Can We Be Good Without God?

Good Without God?

“There are no objective values.”

J. L. Mackie, Ethics: Inventing Right and Wrong

The question in the title is not the interesting one, and it should be answered at once so that it stops getting in the way. Can a person be good without believing in God? Obviously yes. Unbelievers keep their promises, raise their children well, give blood, and go to prison for causes that cost them everything; believers embezzle and abandon their families. Anyone who has lived a while knows this. And the Catholic tradition has never claimed otherwise — it holds that the moral law is written on the heart and knowable by reason, so that people who never heard of Sinai “do by nature what the law requires” (Rom. 2:14–15). A Christian who argues that atheists cannot be decent is contradicting his own catechism.

The real question is not who behaves well but what makes a moral claim true, and where the obligation in it comes from. When we say cruelty is wrong we do not seem to be reporting a preference, or a majority view, or a feeling that natural selection installed in us because it helped our ancestors cooperate. We seem to be saying something is the case — and that it binds, whether or not anyone agrees, and whether or not obeying it costs us. That word “binds” is the whole difficulty.

The dispute is more interesting than the popular version suggests, because unbelievers are divided about it. Some accept the consequence cheerfully: there are no objective values, morality is a human invention, and it works well enough. Others insist that objective moral facts exist and require no God to hold them up. Both replies are given below in their own words, and each has real force. What follows is an argument with the second position mainly, and it concedes a great deal to the first.

Let’s look at what the word “ought” is doing, and ask … If moral obligation is a law, can there be a law where there is nobody to lay it down? And if evolution explains why we feel bound, does it show that we really are — or only that we would feel this way either way?

I invite your engagement with me on these questions. You may leave comments below, but please be sure to read our policy on commenting before doing so.

Atheist/Skeptic View

Jump to the Catholic View ↓

The skeptic holds that morality needs no God — not to be known, not to be practiced, and not to be true. We are social animals with a long history of living in groups that punish defectors; conscience is what that history feels like from the inside. Ethics can be built on the things human beings actually care about, chiefly flourishing and the avoidance of suffering, and adding a deity to the account explains nothing that was not already explained, while raising difficulties of its own.

What They Affirm

Conscience is not distributed by creed. If morality required belief in God, unbelievers would be visibly worse than believers, and they are not. Secular Scandinavia is not a moral wasteland; prisons are not disproportionately atheist. Whatever the source of moral knowledge, it is plainly available to people who hold no religion at all.1

The Euthyphro dilemma, the skeptic holds, has never been answered. Socrates asked whether the holy is holy because the gods love it, or whether they love it because it is holy. Put to the theist: is an act good because God commands it, or does God command it because it is good? If the first, morality is arbitrary — had God commanded cruelty, cruelty would be a duty. If the second, goodness is a standard independent of God, and God has become a messenger rather than a foundation.2

Evolution, on this account, explains the moral sense without remainder. Kin selection explains why we sacrifice for relatives; reciprocal altruism explains why we help strangers who may help us later; group selection and reputation explain the rest. A creature built this way would feel exactly what we feel — guilt, indignation, the sense that some things are simply not done — and would feel it whether or not anything answered to it.3

Grounding morality in God’s nature only moves the question. The standard reply is that the good is not something God consults but what God is. The critic answers: why is that nature good? Either there is a standard by which it counts as good, and we are back to the second horn, or “good” simply means “like God,” which on this reading makes the claim that God is good an empty one.4

And the record of divinely grounded morality is not reassuring. Scripture commands the destruction of the Canaanites and regulates slavery rather than forbidding it; the churches defended both for centuries. The moral advances of the modern world — abolition, religious toleration, the rights of women — were resisted by religious authority as often as they were led by it. A morality anchored in revelation, the skeptic concludes, has proven quite capable of anchoring atrocity.5

