All Truth Is Conditional

Beyond absolutism and relativism

Water boils at 100°C. It is hard to imagine a safer candidate for a plain, simple truth. Yet take it up a mountain and it fails. Add a little salt and it fails. State it fully and the simple truth turns out to be something else: water boils at 100°C at sea level, under standard atmospheric pressure, assuming pure H₂O, measured with calibrated instruments, ignoring impurities and experimental error. The conditions extend much further than the casual claim suggests, and the same is true everywhere you look:

• Objects fall at 9.8 m/s² — only near Earth’s surface, ignoring air resistance, neglecting the relativistic curvature of spacetime.
• The angles of a triangle sum to 180° — only within Euclidean geometry, which itself presupposes a flat metric and the parallel postulate.
• The sun rises at 6 a.m. — only at a given location and date, assuming the Gregorian calendar, a conventional definition of sunrise, and no atmospheric refraction.

Every “simple truth” is compressed shorthand for an entire lattice of background assumptions. Notice what this observation does not say. It does not say the claims are false, and it does not say they are matters of opinion. Water really does boil at 100°C under the stated conditions; that is a fact about the world, not about us. What the observation says is that the truth of the claim lives in the whole conditional, not in the bare sentence. Taking that seriously leads to a position I call Conditionalism, and it is the escape route from the oldest false choice in philosophy.

The False Choice

Public debate about truth is trapped between two dead ends. Absolutists claim truth is context-independent, universal, and eternal. Relativists claim truth is contingent on culture, perspective, or identity. Each side sustains itself mainly by pointing at the other: the absolutist warns, correctly, that relativism corrodes rational discourse and morality; the relativist observes, correctly, that every alleged absolute turns out to rest on somebody’s unexamined assumptions.

Alan Rome speaks for many defenders of absolute truth when he argues that the cure for relativism’s corrosion is a return to belief in absolutes.¹ He is right about the disease and wrong about the cure — resurrecting absolutes repeats relativism’s mistake in reverse. Absolutism denies the hidden assumptions it rests on. Relativism denies the possibility of meaningful discourse across perspectives. Both positions collapse under their own weight, and for the same underlying reason: both treat “true, full stop” and “true for me” as the only options on the table. They are not.

Why Relativism Fails

The strongest global relativism is self-undermining when it asserts, without qualification, that every truth is merely relative. More local forms—about taste, standards, or conceptual schemes—need not make that mistake. Conditionalism’s objection is therefore not that every view called relativist refutes itself; it is that perspective alone cannot determine a claim’s truth without specifying what is held fixed and what evidence can correct the evaluation.

Culture-relative description also cannot, by itself, ground cross-cultural critique. Condemning slavery requires stated evaluative premises and facts about agency, suffering, consent, and coercion. Those premises can be defended across vantage points; their force does not come from pretending no premises are involved.

Why Absolutism Fails

Absolutism is incoherent for the complementary reason: it pretends to speak without conditions while smuggling them in. “Murder is always wrong” sounds unconditional. But it silently depends on a definition of murder (which killings count?), on assumptions about agency and responsibility, and on a commitment to valuing human life. None of those dependencies disappears when we refuse to state them; they just operate in the dark.

Masking the dependencies is what makes absolutist reasoning fragile and absolutist disputes interminable. What one side treats as a self-evident absolute, the other side exposes as a contested assumption — and because the absolutist’s framework has no vocabulary for its own conditions, the dispute can only be repeated, never resolved. Absolutism is not strength. It is blindness.

What Conditionalism States

Conditionalism is the thesis that every truth evaluation depends on conditions, implicit or explicit. This is not the claim that every meaningful sentence must be rewritten as a material conditional, if X then Y. A material conditional can be vacuously true when its antecedent is false, which is not the point. The truth-bearer is better represented as an interpreted proposition together with an evaluation context:

\[ \langle P, C \rangle, \]

where \(P\) is the proposition and \(C\) fixes the material parameters—language, referents, domain, time, model, and standards—needed to evaluate it. Truth is validity of \(P\) under \(C\).

The argument for this has three steps:

1. Every truth claim requires interpretation. A string of words does not evaluate itself; someone or something must fix what it means before asking whether it holds.
2. Every interpretation requires background conditions. Interpretation leans on linguistic conventions, conceptual frameworks, axiomatic systems, observational parameters. There is no view from nowhere to interpret from.
3. Therefore truth attaches to interpreted claims under conditions. The conditions do load-bearing work whether stated or pragmatically supplied. A sentence with no sufficiently fixed interpretation or domain is not yet ready for evaluation.

