Value Claims and Preferences — The Third Distinction Developed v1.0
Value Claims and Preferences — The Third Distinction Developed v1.0
Theoretical foundations: Grant C. Sterling (Eastern Illinois University). Analysis and synthesis: Dave Kelly. Prose rendering: Claude (Anthropic). 2026.
I. The Beginning
The premise is Th10, and the word in it that carries the distinction is actually:
Th 10) The only thing actually good is virtue, the only thing actually evil is vice.
Not what the Stoic counts as good. Not what it is useful to treat as good. What is actually the case. Sterling states the status of the claim explicitly in the sixth excerpt, where he sets out System S and interrupts his own numbered list to mark it:
3) This belief is factually false. (Note that this is not a psychological claim--it is a value claim.)
The belief in question is that externals have value. The parenthesis is doing exactly the work this essay develops. He is anticipating the reader who hears false and translates it into something about psychology — a belief that produces bad outcomes, a maladaptive belief, a belief that does not serve. He blocks the translation before it happens. The belief is false in the way beliefs are false.
Everything below is derived from that. A value judgment is a proposition. It is true or false. Its truth is not a fact about the person holding it.
II. The Impression Arrives as a Claim
Before the distinction can be applied it has to be located, and its location is the impression itself. Sterling’s account, in the seventh excerpt, is that impressions are not raw material awaiting interpretation:
“Those impressions are cognitive, propositional--they are not uninterpreted raw data, but rather ideas that claim that the world is a certain way. I do not see a collection of colors and patterns--I ‘see’ my backpack sitting on the chair in front of my desk.”
He then separates the value-neutral case from the other. His backpack being on the chair is nothing good or bad. But suppose he remembers leaving it on the floor: the impression that now arrives is that his backpack is on his chair, which means someone has intruded on his office, which is a very bad thing.
Read that impression as a sentence and its structure is visible. It contains a report of fact and a verdict, and the verdict is stated in the same voice as the report. Which is a very bad thing does not announce itself as an evaluation. It arrives with the same assertoric force as my backpack is on my chair — which is precisely why it is assented to as easily.
This is what makes assent capable of being correct or incorrect. One cannot assent to a mood or a disposition; there is nothing there to accept. One assents to a claim. If value impressions were reports of how the agent is disposed, assent would have no work to do, since the disposition would already be whatever it is and could not be got wrong.
III. What Correspondence Supplies
Sterling’s statement of the criterion, to the Stoics group on 20 August 2015, is unhedged:
“The only ‘criterion of truth’ that I recognize is correspondence with the facts — correspondence with reality. I reject utterly any notion of ‘truth’ wherein something can be ‘true’ and yet not match reality. And I am an authentic Stoic in this regard — the Stoics were pure realists in this regard.”
He then draws the consequence for the framework, and it is the sharpest statement in the record of what the third distinction protects:
“Pain and death and defeat and unemployment and rejection (etc.) all seem to be bad things. ‘Common sense’ says they are bad things. Our pre-existing notions say that they are bad things. Stoicism says that nevertheless they are not — all these impressions are false, and we must radically revise the way we see the world to embrace the truth. If we undermine the claim that there are objective facts, it is hard to see what justifies us in radically revising our beliefs.”
The argument turns on the word revise. Stoicism does not present itself as an alternative outlook available to those who find it congenial. It presents itself as a correction — it says that what nearly everyone believes about value is wrong and must be changed. A correction requires an error, and an error requires a fact to have been missed. Remove the fact and the revision has nothing to revise toward; what remains is one way of seeing beside another, and the demand to change becomes a recommendation to try something.
One refinement in the same message must travel with the doctrine, because without it the position looks like a claim to certainty it does not make. Sterling separates the criterion of truth from the criterion of knowledge. His example: either the number of oxygen molecules in a Pepsi can is even or it is odd, one of these is true, and no human being can know which. A kataleptic impression guarantees truth because by definition it corresponds to the facts — but billions of sentences are true for which nobody has one. He adds that he believes Barack Obama is President without having a kataleptic impression of it, and so does not know it, while holding it objectively true.
The relevance is direct. To hold that value claims are truth-apt is to make a claim about the subject matter, not about one’s own reliability. It commits the practitioner to there being a fact of the matter. It does not commit him to being right.
IV. What Moral Realism Supplies
Correspondence tells you what would make a value claim true. It does not by itself supply anything for value claims to correspond to. That is moral realism’s contribution, and Sterling’s case for it, on 13 January 2015, is built from the failure of the alternative.
His example is a person he does not care about:
“I care about my daughter. So if some action would benefit me in some way but hurt her, I can see a reason not to do it. But let us say I do not care about what happens to Wanda. So the fact that some action would hurt her does not give me any desire-based reason not to do it. If I can satisfy my desires by exploiting her, why should I not?”