Two Ways of Cashing It Out

Either there are no objective values — and, Mackie argues, we can live with that. On this account moral claims are not reports of fact at all; nothing in the world answers to them, and when we call cruelty wrong we are expressing an attitude and recruiting others to it. That is not nihilism in practice. Morality remains a human institution we have excellent reasons to build and maintain, in the way that law and money are human institutions and none the worse for it.6

Or there are objective values, and they stand on their own. The alternative is to say that some ethical truths are simply basic — necessary, non-natural facts that could not have been otherwise, standing in no need of a maker any more than the truths of mathematics or logic do. That pain is bad is not made true by anyone’s decree, God’s included. It is one of the things that is the case, and asking who is responsible for it is a confusion of categories.7

Endnotes
  1. The point is made throughout the popular literature; for a careful version see Gregory S. Paul, “Cross-National Correlations of Quantifiable Societal Health with Popular Religiosity,” Journal of Religion and Society 7 (2005), and the discussion in Phil Zuckerman, Society Without God (New York University Press, 2008). ↩
  2. Plato, Euthyphro 10a. The modern use of the dilemma against theistic ethics is standard; see Bertrand Russell, Why I Am Not a Christian (1927), and Michael Martin, Atheism: A Philosophical Justification (Temple University Press, 1990), ch. 10. ↩
  3. Michael Ruse and E. O. Wilson, “The Evolution of Ethics,” New Scientist 108 (1985), and Ruse, Taking Darwin Seriously (Blackwell, 1986), where morality is described as an adaptation and, famously, as an illusion fobbed off on us by our genes to make us cooperate. See also Frans de Waal, Primates and Philosophers (Princeton, 2006), on the animal roots of fairness and empathy. ↩
  4. The objection to the “God’s nature” reply is pressed by Erik J. Wielenberg, Robust Ethical Realism, Non-Naturalism, and Normativity (Oxford, 2014), and by Wes Morriston, “God and the Ontological Foundation of Morality,” Religious Studies 48 (2012). ↩
  5. See Deuteronomy 7:1–2 and 20:16–17; Leviticus 25:44–46; Ephesians 6:5. The historical argument is made at length in Steven Weinberg’s much-quoted remark on good people and evil deeds, and in Christopher Hitchens, God Is Not Great (Twelve, 2007). ↩
  6. J. L. Mackie, Ethics: Inventing Right and Wrong (Penguin, 1977), which opens with the flat claim that there are no objective values and then constructs an account of morality as a human device. Mackie calls his position an error theory: ordinary moral discourse asserts objective values, and is systematically mistaken in doing so. ↩
  7. Wielenberg, Robust Ethical Realism, defending necessary, non-natural, brute ethical facts that supervene on natural properties and require no theistic grounding; see also his Value and Virtue in a Godless Universe (Cambridge, 2005). ↩

Christian View

The Christian claim here is much narrower than it is usually taken to be, and stating it narrowly is the only way to state it honestly. It is not that unbelievers are wicked, nor that religion makes people behave, nor that anyone needs Scripture to know that cruelty is wrong. It is that one particular feature of morality (its bindingness, the fact that we are obliged and not merely inclined) is very hard to account for if the world is matter and nothing else — the materialist picture most unbelievers in practice hold, though atheism as such entails neither it nor any particular account of morality — and that the difficulty is not solved by any of the usual replies.

What This Argument Does Not Claim

Four concessions, all of them real, and none of them grudging. Unbelievers do know the moral law; the Church says so, holding that it is written on the heart and reached by reason, so that those who never heard the commandments still “do by nature what the law requires” (Rom. 2:14–15). Unbelievers often keep it better than believers do. Evolution really has shaped our moral emotions, and there is nothing embarrassing in that for a Christian who thinks God made the world by secondary causes. And the record of religious people is in places disgraceful: the defenders of slavery quoted Scripture, and the argument of this page would have told against them exactly as it tells against anyone else.

