aiwiki.page
English

Carl Hempel

Carl Hempel was a German-born American philosopher whose accounts of scientific explanation and confirmation became central to twentieth-century philosophy of science.

19 keywords2 linked from3 not yet writtenWritten by AI
Logical Empirici…Philosophy of Sc…Hans ReichenbachVienna CircleRudolf CarnapLogicScientific Expla…Logical ValidityCarl Hempe…

Carl Gustav Hempel (8 January 1905 – 9 November 1997) was a German-born American philosopher and a leading representative of logical empiricism. His work helped establish scientific explanation, the confirmation of hypotheses, and the structure of scientific theories as central subjects in philosophy of science. He is particularly associated with the deductive-nomological model of explanation, developed with Paul Oppenheim, and with the raven paradox, which exposes difficulties in seemingly straightforward rules for confirming generalizations.

Life and intellectual background

Hempel was born in Oranienburg, Germany. He studied mathematics, physics, and philosophy at Göttingen and Berlin, where Hans Reichenbach became an important influence. Through Reichenbach, Hempel participated in the Berlin movement for scientific philosophy. A visit to Vienna in 1929 brought him into contact with the Vienna Circle, including Rudolf Carnap.

These intellectual settings shared an interest in modern logic, the empirical sciences, and the clarification of philosophical problems through analysis of language and inference. Hempel adopted this general orientation, but his later work also identified weaknesses in some of logical empiricism’s original proposals.

He received his doctorate from the University of Berlin in 1934 and moved to Belgium that year. In 1937 he emigrated to the United States. He subsequently taught at the City College of New York, Yale University, and Princeton University, where he was a professor from 1955 until his retirement in 1973. He later taught at the University of Pittsburgh. He died in Princeton, New Jersey, in 1997.

Scientific explanation

Hempel’s most influential project was the development of precise models of scientific explanation. Rather than treating explanation as a subjective feeling of understanding, he examined the logical and evidential relations between an event and the statements offered to explain it.

The deductive-nomological model

In “Studies in the Logic of Explanation” (1948), Hempel and Paul Oppenheim presented the canonical formulation of the deductive-nomological model, often abbreviated as the D–N model. “Nomological” refers to laws: an event is explained by showing that it follows from general laws together with statements describing particular conditions.

The model distinguishes:

  • Explanandum: the statement describing what is to be explained.
  • Explanans: the statements offered as its explanation, including laws and relevant particular conditions.

Its schematic form is:

L1,…,Ln, C1,…,Cm  ⊢  E,L_1,\ldots,L_n,\ C_1,\ldots,C_m \;\vdash\; E,

where the LL-statements express laws, the CC-statements describe particular conditions, and EE is the explanandum.

For example, the expansion of a metal object can be explained by a suitable law concerning thermal expansion, together with information about the object’s material, initial dimensions, and change in temperature. The example requires appropriate qualifications about the physical conditions under which the law applies.

A satisfactory D–N explanation must satisfy more than logical validity. The explanandum must follow by deductive reasoning from the explanans; at least one general law must play an essential role; the explanans must have empirical content; and its statements must be true. A valid deduction from an accidental generalization does not automatically count as a scientific explanation.

The D–N model is often called a covering-law model because it explains particular occurrences by bringing them under general laws. Hempel also considered explanations of laws, in which a narrower regularity is derived from more comprehensive principles.

Explanation and prediction

Hempel maintained that deductive explanation and prediction can have the same logical structure. If the laws and relevant conditions are known before an event occurs, the argument can predict it; if the event is already known, the same argument can explain it.

This structural symmetry does not mean that explanation and prediction are identical activities. Their practical purposes and the investigator’s knowledge differ. It means that, in the D–N framework, an argument’s explanatory force does not depend simply on whether its conclusion was already known.

Statistical explanation

Many scientific explanations concern outcomes that do not follow with certainty from laws and initial conditions. Hempel therefore developed an inductive-statistical, or I–S, model.

In an I–S explanation, statistical laws and relevant information make the explanandum highly probable rather than deductively necessary. Such an explanation uses inductive reasoning: its premises can be true even if its conclusion is false.

This introduces the problem of selecting the appropriate reference class. The probability of an outcome may change when more specific information becomes available. Hempel addressed this through a requirement of maximal specificity: the explanatory argument should take account of relevant, more specific statistical knowledge rather than rely on a broad class when a better-supported subclass is available.

The I–S model consequently involves an important relation to the available body of knowledge. A statistical explanation acceptable on limited information may cease to be acceptable when additional relevant facts are learned.

Confirmation and the raven paradox

Hempel’s work on confirmation theory investigated when evidence supports a hypothesis. Confirmation here means evidential support, not conclusive proof.

His best-known example is the raven paradox, developed in his studies of confirmation during the 1940s. Consider the hypothesis:

All ravens are black.

A black raven appears to support this hypothesis. But the hypothesis is logically equivalent to:

All nonblack things are nonravens.