And the conclusion:
“There must be some kind of reasons that are utterly independent of my contingent desires. The Stoics think that the fact that I would be harming my parents is a reason not to do something whether I care about them or not. ... If there are no objective moral facts of this sort, then the Stoic project fails utterly.”
Two further elements of that message bear on the present distinction.
The first is his ordering of moral rules. Rules that describe inherent moral considerations — Ross’s prima facie duties, weights that go into the pan — are Type A. Rules of thumb, which record what has usually turned out right, are Type C. And Type C presupposes Type A: one cannot learn from experience which way past weighings came out without already knowing what counts as a weight. Empirical practical wisdom cannot bootstrap itself. Whatever a practitioner has learned about how to live presupposes value facts he did not learn.
The second is the modal claim, from 26 May 2021. Asked what the source of moral truth is, Sterling answers that it has none:
“This has no ‘source’, just as ‘2+2=4’ has no source. It is a fundamental, necessary, and unalterable fact about the universe that, all other things being equal, one should maximize preferred indifferents.”
And on how we know it: by Reason, in the way we know that 2+2=4 and that q follows from if p then q together with p. He adds the complication immediately, and it is not a small one — that we have developed bad habits since childhood of believing that what seems to benefit us is good, so we tend to deny obvious moral truths when they are inconvenient. His illustration is the man who does not repay a debt and pretends he has no obligation to, and is furious when someone fails to repay one to him.
V. The Preferred Is Not a Preference
This is the point at which the vocabulary of the system works against the distinction it needs, and it deserves its own section.
Preferred indifferent sounds like a preference. It is not. The 2021 message just quoted makes the status unmistakable: that one should, all other things being equal, maximize preferred indifferents is stated as a fundamental, necessary, unalterable fact about the universe, in the same breath as 2+2=4. Whether health is a preferred indifferent is not settled by whether the agent prefers health. Sterling’s gloss is that someone who doubts it does not really understand what preferred indifferents are — which is the form a claim takes when its denial is a failure of comprehension rather than a difference in taste.
So the second axis of the fifth distinction — preferred and dispreferred — is itself objective. Both axes are. What distinguishes them is not that one is factual and the other personal; it is that one concerns genuine good and evil and the other concerns appropriate objects of aim. A person may of course also have preferences in the ordinary sense, and those are facts about him. They are simply not what the technical term names.
This matters because the commonest way of holding the system incorrectly is to accept Th10 sincerely while treating the preferred as the zone where one’s own tastes have been given official standing — where I like this has been licensed by the framework. It has not been. The preferred is a further set of claims about the world, and the agent can be wrong about it in the same way he can be wrong about anything.
VI. The Grammar That Hides the Difference
This is bad and I can’t stand this have the same surface urgency and different logical forms, and English lets the second wear the clothing of the first. The result is that a report about oneself gets uttered, heard, and assented to as a claim about the world.
The serviceable test is a question about falsification. Take the sentence and ask what would have to be the case for it to be false. If the answer is a fact about the object — about what health is, about what a promise is, about what harms a rational faculty — the sentence was a value claim. If the only thing that could falsify it is a report about the speaker’s own state, it was a preference, and it was never a candidate for correction in the first place.
Applied honestly this is uncomfortable, because a good deal of what a practitioner takes himself to be asserting turns out on inspection to be the second kind. That is not a defect in the test. It is what the test is for. A preference misidentified as a value claim cannot be corrected, since there is nothing wrong with it; and a value claim misidentified as a preference will not be corrected, since nothing seems to be at stake. Both errors leave the judgment where it is.
VII. The Collapse
The Edelstein record is the worked case, and its interest is that the collapse there is not careless. It is stated as a considered position and defended.
His P1, from 16 September 2019 and recorded in the CPA run as the difference he himself nominates as essential between REBT and Stoicism: there is good and bad in the REBT view, but subjectively rather than objectively — being hit by a truck is bad in terms of his larger goal of avoiding pain. The finding entered against this is Contrary on ethical intuitionism, on the ground that nowhere in the retrieved record is there an appeal to a moral truth apprehended directly and prior to calculation.
Note the structure rather than the verdict. The badness is real, and it is real in terms of something. Every value claim is thereby converted into a relation between an outcome and a goal, and the goal is the agent’s. This is not the denial of value; it is the relocation of its truth-maker into the person. And Sterling’s Wanda case is the exact instrument for testing it: if being hurt is bad only in terms of a goal, and Wanda’s welfare is no goal of mine, then her being hurt is not bad in any sense that constrains me.
His P6 follows from P1 with something close to necessity, though the run records them as separate presuppositions. His clinical recommendation is that whichever of Stoicism or REBT makes more sense to the client is the one to immerse himself in, since each can dramatically improve a life; extended, on 18 September 2019, to encouraging a Muslim client to draw on his faith. Once value is indexed to goals, a framework is an instrument for reaching goals, and the correct framework is whichever serves. There is no further question about which is true, because the framework has been placed among the means.
