Moral Truth and the Limits of Preference
In this essay
Imagine taking your car to a mechanic who knows you cannot check the diagnosis. The repair is simple. He tells you it is complicated, charges for work the car does not need, and sends you home. You never discover the deception. He feels no guilt.
What, exactly, has gone wrong?
There is an easy answer about money: you paid more than you should have. But the question reaches beyond the bill. The mechanic used your dependence on his knowledge against you. He took the trust that made the transaction possible and turned it into an opportunity for exploitation.
Suppose he responds that he feels perfectly comfortable with what he did. Suppose his colleagues approve. Would either of those things make the deception right?
I think the answer is no. His comfort does not settle what he owes you, and our approval does not settle it either. The customer's claim to honest treatment gives us a reason to judge the mechanic's conduct, including a reason to correct our own first reaction.
That distinction is where I want to begin Plumb Lines. When we make a moral judgment, what makes it answerable to anything beyond our preferences?
A plumb line gives a builder a reference for vertical. The builder can dislike the result, misunderstand the instrument, or discover that the wall needs work. The reference remains available to correct the judgment. Morality is harder to examine than a wall, and the analogy supplies no proof that moral truths exist. It does, however, express the question behind this series: can our judgments be corrected by something we did not choose?
I believe they can. I believe some actions are wrong even when the people performing them approve, and that moral inquiry can bring us closer to understanding why. The moral realism I am defending holds that some moral claims are true because there are moral facts and reasons that do not depend for their validity on our evaluative attitudes. Those reasons explain why our judgments can be correct or mistaken. Our judgments answer to them; our approval does not create their authority.
Other accounts of morality can also affirm that an action remains wrong even if everyone approves of it. The dispute concerns what explains that independence. Realism begins with moral facts to which our attitudes should respond. Some of its strongest alternatives begin with an account of our attitudes and seek to explain how moral thought can sustain truth and correction. That disagreement deserves more attention than a contrast between principle and passing preference.
What our preferences can tell us
Preferences matter. They tell us something about the people making them. A desire for security may reveal vulnerability. Anger may call attention to an injury that everyone else has overlooked. Discomfort can be the beginning of moral attention.
Preferences also sometimes settle a question. If you like a particular flavor and I dislike it, we may have learned everything we need to know about which dessert each of us should order. Your enjoyment does not require my agreement.
The mechanic's deception puts a different demand on us. Saying that I dislike being cheated reports my attitude. Saying that he acted wrongly challenges his conduct. It offers a reason he should take seriously even if he enjoys the extra money and has no interest in pleasing me.
That is the difficult part of moral judgment. We expect reasons to reach beyond the person who happens to feel them. Someone else's vulnerability can place a demand on me when helping them is inconvenient. A promise can continue to matter after keeping it stops being attractive.
The existence of such demands is precisely what needs explaining. We cannot establish them merely by speaking forcefully or by pointing out how familiar they feel. But we should also resist explaining them away before we have examined them.
Consider the difference between changing a preference and correcting a judgment. I might once have preferred one kind of music and now prefer another. I do not necessarily owe my earlier self an explanation. But if I once thought taking advantage of an uninformed customer was acceptable, and now think it was wrong, I may understand the change as learning something about what I owed another person.
The important question is whether that understanding can be true. Can I have improved my judgment, rather than merely acquired a new attitude?
A serious alternative to moral realism
The suggestion that moral judgment begins in feeling deserves attention. Morality would be a strange part of human life if suffering never troubled us and generosity never moved us. Any account that leaves out emotion leaves out much of what it needs to explain.
David Hume gave sentiment a central place in moral judgment. In Book III, Part I of A Treatise of Human Nature 1, he argues that moral distinctions derive from sentiment rather than reason. Reason can inform us about circumstances and consequences, but it does not supply the sentiment through which we distinguish virtue from vice. His account asks how human sentiment makes moral evaluation possible; it does not reduce ethics to whichever impulse an individual happens to have.
A. J. Ayer later advanced a sharper account. In Chapter 6 of Language, Truth and Logic 2, he treats evaluative moral utterances as expressions of feeling, rather than factual assertions that are true or false. This view is called emotivism. Expressing disapproval differs from reporting that one disapproves: an expression is not a psychological description of the speaker. Ayer explicitly distinguishes his account from that kind of subjectivism.
Contemporary expressivists have developed much more ambitious accounts. Allan Gibbard connects judgments about what we ought to do with planning how to live in Thinking How to Live 3 (2003). Simon Blackburn calls his program quasi-realism: he seeks to explain how an account beginning with evaluative attitudes can sustain moral reasoning and claims about truth and objectivity. Essays in Quasi-Realism 4 (1993) develops that project.