What remains after those concessions is a question about the grounds, not about the behavior. Both sides can see the moral law. The dispute is over what sort of thing it is that both sides are seeing.

The Word That Needs Explaining

The sharpest statement of the problem was made not by an apologist but by Elizabeth Anscombe, a Catholic philosopher writing for other philosophers in 1958, and it is a strange fact that it is quoted more often by ethicists than by believers. Anscombe noticed that our moral vocabulary is legal vocabulary. Obligation, duty, guilt, binding, and the peculiar emphatic ought that appears in moral contexts and nowhere else, are the words of a courtroom, and they entered ethics from a conception of morality as divine law.

Her point was about coherence, not piety. A verdict requires a court; a command requires someone who commands. If the lawgiver is removed but the legal vocabulary is retained, we are left using words that have outlived the framework which gave them sense, as though a country abolished its legislature and went on issuing statutes. Anscombe’s conclusion was severe, and it was addressed to secular philosophers: the concept of moral obligation ought to be jettisoned if it can be, because it is a survival from a system they have abandoned. That is not a Christian scoring a point. It is a description of the debt.

Euthyphro Has a Third Option

The dilemma is a real one against the crude view that morality is whatever God happens to decree, and Christians who hold that view deserve it. But the classical position has never been that. It is not that God consults a standard of goodness above Him, nor that He invents one by fiat, but that God is goodness, not a being who possesses the property, but the reality in which anything else that is good participates.

The objection that this makes “God is good” empty mistakes the kind of claim being made. When we say that a measurement is accurate we mean it corresponds to the standard, and saying the standard is accurate is not a compliment to it but a statement of what accuracy consists in. Nor is the anchor arbitrary: a nature cannot be otherwise, so the worry that God might have commanded cruelty asks what would follow if God had a different nature, which is like asking what would follow if a triangle had four sides. The horns of the dilemma are “arbitrary command” and “independent standard,” and this is neither. A residue of the dilemma remains and should be admitted: to call God’s nature good, the objector says, is either to apply a standard to it or to say nothing. The reply is that the standard and the nature are one reality under two descriptions, as the meter and the bar in Paris once were — a claim that can be argued against, but not one that leaves the dilemma’s two horns as the only options.

The Commands the Skeptic Quotes Back

The skeptic panel has already made the obvious rejoinder: Scripture itself records commands to destroy the Canaanites, and rules that regulate slavery rather than forbid it (Deut. 7:1–2; 20:16–17; Lev. 25:44–46). If God’s nature cannot command cruelty, what are those?

Notice first what the objection borrows. To indict the conquest texts as atrocity is to measure them against a standard of goodness that stands above ancient Israel, above the churches that defended slavery, above every age’s conscience — which is precisely the objective standard this argument says exists and asks the skeptic to ground. The objection’s force is evidence for the premise.

What remains is a genuine question about texts, and the Church answers it as one. Catholic teaching reads the Old Testament as a divine pedagogy addressed to a people in a brutal age — the Catechism says plainly that its books contain “matters imperfect and provisional” (CCC 122) — and brought to its term in Christ, who forbade retaliation and told Peter to put up the sword. And within the conquest narratives the “utter destruction” formula reads as the war-rhetoric of its time: cities said to be annihilated are inhabited again pages later, and Joshua’s “left none remaining” stands beside Judges’ Canaanites everywhere. However those texts are finally weighed, they pose a question about how God condescended to speak — not a counterexample to what goodness is. The standard by which the skeptic condemns them is the standard whose existence this page defends.

Explaining the Feeling Is Not Explaining the Obligation

The evolutionary account is almost certainly true, and it is beside the point, not because it is too weak but because it answers a different question. It tells us why creatures like us have moral emotions. It cannot tell us whether the content of those emotions is correct, because a story about the causes of a belief is not evidence for the belief. We can explain why humans fear the dark without settling whether anything is in it.

