Against Subjective Truth

And for accurate representation of the world

Previously, on Humean Beings:

I argued that academic freedom should not be seen as an inviolable principle. Rather, it should be seen in functional terms. Academic freedom shields individual scholars to allow them to pursue their ideas even if no one in the broader public is buying what they’re selling, but it does so by putting the reputation of the university on the line. This means that, in order to defend the university, scholars have a responsibility to seriously consider whether the ideas that fall under the aegis of academic freedom are genuinely worthy of that protection. And this is how we should understand the report by Boghossian et al, commissioned by Vanderbilt and WashU. It argues that there are a certain set of ideas in the critical humanities that are not worthy of that protection.

I presented this argument to respond to an excellent piece by Derek Anderson, who launches a two-pronged attack against the Boghossian report. He first argues that following through on the report’s recommendations amounts to an attack on academic freedom - this is the point I was keen to address last time. And his second point is that the work done in the critical humanities is defensible, at least on its own terms. These two lines of argument are obviously related (which is why it’s a bit artificial that I’m splitting my response in two), since if the controversial ideas in the humanities are good - or at least good enough - then they are worthy of protection. Still, I did split my response in two, because we’re going to get deep in the weeds today on epistemology, philosophy of language, and philosophical methodology. So if that’s not your cup of tea, you might want to sit this one out. But I’ll try to make this as accessible as possible.

Before diving in, I’ll emphasize an important point that Anderson, Boghossian, and I are all in agreement on, but which many of the reports’ critics have either missed or discounted. Although the report is concerned with the leftward political skew of the contemporary humanities, this concern is only indirect. The main concern is with the quality of the scholarship being done in the humanities. The question at issue is not whether or not the humanities are “too woke,” or anything like that. The question at issue is whether or not the ideas and methods that characterize the contemporary critical humanities are worthy.

Let’s start with two points that Anderson makes that are more or less spot on. The first is that there are worthy ideas being taught in the postmodernist-influenced critical humanities. Even if postmodernist or poststructuralist philosophy is mostly bullshit - and I think it is - I’d have to be mad to say that it’s entirely bullshit. There are some interesting points in there, and some interesting ways of presenting familiar points. I loathe Foucault - he’s the most massively overrated scholar of all time, a sophist par excellence - but I’m happy I’ve read a bit of his work, and I’ve sometimes encountered situations that seem to genuinely resonate with his theories. I don’t think that every student needs to leave the contemporary university with a basic understanding of Foucault - god forbid! - but I equally think that students should be able to learn and think about his work, with the aid of a professor who can point out what is good and what is juvenile about it. To the extent that a house-cleaning in the humanities is called for, it would be insane to try to remove French or German critical theory altogether, although perhaps it would be best if those subjects were taught in a different manner.

Anderson’s second major point, which I (almost) entirely agree with, is that epistemology is to a large extent theory-laden. What that means is that a certain Cartesian model of epistemology - where we begin from an Archimedean point of certain knowledge (or a set of such points) and then derive everything else from that set of certain points using rational deductive inference - is total bunk. The evidence from the world around us strikes our sense organs in the form of radiation or vibrations (etc.), and our minds process that physical data. In the process of so doing, they rely on our previous conceptions about the way the world is and the way the world works. This means that even our perceptions of our immediate environment are influenced, in at least some small ways, by our preconceptions. And when it comes to the much more difficult task of using our knowledge of the immediate environment as given by our senses to obtain knowledge of things that are remote or small or abstract, our reasoning processes are even more influenced by our preconceptions. There is no such thing as a priori access to the One True Epistemology.

In virtue of this, Anderson urges us to see the critical humanists as simply employing a different epistemology. Their methods and theories might seem odd, but when you go through the effort to interpret them charitably and see how they fit together, you’ll find that what they’re offering is consistent (at least according to their own standards). The critical theorists simply have different preconceptions than we do, and so they parse data differently than we do, and thus end up in a different place. In light of this, we should see the conflict between the critical humanities and the rest of the academy as a kind of conflict of two different epistemological systems. Therefore, there is no epistemological basis on which humanities scholarship might be called into doubt. Or rather, to the extent that we do find such a basis, that’s simply because we’re operating within the preconceptions of our own epistemological system. But that makes us no better off than the humanists, who are operating within their preconceptions. We hit an intellectual stalemate. This is why academic freedom requires us to give the critical humanities full protection.

