. The process of making something intelligible, or saying why certain things are as they are, or the account used to do these things. The account is sometimes called the explanans, and the thing to be explained is called the explanandum. An explanation may do things like reducing the unfamiliar to the familiar, thus enlightening people. Things it does in this way are called its pragmatic features and they are relevant to the question of whether an explanation should be relative to the receiver. A child, a lay adult, and an expert seem to require different explanations of the same thing, since what is already familiar to them will differ. Yet a scientific explanation may be in terms hitherto unfamiliar even to the scientist, the lay person may study science to find ‘the’ explanation of something, and the familiar itself may need explaining, as Newton explained the falling of an apple by associating it with other, apparently quite unrelated, phenomena. Many writers have therefore regarded the pragmatic features as subjective and incidental, and have concentrated rather on various views of the logical form an explanation should take.
On one view, the covering law model, an explanation should state general laws and initial conditions which together logically entail the explanandum. ‘All water heated at normal pressure to 100°C boils’ and ‘This water was so heated’ together explain why this water boiled. The model can be adapted to explain laws themselves, and may have more complex applications, where covering laws are distinguished from supporting laws. However, not just any covering law will do. ‘Whenever water is about to turn to steam it boils’ and ‘This water was about to turn to steam’ do not explain the boiling. This suggests that an explanation must present what is ‘more knowable absolutely’ even though perhaps not ‘more knowable to us’ (Aristotle), and that pragmatic features are relevant, though they need not be subjective (cf. Dray, p. 74). An explanation therefore might present a cause, or something logically prior to the explanandum in the relevant system, as when the explanandum is a mathematical fact.
Often, however, scientific laws say that not all As are Bs, but that a certain proportion are. When the covering law model is extended to use laws of this type, explanations are called statistical or probabilistic. The previous kind, all As are Bs, are called nomological. Nomological explanations are DEDUCTIVE. The explanandum is deduced from the premises. Statistical explanations are usually (not always: see Hempel) INDUCTIVE. For reasons given under CONFIRMATION statistical explanations must be supplemented by Carnaps’ requirement of total evidence (or Hempel’s weaker form, the requirement of maximal specificity).
It is disputed whether any but deductive explanations are properly called explanations at all, and also how many kinds of explanation there are. In particular can the covering law model apply to subjects like history and psychology, and can teleological explanations, i.e. those in terms of purposes or final causes, be reduced to causal explanations? Also do questions beginning with Why?, How? etc. call for the same type of explanation? Is there a basic difference between explaining why-necessarily and how-possibly (Dray)? Can the covering law model account for explaining what something is, or what people are doing?
Further, how is explaining something related to describing it, and also to predicting it? Darwinism explains the variety of species, but does not seem fitted for predicting new species. Thales allegedly predicted the eclipse of 585 BC by consulting records but he could almost certainly not explain it. Do explanation and predictability nevertheless ultimately go together?
So far we have left the explananda vague. When not themselves laws, are they events, states, processes, situations, actions, forbearances, or are they statements describing these? To explain a statement one needs to explain one or more features of the relevant event, while to explain the event itself in its infinite richness would seem a hopeless task. Explaining meanings raises special problems.
After asking what an explanation is, we can ask when one is adequate, and whether the same explanandum can be given more than one explanation.
Explication, when not simply a synonym for ‘explanation’, is the process whereby a hitherto imprecise notion is given a formal definition, and so made suitable for use in formal work. The definition does not claim to be synonymous with the original notion, since it is avowedly making it more precise. (This is a form of logical analysis: see PHILOSOPHY.)
Aristotle, Posterior Analytics, 7lb33–2a6. (‘More knowable absolutely’ and ‘to us’. See also early chapters for when scientific argument is explanatory.)
R.Carnap, Logical Foundations of Probability, RKP 1950, Chicago UP, 1962, chapter 1. (Explication. Cf. W.V.Quine, Word and Object, Wiley, 1960, §§ 53–4.)
W.H.Dray, Laws and Explanation in History, 1957. (Opposes covering law model for history.)
C.G.Hempel, Aspects of Scientific Explanation, Free Press, 1965. (Chapter 12 has full account of covering law model, and defends its adequacy in all fields. Chapter 9 (also in Journal of Philosophy, 1942) defends it regarding history. Cf. also his ‘Explanation in science and in history’, in R.G.Colodny (ed.), Frontiers of Science and Philosophy, 1962, reprinted in P.H.Nidditch (ed.), The Philosophy of Science, Oxford UP, 1968, and in W.H.Dray (ed.), Philosophical Analysis and History, Harper and Row, 1966, Greenwood Press, 1979.
*J.Hospers, ‘What is explanation?’, Journal of Philosophy, 1946, reprinted in A.Flew (ed.), Essays in Conceptual Analysis, Macmillan, 1956, St Martin’s Press, 1966, Greenwood Press, 1981. (Elementary defence of covering law model as adequate for explanations-why of events.)
P.Kitcher and W.C.Salmon (eds), Scientific Explanation, Minnesota UP, 1989. (Specially written papers, including long historical survey, with chronological bibliography. Cf. also Salmon’s Scientific Explanation and the Causal Structure of the World, Princeton UP, 1984, defending appeal to causation rather than to deductive inference or necessitation.)
D.-H.Ruben, Scientific Explanation, Routledge, 1990. (Partly historical and partly constructive. Elaborates realist view, whereby explanation is not just relative to human knowledge or interests.)
C.Taylor, The Explanation of Behaviour, RKP, 1964. (Distinguishes and defends need for, teleological and purposive explanations. T. L.S.Sprigge, ‘Final causes’, Proceedings of the Aristotelian Society, supplementary vol., 1971, assimilates teleological to causal explanations. Cf. also L.Wright, Teleological Explanations: an Etiological Analysis of Goals and Functions, California, 1976, which defends them.)
G.H.von Wright, Explanation and Understanding, RKP, 1971. (Causal and teleological explanations regarding human actions, etc. Good bibliography.)
This is the complete article, containing 1,057 words
(approx. 4 pages at 300 words per page).