These alternatives cannot fairly be summarized as permission to do whatever feels good. An expressivist can condemn the mechanic, give reasons for that condemnation, and urge him to change. In his accessible book On Truth 5 (2018), Blackburn places moral truth at the center of practical inquiry about how to live. The passage concerns reasoning broadly, but it helps explain why rejecting realism need not mean abandoning moral inquiry.
The disagreement concerns what ultimately explains those judgments and their authority. In the publisher's description of Meaning and Normativity 6 (2012), Gibbard's later account is presented as allowing the strongest forms of expressivism and nonnaturalist realism to converge in their theses while differing in their explanations. Nonnaturalist realism treats moral truths as irreducible to the facts described by natural science. Agreement about what is wrong can leave a substantial disagreement about why the judgment is true.
Realism has to earn its answer by explaining why we should begin with independent moral facts, rather than an account of evaluative attitudes that can recover the same moral judgments. Showing that a simple account of passing feelings is inadequate leaves several serious alternatives on the table.
Giving reasons for a moral judgment
Return to the mechanic. We can offer an argument:
- Knowingly deceiving a customer into paying for unnecessary work is wrong.
- The mechanic knowingly deceived a customer into paying for unnecessary work.
- Therefore, the mechanic acted wrongly.
The argument has a form we recognize. We can challenge its moral premise or investigate its factual premise. Perhaps the mechanic made an honest diagnostic error. Perhaps the work was necessary after all. Those discoveries would change our judgment because the reasons for it would change.
Ayer allows this kind of argument when people already share the relevant moral commitment. Establishing that an action falls within a category they condemn can change their response. What he denies is that argument can establish the validity of the underlying evaluative principles themselves. If agreement about the facts leaves a disagreement in basic values, his account offers no further argument to show which system of values is correct. Contemporary expressivists take on the more ambitious task of explaining moral reasoning itself.
We can also ask why the first premise should be accepted. The customer relied on expertise he could not independently assess. The deception denied him the information he needed to decide how to spend his money. These details give substance to the claim that he was wronged.
Moral reasoning therefore asks more of us than the announcement of a reaction. We must attend to the person affected and explain why the circumstances matter. We should be willing to discover that an accusation we enjoyed making was false.
This feature of moral language creates a challenge known as the Frege–Geach problem, associated with Peter Geach's “Assertion” 7 (1965). How do moral claims retain their meaning inside arguments, including statements that nobody is asserting? A person can consider the conditional “If that deception was wrong, encouraging it was wrong” while remaining uncertain about its premise. They have not thereby expressed disapproval of the deception. An adequate account must explain how the moral claim functions in that conditional and in valid reasoning.
Expressivists have worked seriously on that challenge. The existence of moral arguments does not, by itself, establish moral realism. Even someone who thinks our moral beliefs are systematically mistaken can recognize their logical form.
What matters here is the discipline that giving reasons introduces. My reaction becomes accountable to considerations I may have neglected. The customer's interests do not disappear when the mechanic describes his satisfaction. Realism understands this accountability as a response to reasons that apply whether or not either person recognizes them. An expressivist owes a different explanation of how those interests constrain judgment. The argument's logical form does not choose between those explanations.
When an institution gets morality wrong
The stakes become clearer when approval is institutional rather than individual.
From 1932 to 1972, the U.S. Public Health Service conducted its study of untreated syphilis at Tuskegee. The study began with 600 Black men, 399 with syphilis and 201 without it. Researchers did not obtain informed consent and told the men they were being treated for “bad blood.” Participants were not offered penicillin treatment even after it became available. The CDC's history of the study 8 records these facts.
The central moral failure was not a technical mistake about the disease. The men were denied informed participation and available treatment. Their dependence on the people directing the study made the deception especially grave.
When we condemn what happened, I think we mean that the participants were wronged during the study. Their claim to truthful treatment did not begin when a newspaper report brought the study to public attention. Institutional authorization had failed to recognize a demand that already mattered.
This is one reason moral reform matters to the question of realism. Reform can require criticizing the standards by which a community currently judges itself. The fact that a practice has approval is among the things a reformer may need to challenge.
An expressivist can make that challenge too. It would be unfair to insist that anyone who rejects realism must regard the study as acceptable in its time. The judgment that the men were wronged is something both views can affirm and must explain. The case does not settle which explanation is better merely because its injustice is so clear.
I find the realist explanation more persuasive because it locates the reason for correction in the men and their treatment. Their vulnerability, their need for care, and their dependence on truthful information supplied reasons against the study regardless of how the researchers evaluated those considerations. Public exposure changed what people knew and were willing to acknowledge. On the realist account, it did not create the participants' claim to honest treatment.
