I agree. But from here, what follows? There are plenty of unique things in the world, such as the concept of truth. Another is the mysterious thing we can “consciousness.” From the fact that these terms denote unique, sui generis concepts we may infer only that we, as human beings, somehow have the ability to think and speak of such things, and, perhaps even more strangely, people with whom we are speaking know exactly what we’re talking about when we use these concepts. The second is that moral values do not have any “physical embodiment.” In one sense this is obviously true – moral values are not subject to physical laws or are extended in space. But in another sense, moral values certainly cause certain brain-states, which exist in brains and can be measured scientifically. (I want to stay away from the mind-body problem here, however, and so will respectfully beg off this point.)

    The third, however, is the most interesting objection, that we can “detect” (i.e., be aware of) moral values only by way of “an introspective method identical to that used to ascertain subjective emotion.” In other words, Kevin concedes that we, as human beings, can intuit moral values, but that this form of intuition is no different than when we are aware of “subjective emotions.” Two comments here. The first, I think anyway, is obvious, and it is that moral awareness can (and must) be distinguished from emotional awareness. Merriam-Webster defines “emotion” as “a. the affective aspect of consciousness: feeling; b. a state of feeling; c. a conscious mental reaction (as anger or fear) subjectively experienced as strong feeling usually directed toward a specific object and typically accompanying by physiological and behavioral changes in the body.” M-W defines a “moral” as “a. of or relating to principles of right and wrong in behavior; b. expressing or teaching a conception of right behavior; c. conforming to a standard of right behavior; d. sanctioned by or operative on one’s conscience or ethical judgment; e. capable of right and wrong action.” Second, in what way are emotions “objective?” And the reason I ask is that Kevin uses the term “subjective emotion.”

    Of course, one can guess as to why Kevin states that moral propositions are no different from emotional propositions – that’s the project of subjectivism. But stating this and showing this are two very different things, and, so far, Kevin can only state the view. But he does have an interesting thought when he states: “[Morals] impose obligations on witnesses’ behavior. Facts are real, but I can think of no fact that obligates us to change our behavior like morals do. So, are [morals] facts? I cannot see any good reason to see that they are.” I think this requires unpacking a bit here, for nowhere have I argued that “morals are facts” – I have only argued that “moral objectivism is true” is a fact, not that morals themselves are in some sense facts, for the following reasons.

    First, I take a “fact” as denoting a nonlinguistic complex that makes any given proposition true if and only if that proposition refers to something in the world that makes it true. See, e.g., Richard Fumerton, Metaphysical and Epistemological Problems of Perception 8 (1985) (where “nonlinguistic” means only that the existence of language is not a logically necessary condition for the fact’s existence). (And the early Wittgenstein referred to a “fact” as a “state of affairs” when he stated that the world consists “of facts, not things.”) Hence, true propositions correspond to “or picture” facts, whereas false propositions fail to correspond. Id. at 55. Applied to my view, then, when I say “that objective moral truths exist is a fact,” what I’m saying is that I’m accurately describing or “picturing” the basic reality that human beings are at least, in principle, capable of detecting moral truths when place in or given a set of circumstances – such as the set of circumstances involving the sadist on the street.

    In other words, my view holds that moral propositions are intuitively discovered in an analogous way to how we discover mathematical and logical truths, where easy ones are easy and hard ones are hard. To this Kevin has a quick reply: mathematical truths actually do refer to physical facts, that is, any given mathematical truth, such as 2+2=4 “actually does have a basis in physical fact and is observable (verifiable, if you will).” And Kevin purports to show this by replacing the symbols “2” with ** and “4” with “****” to show that **+**=****. Without going into any further detail here, however, I will state simply that I do not believe that replacing one symbol for another in a proposition’s language has any impact on the question of whether that proposition is about the physical world or not. Mathematics is mathematics, namely, a priori reasoning par excellence, and no degree of symbol-switching can convert the a priori into the a posteriori.

    The next interesting claim Kevin asserts is related to who has the burden of proof, which he believes I’ve in some sense misunderstood. (I take particular interest in this claim because, in a draft of my initial post, I included a brief discussion of how the moral subjectivist always falls back on notions of whose burden it is to prove the objective nature of morality, while simultaneously constraining, quite arbitrarily, the realm of admissible evidence relative to the inquiry to evidence that rigs the game.) But Kevin makes a very unconvincing move next, namely, by comparing the view that moral objectivism is the intellectual equivalent of “belief in faeries.” Really?