The conditions usually stay implicit, and ordinary communication requires that compression. “Water boils at 100°C” is serviceable when speakers share the relevant standard conditions. Logical and mathematical claims likewise require a formal language, semantics, and inference rules, though notation and historical origin should not be confused with what makes a theorem follow. Conditionalism’s job is to surface the material conditions when frameworks clash, a claim is pushed beyond its domain, or an utterance remains too underdetermined to evaluate—cases taken up in When Statements Fail.

The position is a close cousin of Quinean holism and of approaches that locate meaning in linguistic practice. Conditionalism sharpens those insights into a claim about evaluation: truth values attach to interpreted propositions under conditions, never to uninterpreted strings. The Three Levels of Truth asks how purpose, correspondence, and coherence contribute without collapsing into a single definition.

Conditionalism Applied to Itself

The thesis must meet its own standard. Conditionalism is asserted under conditions: ordinary inferential and semantic practices, enough stability to identify claims across contexts, and a realist constraint that evidence can make some interpretations perform better than others. Within those conditions, it says that truth evaluation requires an interpreted proposition and a domain. It does not pronounce, from outside every framework, that all conceivable logics or languages must use this vocabulary.

If a rival supplies a genuinely condition-free truth-bearer, or shows that evaluation can proceed without interpretation or domain, Conditionalism should be revised. Its standing is that of a criticizable argument inside the practices it recommends, not an exception to them.

Stronger, Not Weaker

The immediate objection is that conditional truth sounds like diluted truth — that adding “given these assumptions” to a moral claim turns conviction into equivocation. The opposite is the case.

When I say “slavery is wrong,” Conditionalism asks which evaluative and factual premises support the verdict: the standing of enslaved people as agents, the harms imposed, the absence of consent, and the coercive destruction of autonomy. Making those premises explicit makes the claim more inspectable, not automatically stronger. An opponent must contest the values, their scope, or the factual inference rather than trade incompatible verdicts. Given defensible commitments to the agency and standing of enslaved people, the condemnation follows independently of the slaveholder’s approval. Volume 5 undertakes the separate work of defending those commitments and comparing them with moral realism.

Shared Conditions

A society cannot survive without common standards, and the fear that abandoning absolutes means abandoning standards is the deepest motive behind absolutism. The fear is misplaced. What a society needs is not absolute truth but shared conditional frameworks — and these turn out to be what our most robust institutions already are.

Logic, mathematics, empirical science, human rights: none of these is unconditional. Each rests on adopted conditions — inference rules, axioms, methodological commitments, a valuation of persons. Their robustness comes not from being condition-free but from their conditions being so widely adopted and so practically indispensable that opting out carries enormous cost. Within our shared vantage they function as if absolute — we do not renegotiate arithmetic at every transaction — while at the meta-level we remain aware that they are conditional. That dual stance is not a compromise; it is the accurate description of how functioning standards work.

How Far the Conditions Reach

Conditionalism is not only a diagnosis of a philosophical dispute; it is a working method, and the rest of this volume runs on it.

In science, it locates exactly what an experiment tests: never a hypothesis in isolation, but the hypothesis given instruments, auxiliary assumptions, and idealizations — which is why scientific knowledge is permanently revisable without being permanently suspect. In probabilistic reasoning it is even more direct: Bayesian updating is conditional reasoning made quantitative, every credence a credence given evidence and background — the framework I defend against its critics in defense of Bayes. And in the Quantum Branching Universe (QBU) — the Everettian picture introduced in Measure and Credence — conditionality reaches into physics itself: claims about events and outcomes hold truth values only relative to a branch, a theme developed in probability without collapse. The pattern repeats at every level. Wherever truth is being evaluated, conditions are doing the work, and progress comes from dragging them into the light.

The Climb

Conditionalism keeps everything worth keeping from the traditional reverence for truth, including its humility. We are biased, error-prone creatures, and we must climb toward better approximations. But the climb is not toward an impossible Absolute; it is the refinement and expansion of our conditional frameworks — stating conditions more precisely, replacing narrow ones with broader ones, discarding frameworks whose conditions cannot survive being made explicit. The climb is real. The progress is measurable. The grounding is honest.

Conditionalism rejects global relativism’s lack of shared correction and absolutism’s suppression of material assumptions. The alternative is not a midpoint but a method: identify the proposition, bind the conditions that control its evaluation, and expose both to criticism. Conditional truth is not the only vocabulary in which that method can be expressed; it is this book’s proposed account of why the method works.