And before I try to take up the other side of this argument, let me again say that this is a really good point. Epistemology is remarkably theory-laden, and Cartesian approaches to epistemology are, on the whole, simply dogmatic. Still, I think that there is a wedge that we can use to push back against some of the excesses of the critical humanities, to show that there really is a problem with many of their theories. This has to do with their theory of truth.

I single out the issue of truth because this is the issue that the Boghossian report focuses on. Again, the Boghossian report’s central concern is not directly political, it’s about the quality of the scholarship. And the central problem for the scholarship is that it accepts a flawed theory of truth: a subjectivist or a relativist theory of truth.

The problem is not so much that the humanists are drawn to a bad theory. The problem is that this bad theory then goes on to infect the methods that are used in the humanities. An analogy from science might be instructive. Suppose a physicist comes up with a new fundamental physical theory that is pretty out there. There’s nothing wrong with this, even if the theory is, ultimately, totally wrong. This kind of divergent thinking is useful for driving forward progress, even if it sometimes fails. But suppose further that our physicist devises some new expensive experimental equipment whose functioning is premised on the truth of his theory - it will work only if the theory is true. We should be hesitant to get him the funding to build his equipment and, if he does build it, skeptical about the results he claims to have obtained by using it. Conclusions arrived at by methods that are premised on dubious theories are themselves quite dubious.

And the charge against scholarship in the critical humanities is that it is dubious in this way. The conclusions arrived at by humanists are politically left-wing, because they’re arrived at using methods that are slanted to arrive at left-wing conclusions, and those methods are justified by appeal to a theory of subjective truth. (See my earlier summary and analysis of the report for the mechanisms by which this is supposed to have have happened.) Knock down the subjectivist theory of truth, and several other dominos will get knocked down with it, and those will end up being some of the most biased and contentious ideas to come out of the critical humanities. Bad ideas produced by bad methods based on bad theory; we’re trying to isolate the root cause. This is a smart thing to do because, to Anderson’s point, there is plenty that is valuable in the critical humanities. We don’t want to blow the whole thing up! (Well, some conservatives want that, for political reasons, but forget them; we’re scholars.) Rather, we’re trying to cut out a tumor, to save the university, and save the parts of the humanities that can be saved.

So now we get down to it: what is a subjectivist theory of truth, and what’s wrong with it?

That’s a hard question. What makes it particularly troublesome is that there’s very little agreement on what a subjectivist theory of truth is even supposed to be in the first place. But let’s work our way there. Sometimes, a distinction is drawn between things being subjectively true or things being objectively true. That is, there are (perhaps) subjective truths and objective truths. The distinction between subjective truth and objective truth lets us identify three prima facie coherent positions: (1) All truth is objective truth, (2) All truth is subjective truth, and (3) There are both subjective truths and objective truths. The subjectivist theory of truth that the Boghossian report targets is (2), and so that’s what I’ll mainly argue against. Although I think there’s also a strong case to be made against (3).

As for what subjective truth is, it is perhaps easier to start by saying what an objective truth is. Objective truth, I take it, is a two-place relation that exists between a proposition and the world. (“Proposition” is something of a philosopher’s term of art, I explain it here in a piece about why the common “fact/opinion” distinction is bullshit, and that piece is relevant to the current discussion, so you might want to take a peak. But if you just understand “proposition” with its ordinary commonsense meaning, you’ll be fine.) Propositions are representational; that is, they represent the world as being a certain way. If the world is the way the proposition represents it as being, the proposition is (objectively) true. Otherwise, it is (objectively) false. So an objective truth is any proposition that accurately represents the world as being a certain way.

Subjective truth, on the other hand, is supposed to be more complicated than this. It is supposed to be either a two-place relation between a proposition and some evaluative framework, or a three-place relation between a proposition, some evaluative framework, and the world. There are lots of different accounts of what an “evaluative framework” might be, but the potential complications needn’t trouble us much here (although they would have to in a more rigorous treatment of the issue). The basic idea is that a proposition is subjectively true if its truth doesn’t (just) depend on the way the world is; it also depends on something else, something variable, something subjective. This is why the subjectivist theory of truth is also sometimes called a relativist theory of truth (and I’ll use those two terms interchangeably). Truth isn’t just a matter of the way the world is, it holds relative to something that we’ll call a framework.