    If we back up to examine Kevin’s earlier statements, he concedes that human beings are in some sense inclined to believe in the objective nature of objective moral truths, they FEEL that matters moral are objective – that is, that moral objectivism bears the advantage of comporting with our phenomenological experience. It seems very odd to say that humans are phenomenologically inclined to favor faerie belief. This objection is unpersuasive for the simple reason that people do NOT FEEL that there are faeries among us. I will have to spend more time on this notion later, but suffice it to say that it seems a bit strange to say that the burden of proof rests on the person advocating for a theory consistent with phenomenology, rather than on the person advocating for a theory that, Kevin admits, does not jive with common experience or common sense. (It reminds me of the radical solipsist joke: “Sure, I’m a solipsist, and frankly, I’m surprised here aren’t more of us.)

    Next, Kevin again points to the existence of moral disagreement, noting that “moral disagreement does not bode well for moral objectivism [because it is] quite pervasive and often intractable,” and again rests his main objection on the claim that “one cannot think of any fact-based way to convince the other” (where by “fact-based” Kevin here means empirically based). But this is the same argument above, that is, that nonempirical claims are invalid because they cannot be verified for lack of evidence.

    This is related to Kevin’s rejection of my claim that moral disagreement is inevitable in a world of nearly seven million people. For Kevin, the burden is on me to point to – POINT TO – evidence regarding why people disagree when there is disagreement. Again, then, if there is no empirical evidence, for Kevin, my claim fails for lack of a means of verification. As we’ve seen, Kevin claims that mathematical reasoning, a priori reasoning par excellence, is empirically verifiable (a claim I find quite revolutionary, actually), so I can’t simply take two people in mathematical disagreement as a way to show anything. But Kevin has very little to say yet about logical truths. Do may logical truths be empirically verified too?

    Assuming the answer is “no,” then I don’t believe that the existence of logical disagreement in any given circumstance points to – POINTS TO – the inference that there is no such thing as the truth of the matter in any given logical disagreement. There will be no way to empirically prove the truth of the matter in such a circumstance, which is nothing other than to say that the disputants are either both wrong or only one of them can be right. Logical objectivism is left intact. (And, of course, it would be bizarre to hear someone claim that they were a “logical subjectivist,” etc.) In the end, however, Kevin appears to lump logic in with mathematics where he claims that logical truths can be empirically verified.

    This last point seems related to Kevin’s rather flippant dismissal of the notion of truth, which seems a bit ironic given the fact that he and I have been arguing over the truth and falsity of a variety of claims. In any event, Kevin claims that the only way we find truth is empirically, out there, in the world – which is another way of saying that ONLY empirical (a posteriori) truths exist. (This is often referred to as radical empiricism, the pragmatist doctrine argued for by one of Kevin’s heroes, William James.) In other words, if there is no way to scientifically verify the truth of the matter asserted, then we’re misapplying the concept of truth to an area where the application of truth is senseless. Hence, Kevin falls back on behaviorism to explain or justify anyone’s claim to love, e.g., his or her spouse, a claim that Kevin – once again running clearly against the grain of phenomenology – argues has no truth-value (lie-detector tests to the contrary, of course).

    In the end, Kevin winds up with the rather fascinating claim that “belief” in objective mathematical truth, “belief” in objective logical truth, and “belief” in objective moral truth is the equivalent of (or at least comparable to) belief in God. As I demonstrated above, however, Kevin has conceded the phenomenological appeal of belief in moral truths that, I think anyway, has no application whatsoever to faeries, and dismissed comparisons to logical and mathematical truths on the ground that both are fundamentally empirical. I, of course, reject the thesis that no truths but empirical truths exist, and strenuously object to Kevin’s characterization of mathematics or logic being empirical sciences.

    In the end, however, it should be obvious that Kevin and I continue to have our major disagreements with the nature of truth and whether the notion has any application outside of the physical sciences, moral objectivity, the nature of evidence, the role of phenomenology, and who bears the burden of proving fundamental claims about reality. I will attempt to address many of these issues in much more detail on my own blog, for I’ve clearly taken too much space over here!