covering-law model and deductive nomological explanation

covering-law model and deductive nomological explanation

a model of scientific EXPLANATION particularly associated with Carl Hempel and Karl POPPER. In this model the defining feature of a scientific explanation is represented as resting on the operation of general SCIENTIFIC LAWS and initial condition statements (together known as the EXPLANANS) which logically entail the phenomenon thus explained (the EXPLANADUM). The basic model is as follows:
  1. Law(s).
  2. Statements of initial conditions, which point to the applicability of the chosen laws to the case in hand. the explanans
  3. The phenomenon explained (or predicted) as a deduction from the above. the explanandum

The laws involved are universal conditional statements: (‘For any a , if a, then b ), for example, ‘All water heated at the pressure existing at sea level boils at 100° C’. Whilst the basic model thus identifies SCIENCE as involving deterministic laws, ‘probabilistic laws’ (asserting ‘if a , then a certain probability of b’) can also be accommodated. An important distinction also made on the basis of the model is between scientific and nonscientific prediction: rather than being unconditional, scientific PREDICTION is conditional on the occurrence of the relevant initial conditions. The model is proposed as providing a unified account of the role of explanation, prediction and test in science, including a suggested ‘logical symmetry’ of explanation and prediction (see also FALSIFICATIONISM).

Criticism of the covering-law model has concentrated on its claims to provide a defining model of scientific explanation. There is no general acceptance that subsumption under empirical covering laws is a sufficient or a necessary condition for scientific explanation. Successful covering laws can be formulated without these being the basis of satisfactory explanation. A suggestion also made is that scientific explanation involves the identification of underlying causal ‘mechanisms’ which may not depend on empirical covering laws or involve prediction (see EXPLANATORY MECHANISM, REALISM); for example, while Darwinian theory provides an important explanatory account, this is presented in terms of general mechanisms and is not predictive. In sociology in particular it is also clear that important kinds of explanation occur which are dependent neither on covering law nor on general mechanisms (for example, MEANINGFUL UNDERSTANDING AND EXPLANATION). Thus, while the covering-law model may be seen as adequately portraying one form of scientific explanation, there is no widespread acceptance that it provides an adequate overall model of this, or of the range of explanation in general. See also HYPOTHETICODEDUCTIVE EXPLANATION AND METHOD, EXPLANATION.

Collins Dictionary of Sociology, 3rd ed. © HarperCollins Publishers 2000