Moral realists: Denying you're making empirical claims isn't going to help

Photo by Jordi Orts Segalés on Unsplash

(This was written quickly and without proofreading, so please let me know if you spot any errors).

1.0 A comment and a block

Christian Gonzalez recently posted a comment on my debate with David Enoch. Here is the comment in full:

this debate on moral realism basically comes down to whether you have any intuitions at all about irreducibly normative truths existing. if you don’t then there’s no way to get a hold on someone into the non-naturalist realist camp. but this makes me wonder what evidence would even be possible to give a radical empiricist that irreducible normative or moral truths exist. i suspect none is possible and that they’re “deliberatively out of reach”

which reminds me of this phenomenon i see happen with anti-realists whereby after spending enough time in the meta-ethics mines, their commitments to a reductionist, empiricist worldview force the realist intuitions out of them. but that’s not how most people start out. most people, without having their intuitions altered by huffing too much metaethics papers would say “yes killing jews is really wrong” in an irreducibly realist sense. if anti-realists continue to spread their message, i suspect the masses may be deliberatively out of reach in time

I left several comments. The central point of these comments was that Gonzalez was making an unsubstantiated empirical claim.

Unfortunately, Gonzalez has blocked me, so all of these comments were deleted. I don’t know why I was blocked since, to my knowledge, I was not rude in my responses, but simply critical. But it’s anyone’s prerogative to block whoever they want. However, Gonzalez made several remarks that are central to my primary criticisms of contemporary advocacy for moral realism, so I find it important to reiterate and add to my reactions to this comment here.

2.0 Are some antirealists deliberatively out of reach?

First, Gonzalez suggests that if you lack intuitions about irreducibly normative truths existing, there may be no way to get a hold of you. Gonzalez then asks:

but this makes me wonder what evidence would even be possible to give a radical empiricist that irreducible normative or moral truths exist. i suspect none is possible and that they’re “deliberatively out of reach”

I do think some of us are deliberatively out of reach, at least in a conditional way. If a person holds the correct position on a given matter and if they are reasoning well, then they shouldn’t be persuaded to hold a view to the contrary, unless something has gone wrong (i.e., unless they’re not reasoning well, which is inconsistent with the stipulation, anyway, so arguably this is tautologous). So one explanation for why some of us antirealists are deliberatively out of reach is that we’re reasoning well and have arrived at the truth. Gonzalez mocks me here:

What is there to mock? Let me illustrate why mocking this is silly with an example. Suppose that I am looking out of my window. I have good vision, I am well positioned to see what’s there, and I see that there’s…nothing. Now along comes a person who insists there is a tree in my yard. Suppose there is no tree. And suppose I am not insane, am not induced to hallucinate, continue to have well-functioning vision, and so on. Am I deliberatively out of reach with respect to being convinced there’s a tree there?

Yes, I am. And the reason I’m out of reach is because there’s no tree there. The only way for me to come to believe there is a tree there is if I were to make some kind of mistake. So one reason a person can be deliberatively out of reach is if they’re correct and are not making any mistakes.

But how one assesses this remark will turn in part on what’s meant by “evidence.” I wouldn’t demand empirical evidence for truths that are, by their very nature, not empirical. If “evidence” is construed more broadly, then I suppose it would turn on whatever epistemic framework could lead one to the true belief that there are irreducibly normative facts. I don’t know what that would look like, but that isn’t a problem for me, because I think this line of inquiry is premature. Before we can even consider what could serve as evidence for something, we’d first need a clear sense of what the thing is, such that we could assess what could or couldn’t serve as evidence for it in principle. Since I think the notion of irreducible normativity is meaningless, I don’t think it’s possible in principle for there to be any evidence for it, even broadly construed.

So to convince me, at least, one would first have to convince me that I’m mistaken about my more fundamental views about language, meaning, and metaphilosophy. I’d need to be convinced that, e.g., philosophical intuitions are real, that they are a reliable way of determining what’s true, and either that irreducible normativity is meaningful in the respects that I think it isn’t, or that my views about what constitutes meaningfulness are mistaken, and convince me to adopt an alternative view of meaning on which the concept in question is meaningful, then I’d need to somehow obtain, or “have” the concept.