Why accept a subjectivist theory of truth? There are an indefinitely large number of arguments that one might give, but let’s try to simplify this a little by looking at arguments that people do give. Unfortunately, those arguments are generally quite bad. I typically see three different arguments. The first is the argument from the possibility of objectivity. I discussed this in a recent post. This argument says, in short, that it is impossible for anyone to be objective, and therefore no one’s beliefs count as objective truths. The problem with this argument is that it equivocates on the meaning of the term ‘objective.’ Objectivity might be the property of truths described two paragraphs up, which is a matter of accurate representation by propositions. It also might be a property of people, that they have in virtue of being entirely unbiased by any pre-existing beliefs or desires. According to this first argument, no person is perfectly objective in this second sense; therefore, no propositions are objective in the first sense. Put in these simple terms, it’s clear that the argument is bad, although those who make this argument use much more convoluted and obfuscatory language to make it look like a decent argument. It is not. See the linked post for more detail.

The second argument is the argument from the contingency of language. To appreciate this argument, consider the following riddle: “If we called a tail a leg, how many legs would a horse have?” The wrong answer that you might be tricked into giving by the wording of the question is 5. The right answer is 4. If you think the answer is 5, think carefully about what the question is asking. It is not asking “If we called a tail a leg, how many body parts would horses have that are referred to by the word ‘leg’ in the world we’re imagining?” We’re asking how many legs a horse would have in that world. A world where we used language differently would be a world in which people say things like “horses have five legs,” but it would also be a world in which horses in fact have four legs. Unless the change in language causes a change in horse physiology, that is (hard to see why it would!). The world doesn’t change because our language changes (except, trivially, for the parts of the world that have to do with how we use language). The argument from the contingency of language is the argument that comes from thinking that the answer to the riddle really is 5. If we can change how many legs a horse has by changing how we use language, then whether horses have four legs or five depends on the linguistic framework that we use. In one framework, horses have four legs, in another framework, horses have five. And since all propositions are expressed by language, all propositions are subjectively true. But this doesn’t show that the truth of propositions depends on a linguistic framework. It shows that the proposition expressed by a particular sentence can change depending on linguistic framework. The proposition itself - the representation that is the meaning of the sentence - doesn’t change. And thus the truth doesn’t change. So this is another bad argument.

The third argument for subjective truth comes by generalization from paradigm cases. There are some paradigm cases of subjective truths; perhaps we can make a case for subjectivism by identifying what is going on in such cases, and generalizing it. All truths are subjective in the same way that paradigm cases of subjective truth are subjective. So, for instance, matters of personal taste are supposed to be subjective. Perhaps there’s some reason to think that all truths are similarly subjective, just as matters of personal taste are subjective.

But we must be careful here. Let’s allow that there is something about personal taste, like taste in food, that is subjective. My wife loves seafood and can’t drink coffee without a ton of milk and sugar in it, and I’m the exact opposite. I love strong black coffee and can’t stand the taste of fish. And these are both perfectly legitimate matters of taste. But where can we find “subjective truth” here? Consider the claim that I hate the taste of fish (whereas my wife loves it). This is objectively true. “Matt doesn’t like fish” is a proposition that represents me as disliking fish, and it is true because I do have that property. Scientists could study this and figure out that it’s true by putting fish in front of me and watching me pull a face. My wife and I disagree about what to have for dinner, but we don’t disagree about whether Matt likes fish. Ask her whether or not I like fish, and she’ll quickly answer “haha, no.” It is an objective fact that I don’t like fish, and she knows this. She teases me about it all the time! (Just like I tease her about the fact that she gets a milkshake every time she goes to Starbucks even though she calls it coffee.)

This shows that, if we want to put our finger on what a subjective truth consists in, we need to be a bit more careful than just gesturing in the direction of subjective experience. So here’s the sensible thing to say: Consider the claim “Fish is disgusting.” This is something that I might say. My wife, on the other hand, might say “Fish is not disgusting. Fish is delicious.” What should we make of this dispute? A certain kind of hard-headed philosopher might say that the word ‘disgusting’ refers to the property of disgustingness, and fish either has that property or it does not. If it does, Matt is right; if it doesn’t, my wife is right. I’m pretty sympathetic to this hard-headed view. But perhaps it’s fine to say that “Fish is disgusting” is true for Matt, while “Fish is not disgusting” is true for his wife. Neither one needs to be wrong about how appealing fish is; de gustibus non disputandum est. We are both simply stating their own subjective truths. And sure, that seems fine, let’s go with that.