An expressivist can endorse that description, including its independence from approval. The further question is whether the authority of those reasons is best understood as a moral reality our judgments recognize, or explained through a theory that begins with evaluative attitudes. My attraction to realism is that the participants' reasons explain why a community needed to change its commitments. Establishing that it is the better account requires comparing how well both theories explain that need for correction.
The case also warns against complacency. We should ask which practices we now defend because the people around us approve. If moral reform is possible, our own institutions may be among those that need it.
Conscience can require correction
Moral experience often presents itself as recognition. When we see someone deliberately humiliated, the suffering can seem to call for a response before we have decided what to feel. When we remember a betrayal, we may recognize an obligation to repair it even while wishing the memory would go away.
That experience contributes to the appeal of realism. There appears to be something about the person and the act that demands our attention. The judgment feels answerable to its subject.
But the strength of a reaction cannot certify its accuracy. People can feel guilty about things they had no duty to do. They can also feel innocent while treating others badly. Conscience is affected by what we have been taught, which injuries we have learned to notice, and which people we have learned to ignore.
Taking conscience seriously therefore includes examining it. If an action seems obviously wrong, we can ask what features of the action make it wrong and whether we have understood them. If we feel no concern, we can ask whether our comfort depends on failing to attend to the person affected.
The realist's position does not make those questions unnecessary. It gives them a purpose. A judgment can be confident and mistaken. Learning to judge well requires more than increasing the intensity of our conviction.
This is why the plumb line image matters to me. A useful reference can expose an error in the person using it. Believing that moral truth exists should make me more willing to examine the wall I have built.
The questions realism still has to answer
Moral realism faces a serious problem of knowledge. If moral claims can be true independently of our approval, how do we distinguish a reliable judgment from a prejudice that happens to feel authoritative?
Disagreement makes the question harder. Sometimes people disagree because they have different beliefs about what happened or what an action will cause. Sometimes they differ about what matters, even after the facts are clarified. We cannot honestly explain every moral conflict as a misunderstanding that better information will immediately dissolve.
Disagreement alone does not demonstrate that there is no truth. It does show that claiming to know the truth requires work. The realist owes an account of how reasoning, experience, and attention can improve judgment.
Hume's warning about the move from “is” to “ought” appears in the same section of the Treatise as his argument that moral distinctions do not derive from reason. Discovering that a behavior is common, useful to an institution, or advantageous to its practitioners does not by itself establish that it is right. An argument needs to explain why its premises support the obligation it claims.
That warning need not threaten every form of realism in the same way. A nonnaturalist realist can accept that moral conclusions require moral premises. Naturalistic realists, who understand moral facts as part of the natural world, must explain why the particular natural facts they identify have moral significance. The is–ought passage raises a demand for an account of that connection; citing it does not settle the dispute.
Evolutionary explanations raise a further challenge. If our evaluative dispositions were shaped by pressures favoring survival and reproduction, why should we expect them to track independent moral truths? Sharon Street presses that question in A Darwinian Dilemma for Realist Theories of Value 9. It is a challenge to the reliability of judgment, not simply an announcement that morality has a history. A realist needs a reason to trust the relevant capacities, or an account of how reflection can correct their distortions.
Street also develops a constructive alternative, constructivism about reasons 10, in which reasons ultimately depend on an agent's evaluative standpoint while particular judgments remain open to correction. Her criticism of realism is part of an account of how normative judgment works.
Those problems keep this opening essay from being a final proof. The view I am defending is that realism offers a persuasive starting point for understanding moral authority: reasons arising from people's circumstances explain why our attitudes should change. The mechanic and Tuskegee cases show what needs explaining. A fuller defense must compare that account with alternatives that can reach the same moral conclusions.
That account comes with a task: we must investigate how to judge reliably. The existence of moral truth would give us something to seek, while leaving us responsible for how we seek it.
My Latter-day Saint starting point
I approach these questions as a Latter-day Saint. That commitment helps explain why I take moral knowledge seriously, and it belongs openly in the discussion.
In Moroni 7:16 11, the Spirit of Christ is described as given to every person so that they may discern good from evil. I read that as an invitation to expect moral insight beyond the boundaries of my own religious community. Someone need not share my vocabulary or my membership to recognize something I should learn from.
The nearby verses matter just as much. Moroni 7:18 warns against judging wrongly, and verse 19 calls for diligent searching. I understand the gift as received and interpreted by fallible people, whose habits and cultures can distort their judgments. In my reading, access to light brings responsibility for the quality of our judgment. It gives me no warrant to treat every strong religious feeling as a finished moral conclusion.
Paul's discussion of conscience in Romans 2:14–15 12 similarly places moral awareness among people outside the community possessing the law. These passages encourage me to listen for truth wherever it is recognized.
