Lecture 2 — The Standard Objections

John MacFarlane (jgm@berkeley.edu)

Context and Content Lectures, Paris, 2010

Relativism - a first stab

Eduardo: Dulce de leche is tasty.
John: No, it isn’t! It’s just too overwhelmingly sweet.

The received wisdom

That (total) relativism is inconsistent is a truism among philosophers. After all, is it not obviously contradictory to hold a point of view while at the same time holding that no point of view is more justified or right than any other?
—Hilary Putnam

…the contemporary consensus among analytic philosophers is that relativism is not just wrong, but too confused a position to be worth taking seriously.
—Thomas Bennigson

…the label ‘relativistic’ is widely regarded as pejorative, and few philosophers have been willing to mount an explicit defense of relativism.
—Chris Swoyer

Relativism is even sillier than it at first appears. Indeed, if relativism were not so popular, it wouldn’t be worth discussing at all. And even given its popularity, it isn’t worth discussing for long.
—Jamie Whyte

…of all the conceptual options that have ever crossed the mind of the philosophical tribe, none has attracted quite the scorn and ridicule of the relativist.
—Joseph Margolis

Even Rorty repudiates it!

Truth is, to be sure, an absolute notion, in the following sense: ‘true for me but not for you’ and ‘true in my culture but not in yours’ are weird, pointless locutions. So is ‘true then, but not now.’ Whereas we often say ‘good for this purpose, but not for that’ and ‘right in this situation, but not in that,’ it seems pointlessly paradoxical to relativize truth to purposes or situations. —Richard Rorty

A summary of the arguments

  1. Truth relativism is incoherent once you get clear about the truth-bearers.
  2. Truth relativism is self-refuting.
  3. Truth relativism is inconsistent with the Equivalence Schema, the most central principle governing truth.
  4. Nobody has said clearly what truth relativism comes to.

W. Newton-Smith on relativism and truth bearers

  1. The thesis that a sentence could be true in one social group/theory and false in another is trivial, since the sentence might have different meanings in each.

  2. So, an interesting relativism about truth must “focus not on sentences but on what is expressed by a sentence,” a proposition.

  3. But, the thesis that a proposition can be true in one social group/theory and false in another is incoherent.

(Similar arguments can be found in Husserl and some other contemporary writers.)

Newton-Smith’s argument

Let p be the proposition expressed by sentence ‘S1’ in Ψ and by sentence ‘S2’ in Θ . Could it be the case that p is true in Ψ and false in Θ ? No, for it is a necessary condition for the sentence ‘S1’ to express the same proposition as the sentence ‘S2’ that the sentences have the same truth-conditions. To specify the truth-conditions of a sentence is to specify what would make it true and to specify what would make it false. If in fact ‘S1’ and ‘S2’ differ in truth-value, their truth-conditions must be different. If their truth-conditions differ they say different things—they say that different conditions obtain—and hence they do not express the same proposition. Thus if we focus on propositions we cannot find a proposition expressed by a sentence ‘S1’ in Θ  and by a sentence ‘S2’ in Ψ which is true in the one case and false in the other.

Newton-Smith’s argument articulated

  1. Suppose S1 is true in Ψ and S2 is false in Θ . And suppose that S1 and S2 express the same proposition, p.

  2. S1 and S2 differ in truth value (by 1).

  3. If two sentences differ in truth value, they differ in truth conditions.

  4. If two sentences express the same proposition, then they have the same truth conditions.

  5. So if two sentences differ in truth value, they do not express the same proposition (by 3, 4).

  6. So S1 and S2 do not express the same proposition, contrary to our supposition (by 2, 5).

A misguided response

Accept Newton-Smith’s argument and try to spin relativism as a view about the truth of something other than propositions and sentences—for example, utterances or assertions.

The self-refutation argument

Strong global relativism

Everything is true merely relatively, and that nothing is true absolutely — that is, nothing is true for everybody, or relative to every perspective.

The dilemma
  • If the global relativist says that Strong global relativism is true for everyone, then she is acknowledging that there is at least one non-relative truth.
  • If she concedes that Strong global relativism is false for someone, then… (?)
    • her position is contradictory
    • her position is pragmatically undermined
    • her position can’t be coherently formulated

Plato’s Theaetetus 40a-b

SOC: Secondly, it has this most exquisite feature: Protagoras admits, I presume, that the contrary opinion about his own opinion (namely, that it is false) must be true, seeing he agrees that all men judge what is.