And the account has a sting its friends do not always notice. If our sense that some things are simply forbidden was installed because it helped our ancestors cooperate, then it was installed for its usefulness and not for its accuracy, and we have no reason to expect it to track any truth. Michael Ruse, who has argued the evolutionary case as forcefully as anyone, accepts this and says plainly that ethics is an illusion put in place by our genes to make us cooperate. That is not a defense of morality against the theist. It is an agreement with the theist about what morality would have to be if there is no God, offered by someone content with the result.

Free-Standing Values Are a Large Bill

The more serious reply says that objective moral facts exist as brute necessities, like the truths of mathematics, requiring no author. This deserves respect; it is not evasion, and it may be the strongest position available to an unbeliever. But notice what has to be bought.

Mathematical truths do not tell anyone to do anything. Moral truths do, and obligation looks like a relation between persons rather than a property floating free of them. I can be obliged to a neighbor, to a promise, to a child; it is much harder to say what it would mean to be obliged to an abstract fact, or how a necessary truth about pain could impose a demand on me and hold me guilty when I ignore it. This is the point C. Stephen Evans presses: duties are the sort of thing that get laid on someone by someone, so an account of obligation without a person in it has kept the word and dropped the thing it named.

The contemporary versions deserve to be named, because they are serious and their costs itemize differently. Erik Wielenberg’s “robust realism” bites the bullet whole: moral facts are non-natural, brutely necessary, and causally inert — and our brains, shaped by forces with no interest in moral truth, are said nonetheless to line up with them, a lucky alignment his own book must labor to defend. Reasons-based realism in the line of Parfit and Scanlon translates goodness into facts about what we have reason to do — which relocates the mystery rather than solving it, since a normative reason binding every rational creature is exactly the queer item in a smaller font. And constructivism, which builds morality from what agents would agree to, delivers obligation at the price of objectivity: what we construct we can reconstruct, and the tyrant’s community constructs otherwise. Each position buys the argument’s premise — real, binding moral truth — on credit. The theist can say where such truth lives and what gives it the right to command: a Person whose nature is the Good.

The Best Atheist Philosopher Grants the Premise

It is worth noticing who agrees. J. L. Mackie, the most formidable atheist writer on this subject, argued that objective values would be entities of a very queer sort, unlike anything else in the universe, requiring a faculty of moral perception unlike any other faculty we have. He was right, and he drew the consistent conclusion: since the world contains no such things, there are no objective values, and ordinary moral talk is systematically mistaken.

The theist accepts Mackie’s premise entirely and denies his minor. Objective values are queer: strange, unlike the furniture of a purely material world, exactly as he said. Where the two part company is that Mackie inferred there are none, while the theist observes that they show every sign of existing and asks what sort of universe would contain them. The dispute is not about whether moral facts fit comfortably into naturalism. Both sides agree they do not.

What the Denial Costs

The price is usually paid quietly, so it is worth setting out. “Torturing a child for entertainment is wrong” becomes a statement about the speaker rather than the act. Moral progress becomes unintelligible, since progress is movement toward something, and if there is no standard there is only change, and abolition was not an improvement on slavery, merely a later fashion. And the lone reformer, who is right while his entire society is wrong, has to be redescribed as someone deviating from his community’s values, which is precisely what his enemies said at the time.

Almost nobody can live in that account, and the ones who defend it do not try to. The man who says morality is a useful fiction still resents being cheated, and resents it as a wrong done to him rather than as a breach of a convention he happens to favor. That is not a proof. But when a theory cannot be inhabited even by the people who hold it, that is worth something.

The Reply in Brief

Grant everything that should be granted, that unbelievers know the moral law and often keep it better, that evolution shaped our moral feelings, that religious people have used God to justify appalling things. One thing is still unaccounted for. Obligation is law-language, and law is the sort of thing somebody lays down. Explaining why we feel bound is not explaining that we are; brute values keep the vocabulary of duty while removing the person to whom duty could be owed; and the ablest atheist in the field agreed that objective values would be very queer things indeed.