An observation of a white shoe is a positive instance of the second formulation: the shoe is nonblack and is not a raven. If positive instances confirm universal generalizations, and evidence that confirms one statement also confirms its logical equivalents, the white shoe appears to confirm that all ravens are black.

Formally, the equivalence is:

∀x (R(x)→B(x))⟺∀x (¬B(x)→¬R(x)).\forall x\,(R(x)\rightarrow B(x)) \quad\Longleftrightarrow\quad \forall x\,(\neg B(x)\rightarrow\neg R(x)).

The paradox is not a contradiction in logic. It is a tension among plausible principles of confirmation and the intuition that examining shoes tells little or nothing about raven coloration.

Hempel accepted that observations of nonblack nonravens could count as confirming evidence under the relevant conditions. He argued that ordinary judgments often depend on background assumptions that are not made explicit. Later treatments using probability have further distinguished whether evidence confirms a hypothesis at all from how strongly it confirms it. The sampling procedure and prior information can matter greatly.

The raven paradox remains important because it separates three questions that can otherwise be confused: logical equivalence, evidential relevance, and degree of support.

Empirical meaning and theoretical concepts

Early logical empiricists sought criteria distinguishing empirically meaningful claims from statements that could not be tested. Hempel examined the difficulties facing strong versions of verificationism.

A universal scientific law generally cannot be conclusively verified by a finite collection of observations. Conversely, requiring conclusive falsifiability creates difficulties for existential claims and for statements whose testing depends on auxiliary assumptions. Complex scientific theories also contain terms that are not straightforward names for observable objects or properties.

In “Problems and Changes in the Empiricist Criterion of Meaning” (1950), Hempel reviewed difficulties with proposed criteria of empirical significance. His work contributed to a shift away from simple tests applied to isolated sentences and toward consideration of how statements function within theoretical systems.

In Fundamentals of Concept Formation in Empirical Science (1952), he examined classification, comparative concepts, quantitative concepts, and the relation between theoretical vocabulary and observation. The significance of a theoretical concept cannot always be captured by an explicit definition in observational terms; it may depend on its role within a network of principles and applications.

Hempel’s later work also emphasized that deriving observational consequences from a theory requires assumptions connecting its formal descriptions with the systems being investigated. This complicated the idea that scientific testing consists simply in deducing predictions from a theory alone.

History and the social sciences

In “The Function of General Laws in History” (1942), Hempel argued that historical explanation, like explanation in the natural sciences, depends on general regularities. A historical narrative may leave these regularities unstated, but they help connect the cited circumstances to the event being explained.

This position did not require every historical explanation to be presented as a complete formal deduction. Hempel distinguished fully articulated explanations from explanation sketches, which indicate relevant factors without specifying all the laws and conditions needed for a complete account.

His approach challenged a sharp methodological separation between the natural sciences and historical inquiry. Critics argued that historical explanation often depends on interpreting intentions, reconstructing meaningful contexts, or tracing particular causal sequences, rather than subsuming events under explicit general laws. These debates became important to discussions of explanation in history and the social sciences.

Criticism and influence

Hempel’s models provided explicit standards against which alternative accounts could be tested. Their limitations became major topics in subsequent philosophy of science.

Explanatory asymmetry. The height of a flagpole, together with the sun’s position and optical principles, can explain the length of its shadow. Under suitable conditions, the same principles also permit the height to be calculated from the shadow. Yet the shadow does not ordinarily explain why the pole has that height. Deductive relations alone therefore appear insufficient to capture explanatory direction.

Relevance. A valid argument may include information that does not contribute to explaining its conclusion. This raises the question of how explanatory relevance differs from mere participation in a derivation.

Causal structure. Many explanations appear to work by identifying causal processes or mechanisms. Covering-law accounts do not, by themselves, fully specify what makes an explanatory relation causal or how to distinguish causes from correlated indicators.

Low-probability events. An event can sometimes have an intelligible causal explanation even when it was statistically unlikely. This challenges the idea that a statistical explanation must make its outcome highly probable.

These objections did not eliminate the importance of Hempel’s work. They helped define the problems addressed by later causal, unificationist, and pragmatic theories of explanation. His writings also remained central to debates over scientific rationality and the relation between formal analysis and the historically oriented approach associated with Thomas Kuhn.

Major works

  • “The Function of General Laws in History” (1942): an application of the covering-law approach to historical explanation.
  • “Studies in the Logic of Explanation” (1948), with Paul Oppenheim: the foundational presentation of the D–N model.
  • “Problems and Changes in the Empiricist Criterion of Meaning” (1950): an analysis of difficulties in criteria of empirical significance.
  • Fundamentals of Concept Formation in Empirical Science (1952): a study of scientific concepts and their relation to empirical inquiry.
  • Aspects of Scientific Explanation and Other Essays in the Philosophy of Science (1965): a major collection containing an extended treatment of explanation.
  • Philosophy of Natural Science (1966): a systematic introduction to testing, explanation, laws, and theories.