THEOD: Undoubtedly.

SOC: And in conceding the truth of the opinion of those who think him wrong, he is really admitting the falsity of his own opinion?

THEOD: Yes, inevitably.

Hales’ formal reconstruction

Steven Hales shows that global relativism is self-refuting if we assume a principle analogous to S5.

◇ p
it is relatively true (true for someone) that p
□ p
it is absolutely true (true for everyone) that p
◇ □ p ⊃ □ p

But why should a relativist accept S5 for these operators?

Pragmatic incoherence

The content of the relativist’s claim is inconsistent with a true description of what she would be doing in asserting it.

Compare: “I am not asserting anything.”

[E]ven if we can make some sense of the description of p as ‘being true for x’…Protagoras is still asserting that ‘p is true for x’ and ‘p is not true for y’; these propositions he is taking to be true. It has to be true not only for x but for everybody that ‘p is true for x’ since this exactly what is involved in asserting that ‘man is the measure of all things.’ The fundamental criticism of Protagoras can now be put thus: to engage in discourse at all he has to assert that something is the case.
—John Passmore

Assertion and absolute truth

Key idea: to assert that p is to put p forward as true absolutely (true for everyone).

No amount of maneuvering with his relativizing qualifiers will extricate Protagoras from the commitment to truth absolute which is bound up with the very act of assertion.
—Myles Burnyeat

Assertion and relative truth?

Why shouldn’t the relativist say that in asserting p, one is putting p forward as relatively true—perhaps, as true relative to oneself?

Not clear what this means, but also not clear it’s an untenable position. (We will try to get clearer about this in Lecture 5.)

Regress of formulation?

The relativist faces a regress in formulating her own position.

  1. I’m only putting my thesis forward as true for me.

Is (1) being put forward as true absolutely?

then the relativist has conceded that there is at least one relative truth, and is refuted.

then the relativist faces a regress:

  1. I’m only putting (1) forward as true for me.
  2. I’m only putting (2) forward as true for me.

But why is this bad?

Burnyeat’s diagnosis

The problem lies in the complexity of the propositions to which we will be led in the iteration from (1) to (2) to (3)…


Protagoras, as Socrates keeps saying, is a clever fellow, but he is not so clever that there is no limit to the complexity of the propositions he can understand and so judge to be true. Therefore, the relativist prefix ‘It is true for Protagoras that…,’ unlike the absolute prefix, admits of only limited reiteration.

Boghossian’s diagnosis


The upshot is that the fact-relativist is committed to the view that the only facts there are, are infinitary facts of the form:

According to a theory that we accept, there is a theory that we accept and according to this latter theory, there is a theory we accept and … there have been dinosaurs.

But it is absurd to propose that, in order for our utterances to have any prospect of being true, what we must mean by them are infinitary propositions that we could neither express nor understand.

Putnam’s diagnosis


The argument is that the relativist cannot, in the end, make any sense of the distinction between being right and thinking he is right; and that means that there is, in the end, no difference between asserting or thinking, on the one hand, and making noises (or producing mental images) on the other. But this means that (on this conception) I am not a thinker at all but a mere animal. To hold such a view is to commit a sort of mental suicide.

What is left of the self-refutation argument?

Some questions for the relativist to answer:

Answers in the following lectures.

Is local relativism immune?

Local relativism about truth
Some propositions (in certain domains) are true only relatively.

The local relativist can grasp the horn of the dilemma that was not available to the global relativist:

It is absolutely true that claims of taste are true only relatively.

Not so fast…

The real problem the self-refutation argument raises for the global relativist—explaining how putting something forward as true not absolutely, but only relatively, could count as an assertion—is equally pressing for the local relativist.

The equivalence schema

The objection is that relativism about truth is incompatible with the

Equivalence schema
The proposition that Φ  is true iff Φ .

This (or a suitably qualified version) seems fundamental to our use of the truth predicate.

  1. For all P, if Billy asserted that P, P.
  2. Every proposition Billy asserted is true.

We need the Equivalence schema to get the inferential power of (a) from (b).