Doctrine and Covenants 93:24–30 13 also gives truth a place beyond personal preference. Verse 24 describes truth in terms of knowledge of things as they are, were, and will be; verse 30 speaks of truth's independence within its sphere. I take these passages to support an understanding of truth that we receive and learn to recognize. They do not, by themselves, settle every philosophical question about the relationship between God, moral law, and obligation.
A reader who does not accept these scriptures has no reason to treat them as a shared premise. My philosophical argument must still be assessed on its reasons. Their place here is to make my own starting point visible and to explain a responsibility I believe follows from it.
For me, repentance makes that responsibility concrete. It involves admitting that my previous judgment or conduct needs correction. I can sincerely believe I acted well and still need to reconsider what I owed another person. Faith ought to make that recognition possible.
Beginning with our own judgments
The question “Is this right, or do I merely want it?” is a useful beginning. It is also incomplete. Sometimes our desires are well formed. We want something because we recognize its goodness. Sometimes what we call principle is a preference protected by more impressive language.
The task is to examine the reasons connecting our judgment to the situation. T. M. Scanlon's What We Owe to Each Other 14 (1998) offers a helpful account of justification to others. His contractualism asks whether an action would be permitted by principles that no one could reasonably reject as a basis for informed, unforced general agreement. That test requires us to examine the reasons people have for objecting; actual acceptance does not settle the matter.
Drawing on that concern with justification, alongside questions about factual error and my own interests, I want to ask:
• Which facts would change my judgment if I learned I had them wrong?
• What reason could I offer to the person who bears the cost of my decision?
• Could they reasonably reject the principle I am using to justify my decision?
• What might I be overlooking because the present arrangement benefits me?
The third question asks about the affected person's reasons for rejecting a principle, which requires more than imagining whether I would accept the decision in their position. These prompts draw on Scanlon without reproducing his whole theory. They do not resolve every disagreement, but they make it harder to treat my own comfort as the measure of what I owe.
The mechanic's satisfaction cannot remove the customer's claim to honesty. An institution's approval cannot excuse its failure to treat people justly. And my confidence cannot establish that I have understood either case correctly.
I believe moral realism gives a persuasive explanation of those observations by treating our judgments as answerable to reasons whose authority our approval does not create. The point of seeking a plumb line is to discover where our judgment needs work.
That is the work I want Plumb Lines to take up. The next question follows directly: if getting morality right involves more than getting our preferences aligned, how do we come to know what is right? A fair procedure may help us reason together. We still need to ask what makes the procedure fair and whether its outcome deserves our trust.
Sources
1. David Hume. A Treatise of Human Nature (1739–1740), Book III, Part I, Sections I–II. Project Gutenberg edition.↩
Read source at gutenberg.org ↗
2. A. J. Ayer. Language, Truth and Logic, Chapter 6, “Critique of Ethics and Theology.” Second edition (1946); Dover reprint (1952). ISBN 9780486200101.↩
Read source at store.doverpublications.com ↗
3. Allan Gibbard. Thinking How to Live. Harvard University Press, 2003.↩
4. Simon Blackburn. Essays in Quasi-Realism. Oxford University Press, 1993.↩
Read source at academic.oup.com ↗
5. Simon Blackburn. On Truth. Oxford University Press, 2018. Linked excerpt: “Fake news: a philosophical look at biased reasoning,” OUPblog, September 1, 2018.↩
6. Allan Gibbard. Meaning and Normativity. Oxford University Press, 2012. Publisher description reproduced by the University of Michigan Department of Philosophy, December 11, 2012.↩
Read source at lsa.umich.edu ↗
7. Peter T. Geach. “Assertion.” The Philosophical Review 74, no. 4 (1965): 449–465. DOI: 10.2307/2183123.↩
8. Centers for Disease Control and Prevention. The U.S. Public Health Service Untreated Syphilis Study at Tuskegee. Historical timeline.↩
9. Sharon Street. “A Darwinian Dilemma for Realist Theories of Value.” Philosophical Studies 127, no. 1 (2006): 109–166. DOI: 10.1007/s11098-005-1726-6.↩
10. Sharon Street. “Constructivism about Reasons.” In Oxford Studies in Metaethics, volume 3, edited by Russ Shafer-Landau, 207–246. Oxford University Press, 2008.↩
Read source at academic.oup.com ↗
11. Book of Mormon. Moroni 7:16, 18–19. The Church of Jesus Christ of Latter-day Saints.↩
Read source at churchofjesuschrist.org ↗
12. New Testament. Romans 2:14–15. The Church of Jesus Christ of Latter-day Saints.↩
Read source at churchofjesuschrist.org ↗
13. Doctrine and Covenants 93:24, 30. The Church of Jesus Christ of Latter-day Saints.↩
Read source at churchofjesuschrist.org ↗
14. T. M. Scanlon. What We Owe to Each Other. Belknap Press of Harvard University Press, 1998.↩