So the argument is not that the unbeliever cannot be good. He can, and the Church says he can. It is that when he is good he is obeying something he cannot quite account for, and that the law he finds already written, which he did not put there and cannot repeal, is better explained by there being someone who wrote it.

The full essay works through Anscombe’s 1958 argument in detail, sets out Robert Adams’s account of God as the Good itself against the Euthyphro dilemma, examines the evolutionary debunking argument and what it does to the debunker’s own beliefs, takes up Wielenberg’s brute ethical facts at their strongest, and asks what natural law implies about the atheist who keeps a commandment he does not believe in.

Summary

Can a person be good without believing in God? Yes, and no Catholic should say otherwise: the moral law is written on the heart and reached by reason, so that those who never heard the commandments “do by nature what the law requires” (Rom. 2:14–15). The argument here is not about who behaves well, but about what obligation is, and what sort of world contains it.

The moral vocabulary is legal vocabulary. Duty, guilt, binding, and the emphatic ought that turns up in ethics and almost nowhere else are the words of a courtroom. Elizabeth Anscombe argued in 1958 that they entered ethics from a conception of morality as divine law, and that keeping them after discarding the lawgiver leaves us using terms that have outlived the framework which gave them sense, like a country that abolished its legislature and went on issuing statutes. She was writing to secular philosophers, and told them to drop the concept if they could.

Euthyphro has a third option. The dilemma defeats the crude view that good means whatever God decrees, and Christians who hold that deserve it. The classical position is neither horn: God does not consult a standard, nor invent one, but is goodness, the reality in which whatever else is good participates. Calling the standard good is not flattery of it but a statement of what goodness consists in. And a nature cannot be otherwise, so the worry that God might have commanded cruelty asks what would follow if a triangle had four sides. A residue remains — is calling God’s nature good applying a standard to it? — and the answer is that standard and nature are one reality under two descriptions.

Explaining the feeling is not explaining the obligation. The evolutionary account of moral emotions is very likely true and answers a different question, since the causes of a belief are not evidence for it. It also cuts the wrong way for its friends: a sense installed because it aided cooperation was installed for usefulness, not accuracy. Michael Ruse accepts this and calls ethics an illusion our genes fobbed off on us, which is not a reply to the theist but an agreement about what morality would be if there were no God.

Free-standing moral facts keep the word and lose the thing. The strongest unbelieving reply is that ethical truths are brute necessities like mathematical ones, needing no author. But mathematical truths do not tell anyone to do anything. Obligation looks like a relation between persons: one can be bound to a neighbor, a promise, a child. What it means to be bound to an abstract fact, or how a necessary truth about pain could hold a person guilty, is the question C. Stephen Evans presses, and it does not have a comfortable answer.

The ablest atheist on the subject granted the premise. J. L. Mackie argued that objective values would be entities of a very queer sort, unlike anything else in the universe, knowable only by a faculty unlike any other we possess. The theist agrees entirely. Both sides hold that moral facts fit badly into a purely material world; Mackie concluded there are none, while the theist observes that they show every sign of existing and asks what sort of universe would contain them.

And the denial is expensive. “Torturing a child for entertainment is wrong” becomes a remark about the speaker. Moral progress becomes unintelligible, since progress is movement toward something and abolition would be merely a later fashion than slavery. The lone reformer, right while his whole society is wrong, has to be redescribed as a deviant from his community’s values, which is exactly what his enemies called him.

In short: the unbeliever can be good, and often is. But when he is good he is keeping a law he did not lay down and cannot repeal, and whose bindingness he cannot easily explain. That the law is there is not in dispute between us. The question is only whether anyone wrote it.

For a more complete essay on this topic, download Can There Be Good Without God? (.docx) or (.pdf)

Next: What Is the Bible?