That, at least, is a rough sketch of what would be needed to convince me. I can’t speak for other antirealists, but I am confident many of them think differently than I do so the same conditions wouldn’t apply.

3.0 Native antirealism

Gonzalez continues:

which reminds me of this phenomenon i see happen with anti-realists whereby after spending enough time in the meta-ethics mines, their commitments to a reductionist, empiricist worldview force the realist intuitions out of them.

This is an anecdotal claim, but I’ll take his word for it that he’s observed this at least once. I’ll also bracket my objections to the notion that people have “intuitions.” There are some construals of this claim on which I’d deny anyone has ever had intuitions, while on others I’d grant that they do. Let’s grant for the sake of argument that some acceptable conception of intuition is in play (even if this isn’t actually the case).

However, it’s still an empirical claim. Gonzalez is outlining a causal process and a mechanism where people move from one psychological state (having realist intuitions) to another state (not having realist intuitions) via the process of studying metaethics and possessing (or developing) a reductionist, empiricist worldview. And Gonzalez’s subsequent remarks strongly allude to extrapolating from this anecdote to more general claims about human psychology.

How often does the process Gonzalez describe happen? Who knows. But I am confident we can find at least one, and probably quite a few instances of this among contemporary philosophers or people who’ve studied philosophy. An important question, though, is how they obtained those realist intuitions in the first place. I don’t think we have good evidence most people had realist intuitions prior to studying philosophy. Instead, a more defensible trajectory is that studying philosophy first causes people to acquire realist intuitions, then, for some people, they lose them at some later point.

Speaking for myself, it’s possible I once had realist intuitions but I can no longer recall if I ever had them. As far as I recall having any explicit position at all, I’ve held an antirealist view. As far as I know, studying philosophy did not cause me to lose realist intuitions. If so, I would appear to be a native to antirealism.

4.0 Huffing too much metaethics

The important part of Gonzalez’s remarks follow from the previous comment. After stating that he has witnessed people lose their realist intuitions by studying metaethics, Gonzalez continues:

but that’s not how most people start out. most people, without having their intuitions altered by huffing too much metaethics papers would say “yes killing jews is really wrong” in an irreducibly realist sense.

This is an empirical claim about human psychology. Note two key phrases here:

• most people

• would say

Gonzalez has presented an empirical hypothesis about human behavior. Specifically, a statistical claim that more than fifty percent of the population in question (“people”) would say a certain thing, presumably under some counterfactual condition in which they were prompted with some kind of stimulus that one predicted would elicit the response. This latter presumption is based on the use of the term “would.”

Statistical claims about human behavior are not only empirical claims, they are paradigmatic empirical claims. What Gonzalez has said here could be used in a psychology textbook as an example of an empirical hypothesis.

And yet there is no good evidence this claim is true at all. There is a relatively new and growing literature in experimental metaethics that assesses whether nonphilosophers are moral realists, have realist intuitions, speak or think like realists, and so on. To my knowledge, there are no studies that directly assess whether people think morality is irreducibly normative. Nevertheless, we don’t require specific, directed studies to make informed judgments about what is or isn’t the case, empirically speaking (some people mistakenly think that in order to justify any empirical claim, you must have a specific study addressing that claim. This is absurd and not worth considering further).

Gonzalez, like many other realists, insists on making sweeping empirical claims, but is unable or unwilling to support those empirical claims by appealing to (or even gathering) empirical evidence.

When I criticize realists in this way, some will push back by appealing to the empirical literature. This is rare, but I do appreciate it when they do. This is not what Gonzalez did. Gonzalez posted a critical response to me, then while I was writing a response to it, he blocked me.

This is a common experience for me. Realists make empirical claims. I criticize them for doing so. And then, instead of acknowledging that they’re making empirical claims and either retracting those claims as unsubstantiating or substantiating those claims, they:

• Ignore me

• Deny the claims they’re making are empirical

Impressively, Gonzalez has opted to do both.