What is lost is stated most economically in Sterling’s own 2015 formulation: without objective facts, Stoicism cannot call these impressions false. It can only say they are inconvenient, or culturally contingent, or personally unhelpful. The vocabulary survives entirely. False value judgment can still be said. What it now means is judgment that is not working out.
VIII. Two Softer Forms, Inside the Practice
The doctrinal collapse is rare. Two paraphrases of it are not, and both occur among people who would affirm Th10 without hesitation.
The therapeutic paraphrase. The correction gets stated as that judgment isn’t serving you. The words are close to the corpus’s own and the logic is inverted. If serving is the standard, then a judgment that serves is to be kept, whatever its truth — which is the beneficial lie arriving in the phrasing of the correction itself. The corpus form is a verdict: this judgment does not correspond. Whether the agent is better off for holding it is a separate question and not the one being answered.
The school paraphrase. Th10 gets stated with a qualifier: the Stoic holds that only virtue is good. The qualifier looks like scholarly care and is a change of subject. It converts a claim about reality into a report of a position. Sterling’s 2015 formulation refuses the softening in advance — the foundation of Stoicism is that things not in our control are neither good nor evil, and these are taken to be objective facts. A man who can only state the doctrine in the third person has already declined to assert it.
IX. Why This Is Not Arrogance
The objection arrives predictably: to hold that value claims are objectively true or false is to claim a certainty about morals that nobody has.
The truth-knowledge distinction disposes of it. Sterling’s Pepsi can establishes that a proposition’s having a determinate truth-value is independent of anyone’s being able to determine it. Holding that there is a fact about whether health is genuinely good is not a claim to know every such fact, or to be immune from error about them, or to be better placed than one’s neighbour.
His own 2021 remark cuts the other way from arrogance. The reason moral truths are hard is that we have spent our lives acquiring the habit of calling good whatever seems to benefit us, and so we deny obvious moral truths when they are inconvenient. That is offered as a general condition, not a description of other people. The realist’s position is that there is something to be wrong about, and that he is very likely wrong about a good deal of it.
The genuinely immodest position is the other one. To say that value is indexed to my goals is to say that my goals settle the matter — which is a larger claim about myself than any realist makes.
X. Holding It Open in Practice
Three markers, in ascending order of usefulness.
Watch for the indexing clause. For me, in terms of, given what I want, from where I’m standing. Each converts a claim into a relation, and each is nearly always added silently.
Ask what would falsify it. Not whether it is true — that is the harder question and comes later — but what kind of thing would settle it. The answer identifies which sort of sentence was uttered.
And state the correction as a verdict. Not I’d be better off not thinking this but this is not the case. The first is a preference about preferences. The second is the only form in which anything gets corrected.
XI. What Remains Open
Th10’s own warrant. The corpus’s finding is that Th10 admits presentation and defeater removal and nothing else — the system’s most consequential proposition is its least transmissible. That is a consequence of ethical intuitionism rather than an embarrassment, but it means this distinction is asymmetrical in an important way: it establishes that value claims answer to facts, and it cannot establish, to someone who does not see it, which facts. The distinction defends the form of the claim. The fifth commitment carries the content.
Where correspondence stops and intuitionism begins. Correspondence tells you what makes a value judgment true; it is silent on how anyone knows which are. Sterling joins the two at the end of the 2015 realism message — if there are objective moral facts, we must have some means of knowing them, they cannot be sensed, how do we know them? — and the answer belongs to a different commitment. The seam should be marked rather than papered over, since a practitioner who expects correspondence to deliver knowledge will conclude the system has failed him when it does not.
Whether agent-indexed realism is a coherent position. Edelstein’s P1 is recorded as good and bad being real but subjective, which the run maps as agent-indexed rather than eliminativist. Whether that is a stable third position, and whether it survives the Wanda case, is not settled by the run and was not the run’s question. It is worth settling, because it is the position most contemporary readers actually hold, and refuting subjectivism is not the same as refuting it.
False is not incoherent. Sterling is careful about this and the corpus sometimes is not. On 3 September 2019 he declines to say that Aristotle was absurdly self-contradictory in allowing that ugliness can impinge on eudaimonia, or that most people are engaged in unnoticed self-contradiction in valuing what they do not control. A false value judgment is an error about a fact, not a contradiction — the man who holds it has not violated logic, he has got something wrong. Documents that write as though the ordinary valuation were incoherent are claiming more than Sterling does, and are also less persuasive for it.
Theoretical foundations: Grant C. Sterling (Eastern Illinois University). Analysis and synthesis: Dave Kelly. Prose rendering: Claude (Anthropic). 2026.
0 Comments:
Post a Comment
<< Home