But if we have an instance of a subjective truth here, it’s a very thin kind of subjective truth. It is subjectively true for me that fish is disgusting. But in virtue of what is that so? What makes it the case that fish is disgusting (for me)? Just this: that I don’t like fish. But the fact that I don’t like fish is an objective truth! Scratch the surface of this subjective truth, and we find an objective truth hidden directly underneath. This is not an objection to the idea that “Fish is disgusting” is a subjective truth. We’re allowing, at least for the sake of argument, that it is a subjective truth. The point is just that this is a thin subjective truth, because it is fully grounded in an objective truth.

Other subjective truths are thin in this way. Consider facts about currency. A $20 bill is currency in the US, but not elsewhere in the world. Give a $20 to a merchant in China, and they’ll give it back to you. “That’s not money.” And, to be charitable, they’d be right to say that. The $20 isn’t money in China, but it is money in the US. For a US denizen, it’s money; for a Chinese denizen, it’s not. Whether the $20 is money is a subjective truth. But again, it’s thin. The $20 is money in the US entirely in virtue of the laws and customs of the US, and it’s not money in China entirely in virtue of the laws and customs of China. But those truths I just stated about laws and customs are objective truths. So, again, we’ve found some thin subjective truths, subjective truths which obtains entirely in virtue of objective truths.

At this point, you might expect me to help illustrate the point by giving an example of what a non-thin subjective truth might look like. The problem is, I can’t do that, because I can’t think of a single plausible example. All subjective truth is thin subjective truth.

Let’s make this point a bit sharper. Subjective or relative truths are truths that obtain in relation to some framework or another. That a piece of paper is currency depends on a certain monetary or legal framework. That fish is disgusting depends on my own personal tastes in food. This is why these facts are relative or subjective rather than being objective. What makes them true is not (just) something about the way the world is, it’s also a relationship to a truth-making framework of some kind. But the facts about the truth-making framework are objective facts. Thus, the problem for subjective theories of truth arises when we take a step back and ask about the framework and the fact’s relationship to it. “‘The fish is disgusting’ is true for Matt in virtue of the fact that Matt doesn’t like fish.” Is that sentence, the sentence in quotes, a subjective truth or an objective truth?

In asking that question, we’re beginning to tumble down the rabbit hole into some abstract and highly controversial theoretical concerns. Before we do, let’s pause for a moment to take stock of where we’re at. We’ve seen that there are three possibilities: that all truths are objective, that all are subjective, and that some are objective and some are subjective. A mixed theory, on which some truths are objective and some are subjective, remains a live option (although in discussing the way in which certain paradigmatic subjective truths are thinly subjective, I’ve perhaps given an indication of why I’m not sympathetic to that mixed theory.) But our target is the theory that all truth is subjective. We’ve also said that truths are subjective if they consist in a relationship not just between a proposition and the world, but in a three-place relationship between a proposition, some framework, and the world. And the crucial question we’re asking now is: for any proposition describing such a three-place relationship, is that relationship an objective truth or a subjective truth? If any of that is unclear, perhaps re-read the above section until it’s clear, because the discussion below assumes that you’ve fully understood the above points.

Good?

Ok.

The line of argument I’m pressing here is related to “reflexivity,” which is the name that has been given to the problem of applying a subjective theory of truth to itself. Is subjectivism subjectively true or objectively true? The usual way of pressing this dilemma goes back to Plato’s dialogue, Theatetus. If subjectivism is objectively true, then there is at least one objective truth, and so not all truths are subjective. But if subjectivism is subjectively true, then subjectivism isn’t really true; it’s just, like, your opinion, man. And who cares about your opinion? Checkmate.

But this isn’t checkmate at all. Any subjectivist will say that subjectivism is subjectively true. All truths are subjectively true, including subjectivism itself. That’s just what the view says. And why is that a problem for subjectivism? It’s not self-contradictory. You might say that you don’t care about subjectivism if it’s just the subjectivists’ opinion. But (1) talk of ‘opinion’ is a red herring; subjectivism is a theory of the nature of the truth relation that is at most tangentially connected to issues of opinion, and (2) even if this is just the subjectivist’s opinion, it is their opinion, which they hold consistently. So where’s the problem for the subjectivist? They’re not doing anything wrong (by their own lights.)