5.0 First they ignore you, then they deny their claims are empirical

I was in the middle of writing a response to this post when I was blocked, and this prompted me to write an article about this instead. This was a direct response to my original note where I made a shortened version of the some of the remarks above, most critically that Gonzalez’s claim about what “most people” would “say” is an empirical claim. Thus, this is clearly intended as a critical response to that claim. According to Gonzalez:

To think phenomenology submits to empirical study is to misunderstand the nature of perspectival knowledge. Self-reporting is a horrendous instrument for getting at what people actually experience, because most people don’t see their own experience clearly. Asking the masses about the structure of their own consciousness is asking for an unreliable account. There are things that talented meditators know about the structure of consciousness that lay people do not. And, crucially, there is no empirical analysis to know who is who. The general public can be, and in fact are, deluded about what they experience. Once one Sees a structural feature of consciousness for oneself (that we perceive normative reasons to act agnostic of our desires, for instance) its truth is self-evident, and one can be confident it holds universally for any consciousness similarly structured.

Can you prove this to other people? No! And that’s precisely the epistemic gap between first-person and third-person truths. Can you be wrong about what seems self-evident? Of course! And yet self-evident first-person truths about consciousness exist nonetheless. These are simply the cards we’re dealt when doing research into the nature of consciousness.

Gonzalez seems to reject my claim that he was making an empirical claim on the grounds that phenomenology isn’t subject to empirical scrutiny, and because self-report is a bad method for figuring out what people’s experiences are like.

This is confused in a number of ways.

First, facts about what people experience are psychological phenomena. If one wants to claim they’re somehow empirically inaccessible, they’re welcome to make a case for this, but this would be irrelevant. Christian made a claim about what most people would say. Not only is this an empirical question because it’s a statistical claim about human behavior, how is that not the sort of thing you’d study using self-report? It’s a claim about what people would say!

Setting aside that obvious and massive problem, even if this somehow wasn’t an empirical claim, but a claim about what they’d “experience” in some empirically inaccessible way, Gonzalez is going to face a serious problem: even if the content of any given person’s experience included elements that could not be empirically evaluated, statistical claims like “most people” have these experiences are still empirical claims that can only be assessed empirically. Here’s why.

Gonzalez draws a distinction between a kind of perspectival knowledge, which is a kind of first-person knowledge that isn’t amenable to the third-personal nature of scientific tools. The problem is that first-person knowledge of the sort Gonzalez describes only gives you access to your own experiences, it does not give you access to other people’s experiences. As such, even if you can obtain first-person knowledge, and even if it isn’t subject to empirical scrutiny, what you cannot do is obtain first-person knowledge of the content of other people’s first-person knowledge, which is exactly what you’d need to do to confirm that “most people” have the exact same experiences as you. If one insisted you could instead use third-personal knowledge to do so: great, then it’s an empirical claim. So Gonzalez faces a dilemma:

(1) If Gonzalez’s claim about what “most people” would “say” isn’t an empirical claim, then, because it’s a third-personal claim about the content of other people’s first-person experiences, it isn’t a claim that one can obtain knowledge about through first-personal means, and thus Gonzalez wouldn’t be able to justify this claim. In short, neither Gonzalez nor anyone else can obtain knowledge of other people’s first-personal experiences via exclusively their own first-personal experience

(2) This is, in fact, an empirical question, and either our empirical tools aren’t good enough to settle the matter (in which case Gonzalez wouldn’t be justified in making the claim in the first place), or they are, in which case Gonzalez has made an unsubstantiated empirical claim.

In short, Gonzalez makes a very serious conflation. Gonzalez mistakenly reacted to my critique as if I were claiming that first-person knowledge can be studied empirically. While I think it can be, this is irrelevant, because that wasn’t even what I was doing. What I was doing is saying that Gonzalez made a third-personal claim: a statistical claim about what most people would say.

