Carl Hempel

Carl Gustav Hempel (1905–1997) was a German-American philosopher of science whose work defined the analytic tradition's approach to scientific explanation, confirmation, and theory structure. A student of the logical positivists Rudolf Carnap and Hans Reichenbach, Hempel transformed the epistemological concerns of the Vienna Circle into precise logical models — most famously, the Deductive-Nomological (D-N) model of scientific explanation and the Raven Paradox, which exposed deep tensions in probabilistic confirmation theory.

Hempel's career traces the arc of twentieth-century philosophy of science: from the confident formalism of logical positivism, through the recognition that no purely syntactic model captures scientific practice, to a mature position that acknowledged the irreducible role of pragmatic and contextual factors in theory appraisal. His intellectual biography is, in miniature, the biography of a discipline learning its own limits.

The Deductive-Nomological Model

In his 1948 paper with Paul Oppenheim, Hempel proposed that a scientific explanation is a deductive argument: the explanandum (the phenomenon to be explained) follows logically from a set of premises consisting of universal laws and initial conditions. A prediction has exactly the same logical structure — it differs from an explanation only in that the explanandum is known before the argument is constructed, while the prediction concerns a future or unobserved event.

This symmetry thesis — that explanation and prediction are structurally identical — was radical and influential. It implied that explaining why the sun rose yesterday and predicting that it will rise tomorrow are the same kind of cognitive achievement: both demonstrate that the outcome was necessitated by law-governed antecedent conditions. The model gave philosophers of science a clear target: any putative explanation that could not be reconstructed as a valid deductive argument from laws was not a genuine explanation.

The D-N model faced decisive objections. Explanatory asymmetry: we can deduce the height of a flagpole from the length of its shadow and the angle of the sun, but we do not explain the flagpole's height by citing its shadow. The problem of irrelevance: adding irrelevant premises that happen to be true laws does not improve an explanation, yet the D-N model cannot exclude them. The problem of low-probability events: a deductive model cannot explain why a particular atom decayed at a particular time if the governing law is probabilistic. These problems pushed Hempel to develop the Inductive-Statistical (I-S) model, but the I-S model suffered from analogous difficulties and never achieved the clarity of its deductive predecessor.

The Raven Paradox and Confirmation Theory

Hempel's 1945 Raven Paradox (or 'Paradox of the Black Ravens') exposed a tension in naive inductivism. The hypothesis 'All ravens are black' is logically equivalent to 'All non-black things are non-ravens.' A white shoe is a non-black non-raven, so it confirms the second formulation. By logical equivalence, it should confirm the first. But a white shoe seems irrelevant to ornithology.

The paradox is not a trick. It reveals that confirmation depends on background knowledge and relevance relations that purely syntactic logic cannot capture. A white shoe confirms 'All ravens are black' only if you already know that ravens are a small subset of non-black things — a piece of background information that the formalism does not include. Hempel himself accepted the 'equivalence condition' and argued that the intuition of paradox rests on a confusion: we judge the white shoe irrelevant because we tacitly restrict our inquiry to birds, but the unrestricted logical formulation makes no such restriction.

The deeper lesson, which took decades to absorb: confirmation is not a two-place relation between evidence and hypothesis. It is a three-place relation between evidence, hypothesis, and auxiliary assumptions — and the auxiliary assumptions are doing almost all the epistemic work. Hempel's paradox was one of the main routes by which philosophy of science discovered that pure formalism is insufficient.

From Formalism to Pragmatics

Hempel's later work — particularly 'Aspects of Scientific Explanation' (1965) — acknowledged what the criticisms had made unavoidable. Explanation requires not merely deductive entailment but causal relevance, pragmatic appropriateness, and contrastive focus. Why did the window break? Because a stone hit it. The D-N model can derive the breaking from the stone's momentum and the glass's fragility. But what we want to know, pragmatically, is why the stone's impact rather than the breeze's pressure caused the break — a question about causal selection that the formal model does not address.

Hempel never abandoned the aspiration to formal precision. But he came to see that formal precision captures only the skeleton of scientific reasoning. The flesh — the decisions about which factors to include, which to idealize away, which contrasts to foreground — is irreducibly contextual. This is the trajectory that leads from Hempel to contemporary mechanistic explanation, to Philip Kitcher's unificationist account, and to the recognition that scientific understanding is as much a social achievement as a logical one.

Hempel's legacy is not a solved problem but a precisely articulated failure: the demonstration that no purely syntactic model of explanation or confirmation can succeed, and that the attempt to build one produces the conceptual machinery by which we now understand why it cannot.

The Raven Paradox is not a curiosity to be dissolved. It is a map of the boundary where formal logic ends and scientific judgment begins — and Hempel was the cartographer who drew the line so clearly that we have been navigating by it ever since.