In fact, among subjectivists, embracing reflexivity is often seen as a paradigmatic kind of epistemic virtue. This is because, for the typical postmodernism-influenced subjectivist, embracing subjectivism isn’t really about endorsing a particular kind of formal model of truth-functional semantics. They might be committed to the formal bit, but that’s not what they care about. Rather, embracing subjectivism is often seen as a kind of epistemic humility. It’s an acknowledgment that other people have other views, and they can make arguments in favor of their views that they themselves find plausible. And even if we don’t find their arguments plausible, they don’t find our views or our arguments for them plausible. So who’s to say who’s right? It’s all subjective. So “reflexivity,” applying subjectivism to itself, doesn’t raise any problems. Indeed, it’s just an opportunity to be maximally humble. I’m humble about the very theoretical basis of my humility. That makes me the most humble. (What’s so good about being the most humble? Well, that’s good from my perspective; but maybe you think different, and who’s to say? I’m humble about my humility and also about the value of being humble about my humility. This is intellectual sophistication.) The person who would try to stop this endless cycle of humble-bragging by trying to find an objective groundwork to stand on is typically viewed with a combination of deep suspicion and withering contempt.

But there is a problem with applying a theory of subjective truth to itself, and it’s not the simple gotcha of “Is that an objective truth?” The problem is illustrated by Paul Boghossian, he of the report, in his book Fear of Knowledge. The master argument of that book is that bottomless subjectivism is really incoherent. Here’s why.

The objectivist theory of truth works because if a proposition is, by itself, sufficient for representing the world as being a certain way, then it’s straightforward that the world either is that way or it’s not. So if we need a framework in order to assign a truth value to a proposition, then the proposition by itself is not sufficient to represent the world as being a certain way. The proposition requires supplementation by a framework in order to be genuinely representational. That is, if a proposition P is true only relative to a framework F, then it’s not just the case that we need F to tell us whether or not P is true, but F is needed to help us understand what P is even saying in the first place. Consider again the sentence “Fish is disgusting.” If that’s true for me but not true for my wife, then there’s got to be something about the meaning of “fish is disgusting” that maps that sentence onto my own tastes when I say it, and maps that sentence onto my wife’s tastes when she says it. There has to be something like that, because that’s what makes the sentence subjectively true or false depending on who says it. If there’s nothing like that, then “fish is disgusting” isn’t subjectively true or false, it’s just meaningless arglebargle.

The cost, then, to being a subjectivist about frameworks is that this amount to putting off intelligibility. Frameworks supply meanings. A proposition P is only true relative to some framework. But what about that framework? If a framework, framework-A, is true only relative to some other framework, framework-B, then framework-A is meaningless without framework-B, and so without a framework-B, framework-A cannot make P meaningful. Framework-A cannot do its meaning-making work without framework-B. But framework-B is itself meaningless without the assistance of another framework, framework-C. And so on… A computer programming metaphor might help here. A common operation in computer programs is to call or lookup a value from some library or table. But that library or table might itself contain another call function which directs the code to another library or table… and this process can be repeated indefinitely, with the logic of the code chasing call function after call function. And this is all fine, so long as the chain ends with an actual value that it can propagate back through that series of call functions. If there’s an endless chain of lookups, the program will hang and your computer will crash. The same goes if the lookups loop. Similarly, then, if your semantics doesn’t bottom out somewhere in some facts about a framework that are true without reference to some other framework, then your semantics will either loop or continue on an endless regress. But either way, they’ll crash, and you’ll never end up saying anything meaningful. Of course, we do end up saying meaningful things, at least sometimes. Therefore there are objective truths, if only objective truths about frameworks.

Here’s another way to put the basic point. We’ve said that we’re looking at the relationship between (potentially) three different things: a proposition, the world, and a framework. Propositions are supposed to be representational; they represent the world as being a certain way. So for any given claim, is the proposition by itself sufficient to represent the world as being a certain way, or is it not? If it is, then we have all we need for objective truth, the world either is the way the proposition represents it as being, in which case the proposition is true strictly in virtue of that relationship between the world and that proposition, or it isn’t, in which case the proposition is false strictly in virtue of that relationship between the world and the proposition. The framework falls out of the picture; this is objective truth. So a subjectivist would have to say that the proposition doesn’t represent the world by itself. But if the framework plays a role in determining the way that the proposition represents the world as being, then we need the framework to say how the proposition represents the world as being in the first place. Without the proposition to help, we don’t yet have a complete representational meaning. This is what gives rise to Boghossian’s problem, where we either need to say that there’s an objective fact about how a framework can give a proposition meaning, or else we end up an endless meaningless arglebargle of frameworks-about-frameworks. (Sorry to repeat myself a bit for these last three paragraphs, but the point is worth restating a couple different ways because it’s both rather abstract and highly important.)