Ramsey vs. James

In “The Nature of Truth,” Ramsey criticizes philosophers who “produce definitions of truth according to which the earth can be round without its being true that it is round”:

…according to William James a pragmatist could think both that Shakespeare’s plays were written by Bacon and that someone else’s opinion that Shakespeare wrote them might be perfectly true ‘for him.’

A compatibilist approach

The relativist ought to hold that if

Dulce de leche is tasty

is true relative to some perspectives and false relative to others, so is

It is true that dulce de leche is tasty.

Work for the relativist

  1. Explain how the monadic truth predicate and the relativized one are related.
  2. Explain why the relativized one should be called a truth predicate at all.

We’ll return to this in Lecture 4.

What does it mean?

What does it mean to say that a proposition is “true for Eduardo” or “true relative to Eduardo’s tastes”?

Uses of “true for” in ordinary language

“True for” does have an ordinary use, but it’s not the one we’re after. One use is to specify the domain for which a generalization holds.

For example:

…while the doors to high civil, military and academic office have been opened to merit for members of other communities, this has not been true for Muslims.


the doors to high civil, military and academic office have been opened to merit for members of their communities

Examples of the generalizing use

Some of those religious leaders commute between homelands and host countries, not only providing religious continuity but also maintaining political contacts. That has been particularly true for Turks, Pakistanis, Iranians, Sikhs, Jews, and Palestinians.

About half of all fatal heart attacks occur in women, but a woman who has a heart attack has a twofold risk of dying within the first two weeks, compared to her male counterpart. This is especially true for women under the age of forty-five.

It is well known that certain human subjects are especially resistant to the gas and I have frequently found this to be true for dogs.

“The pianos used by Mozart, Beethoven or Chopin were radically different from the large, loud black instruments found in all modern concert halls,” Moroney said. “The same is true for violins.”

Not only was it useless to try to derive the satisfaction she needed from another individual through subordinating her needs to that person’s, but even attempting to do so was destroying her. Generalizing her insights, she concluded that what was true for her was true for all people.

The intentional use

p is true for x” = “x takes p to be true”

Here the work is done by “for,” since we don’t really need the “true”:

For Sarah, that fossil is less than 5000 years old.

For John, those socks are the same color.

For Elroy, ant mounds are space stations.

The upshot

I think this is the principal challenge for truth relativism, and the one the existing literature has made least progress in answering.

Pragmatist theories

One strategy is to start with an existing theory of the nature of truth and show that, when followed through, it implies that truth is relative.

William James:

  1. Truth is “only the expedient in our way of thinking.”
  2. What is expedient for one person need not be expedient for another.
  3. What is true for one person need not be true for another.

…in any concrete account of what is denoted by ‘truth’ in human life, the word can only be used relatively to some particular trower.

Epistemic theories

Begin with an epistemic account of truth as some sort of idealization of rational acceptability: true sentences are those which disinterested inquirers would assent to under ideal conditions, or at some idealized ‘limit of inquiry.’ The relativist simply adds that different communities of inquirers, starting from different sets of assumptions about what is plausible, noteworthy, explanatory, etc., might approach different limits. Thus, on the appropriate epistemic conception of truth, conflicting conclusions could be true for different communities.
—Thomas Bennigson

This might work even if we assume that any two communities of idealized enquirers would have access to the same observations and experimental results. For what is confirmed by these observations depends on background “prior probabilities.”

These routes aren’t promising

Correspondence theories

Meiland’s proposal: “φ is true for Jones” means “φ corresponds to reality for Jones.”

But what does “corresponds to reality for Jones” mean?

… although this question is embarrassing in the sense that it is difficult for the relativist to give any useful answer to it, nevertheless the relativist is in no worse a position than the absolutist at this point. … relativism is not to be faulted for being unable to give an account of that which the absolutist cannot give an account of in his own position either.
—Jack Meiland

An insight and a strategy

Meiland’s insight

The relativist should not be held to a higher standard in explicating truth than the absolutist.

A strategy

Look at the best non-relativist explication of truth, and explicate relative truth in a similar way, using similar materials.

We’ll return to this project in Lecture 5.

Summing up

None of these objections are knock-down arguments, but they do point to further work the relativist needs to do.

In the remaining lectures, we will develop a relativist position that discharges these tasks.