I think Gonzalez may be very mixed up here, because I suspect they take what most people would say in these cases to reflect some kind of ineffable, first-person experience they have, and the content of which cannot be confirmed or disconfirmed empirically. But even if this is what Gonzalez was doing, then he’s simply entangled a bunch of first-person claims in a broader, third-personal empirical claim about the statistical regularities of the occurrences of certain first-person experiences. In other words, Gonzalez has nested first-person claims inside a third-person, empirical claim, then wants to treat the latter as inheriting the epistemic qualities of the former. But that’s not how things work.

For comparison, suppose there is a sui generis experience associated with tasting chocolate. And suppose the content of that experience cannot be assessed empirically. Okay. Fine. Then there’s a certain kind of knowledge, knowledge of what it’s like to taste chocolate, that cannot be studied empirically. But if one now wants to say that:

Most people would say that they have this specific experience when tasting chocolate, and are, in fact, experiencing it in this sui generis way

One is now mixing first-person claims with third-person empirical claims. Even if first-person knowledge could enable you to know what it’s like to taste chocolate, it cannot, by itself, provide you with knowledge that most people have the same experience when they taste chocolate, nor can first-person knowledge tell you anything about what they’d say.

First-person knowledge simply can’t tell you anything about other people’s experiences. So Christian could say that they have knowledge of P, and I could say I have knowledge of Q, and so on. This kind of ineffable, first-personal access to one’s own experience doesn’t and can’t justify population level claims about other people’s experiences.

In other words, if certain kinds of knowledge are fundamentally private, and hinge on one’s own personal experience, and if what Christian intended to say all along was that most people have the private experience of experiencing morality as “irreducibly realist,” this latter claim isn’t one that Christian or anyone else could possibly be justified in making by appealing to their own first person experience, for the simple reason that it’s a statistical claim about other people’s experiences. So even if we had private, ineffable experiences the content of which cannot itself be studied, any claims that most of us have a given kind of experience are themselves empirical claims. If Christian wants to insist such claims cannot be assessed empirically, either, that’s fine: then we can’t assess the truth of these claims at all, because they’re empirical claims and if empirical methods can’t settle the matter, nothing can. Introspection can’t tell you what other people’s experiences are like.

Furthermore, Christian, like many critics of empirical psychology, is under the mistaken impression that the only empirical tools available to us are self-report. This is false. One must have a profoundly impoverished understanding of psychology to think that we exclusively rely on self-report, or that this is the only method available for evaluating how people think. Christian continues:

Asking the masses about the structure of consciousness is like consulting a random pedestrian instead of a seasoned meditator. They can be, and in fact are, deluded about what they undergo. Once one Sees a structural feature of consciousness for oneself (that we perceive stance-independent normative reasons to act, for instance) its truth is self-evident, and one can be confident it holds universally for any consciousness similarly structured.

Again, the only methods available don’t involve merely asking about questions and awaiting self-report. And this latter claim is extremely dubious: people who have introspected on consciousness arrive at different conclusions on a regular basis; claims of the truth being self-evident are highly questionable and note that no arguments or reasons are given for thinking that they are; Christian simply asserts that this is the case.

6.0 Source???

Christian’s reaction to me exhibits a pattern I’m becoming too familiar with. Realists often do the following:

1. Make highly contestable empirical claims

2. Are unable or unwilling to furnish any empirical evidence for those claims

3. Make a bunch of excuses for why they don’t have to, e.g., because the claims in question are “obvious,” or, in this case, because they somehow aren’t empirical claims in the first place

4. Disengages from me when I press them on the point (in this case, by blocking me, and, consequently, deleting all of my comments)

Another thing many realists do is follow up on Step #4 by mocking me. Gonzalez opted to do this as well, with this unpleasant meme:

I’m unironically with the guy on the right.

Realists can’t have their cake and eat it, too. If you want to say most people are moral realists, or have realist phenomenology, or would say things that sound like they endorse or are committed to realism, you are doing psychology.

You can either acknowledge this, and try to do psychology well, or you can deny this, and continue to do bad psychology.