This argument assumes that the role frameworks play is some kind of meaning-completion, without which a proposition is meaningless and not successfully representational. We’ve just seen that, if frameworks play this kind of role, subjectivism ends up being incoherent. But a subjectivist might try to cash their view out in some other way. What other way might that be?

Perhaps the view is that the meanings of sentences aren’t really representational propositions at all. They’re merely expressive, or serve some other sort of coordinative purpose. That’s doubtlessly true of some sentences. But it’s absurd to think that that’s the function of all sentences. “Paris is the capital of France” represents Paris as being the capital of France.

Perhaps the view is that subjective propositions are representational, but they just represent elements of our frameworks. This is the sense in which a sentence like “Fish is disgusting” is subjectively true: it simply represents an element of my personal framework of taste in food. But this isn’t a recipe for a subjectivity about all truth. That’s because my personal framework of taste in food is part of the world, and thus in representing my framework of personal taste in food, the subjective truth that “fish is disgusting” is representing an aspect of the world. That’s why I called this kind of thing “thin subjective truth;” it looks a whole lot like objective truth if you think about what’s going on here for a second. I won’t say you can’t call this a “subjective truth” if you want to, but this isn’t the kind of account of subjective truth that the subjectivist needs.

Perhaps the best hope for the relativist is to say that truth isn’t a matter of accurate representation at all. Propositions might represent the world as being some way, but whether or not the proposition is true is not settled simply by the question of whether the world is the way it’s represented as being. Truth is not a matter of accurate representation of the world, but rather a matter of representation-plus-something-else-having-to-do-with-frameworks. This is the kind of view that some of the most sophisticated subjectivists tend to pursue, since there’s no obvious contradiction here. But there’s still a huge problem with this view, which is that no one has any idea what it means. It’s not enough to have a formal system where truth is an output of a three-argument function rather than a two-argument function. We need an interpretation of that formal system to make sense of that formalism, and there’s nothing plausible in the offing. There are two main problems here. First, it is plausibly true by definition that truth is entirely a matter of accurate representation. If you say that there are propositions that represent the world as being a certain way, and the world is that way, but we don’t yet know whether or not that proposition is true, then you just don’t understand what truth is. Second, the whole project is undermotivated. We don’t have any workable examples of deep subjective truths that we can use as models to build a larger theory around. The only reason to think that there is any such thing as deeply subjective truth is that certain relativist theories imply that there is. But the standard arguments in favor of those relativist theories are, as we saw above, deeply confused. And the theory itself doesn’t make any sense. So what are we even doing here?

One place where someone like Anderson might object to the argument of the last paragraph is that I said that a certain idea about truth is true by definition. But that’s a bit of a weasel. What are these “definitions” that make words true? There is nothing necessary about them. We can use words however we like. Perhaps everything I’ve argued above is what happens if you set out to use the word ‘truth’ in a certain way, i.e. as describing a two-place relationship between a proposition and the world. But if you use the word ‘truth’ in another way, i.e. to describe a three-place relationship between a proposition, the world, and a framework, then none of my arguments work.

This is actually a good point. The subjectivists might very well be using their words in a different way, and are pursuing a fundamentally different program in epistemology, and that’s, in it’s own way, fine. Trying to argue against this will commit you to a kind of Platonism that is indefensible. We’re not discerning the form of Truth via a priori reflection.

But this point is limited. Because while language is malleable, there are things that we mean by our words. So I put it to you: does what I’ve argued seem straightforward from the perspective of how you use the word ‘true?’ If so, then we’re all on the same page, except for the subjectivists who are off on their own playing language games.

What we care about in academic inquiry isn’t whatever-anyone-might-happen-to-mean-when-they-use-the-word-‘true.’ What we care about is truth - which is to say, truth as we understand it. And if you want to use the word ‘truth’ to talk about something else, then you may do so, but don’t expect us to care about your subject of interest. (By silly analogy: suppose some group decided to use the word ‘truth’ to mean ‘a giant bowl of lime jello.’ They’d be well within their rights to do so. But if they then declared themselves to be passionate about the pursuit of truth, we should understand their interests as being primarily concerned with dessert, and would be well within our rights to say that such a pursuit is not consonant with the university’s search for truth. We wish them the best in their quest for jello, but there’s no reason to allot tenure lines towards it just because they use the word ‘truth’ to describe their quest.)

Whew! That was a lot, so let’s sum up the main points. There are two potential accounts of truth, one that understands truth as a two-place relation between a proposition and the world, and one that understands truth as being a relation between a proposition and some evaluative framework (and perhaps the world as well). While the former conception of the truth is straightforward in many ways, the latter conception of the truth proved to be both undermotivated and underdeveloped. And while I don’t want to go so far as to say that any subjective theory of truth is incoherent or self-contradictory, the difficulties with trying to explain how such a thing might work all point in the direction of objective truth being our common-sense concept of truth, with subjective truth being something much more strange and elusive. So when we academics say that they are pursuing the truth - the sign over the gate says “Veritas” - we should be understood as saying that we care about objective truth. Subjective “truth” is a sideshow at best.

So what are the upshots? First, I’ll begin with a concession. I have just engaged in a bit of what philosophers call “pounding the table,” insisting on my way of viewing things rather than some other way of viewing things. Anderson might well respond that it’s totally consistent and coherent to care about subjective truth and pursue subjective truth rather than objective truth, and to have methodologies oriented around doing so. I haven’t shown any essential defect in that project, although I hope to have given reasons to doubt it. But by the same token, it is equally consistent and coherent to care about lime jello and pursue lime jello rather than objective truth, and to have methodologies oriented around doing so. There’s no essential defect in this project either; it sounds delicious! The question is whether the pursuit of either of these things - lime jello or whatever it is that subjectivists about truth are on about - are part of the core mission of the university. And again, to concede to Anderson’s argument: we can’t deduce the core mission of the university a priori. There’s just what we contingently happen to care about. But I stand for a university that cares about objective truth - or, as it is commonly known, truth.

Second, I’m not arguing that research into subjective truth should be cut off. Perish the thought! If relativists can come up with better ways of explaining and defending their view, that would be awesome. There are plenty of philosophers at work on that task, and I wish them the best. But unless and until there are better explications and defenses of relativism in the offing, any methodology that is premised on the acceptance of relativism is ipso facto suspect; and any discipline that routinely uses such a methodology is presumptively hogwash.

Third, this is not to say that academic work in the humanities or the various “interpretive disciplines” is all presumptively hogwash. There is plenty of good critical humanistic work to be done. Work in the humanities needn’t consist in investigation of objective truths. It can, for instance, be done by applying different interpretive lenses to texts, to social practices, and to much else besides. But the kinds of subjective truths that come out of this kind of interpretive exercise should be understood as thin subjective truths, consisting entirely of objective facts about the relationship between a critical lens and its object of study. Applying those critical lenses can be a worthy and edifying pursuit, but let’s not let the interpretive nature of this work make us lose sight of the fundamental anchoring of the discipline in truth.

Fourth, this still leaves a lot of room for epistemic pluralism. Epistemology is theory-laden, and there’s lots of coherent and defensible theories out there that we might build our conceptions of rationality on. But these different epistemologies should be understood as offering up different ideas about how we might get to the ultimate goal of truth. Subjectivist theories that deny the existence of such a goal are playing a different and less defensible game.

The Boghossian report argues that subjectivism about truth is the cancer eating the humanities from within. If scholars don’t see their aim as putting forward propositions that accurately represent the world but only as putting forward claims that are valid from the point of view of some evaluative framework, their work loses its essential grounding in reality. And it’s hardly any surprise if the preferred evaluative frameworks of the humanities follow the politics of the humanists to the left. But if we reject subjectivism about truth as confused, we re-establish a correspondence with reality as the overriding goal of all academic work. Politics may still play a role in the academy. But if the goal of academic work is truth, the corrupting influence of politics may be greatly mitigated.

Those who reject this position put themselves at odds with the mission of the university, and thereby sacrifice the protections of academic freedom. That is the conclusion of the Boghossian report. I agree with it.