Types of Explanation

To explain something is to make it plain, and there are various ways of doing this. Brown and Atkins set out their analysis, developed within an educational context, and define explaining as 'an attempt to give understanding of a problem to others':(1)

Our typology consists of three main types of explanation. these may be labelled the Interpretive, the Descriptive and the Reason-Giving. They approximate to the questions, What?, How?, and Why? Interpretive explanations interpret or clarify an issue or specify the central meaning of a term or statement? Descriptive explanations describe processes, structure and procedures? Reason-giving explanations involve giving reasons based on principles or generalisations, motives, obligations or values.(2)

Brown and Wragg have developed these ideas in a useful school text entitled 'Explaining'.(3)

I have illustrated some of the different types of explanation in my Guide to Science and Belief by explaining Harrison's prize-winning chronometer in the following ways:

? Interpretive explanation (answering the question WHAT is it?): - an intrument for telling the time very accurately, a special kind of clock.

? Descriptive explanation (answering the question HOW is it constructed?): - from cog wheels, springs, pivots, hands and a balance wheel.

? Reason-giving (scientific - 'principles or generalisations') explanation. (answering the question WHY does it keep time so accurately?): - the expansion of a lever with changing temperature causes a compensating change in the length of the hairspring.

? Reason-giving (motives) explanation (answering the question WHY was it invented?): - to determine longtitude accurately at sea in order to save sailors' lives and time. [pp 40f]

? Reason-giving (motives) explanation (answering the question WHY was it invented?) - a third reason-giving explanation, again a 'motives' type, was the attraction of the very large prize money offered.

The important point for our present concern is that there are many different types of explanation which serve different functions. Furthermore, as Flew succinctly observes, explanations answering different questions are not necessarily rivals. . . The first moral, therefore, is that there is not just one single, the explanation for anything which we may wish to have explained. There may instead be as many, not necessarily exclusive, alternative explanations as there are legitimate explanation-demanding questions to be asked.(4)

This is an important point for science-and-religion because there is a tendency, in some popular literature on science, to treat scientific explanations as the only ones or at least the best ones. This is particularly noticeable in some popular writings on cosmology where the inference is that if a complete scientific explanation can be found, God is somehow ruled out of the picture, as for instance is suggested in these comments by John Gribbin:

The known laws of physics, operating in accordance with the observed behaviour of the Universe on the largest and smallest scales, are alone sufficient to explain how everything came into being, spontaneously, at a definite moment of creation about 15 thousand million years ago. The ultimate question of metaphysics has indeed been answered. Is there, then, still a role for God?(5)

But explanations of the world in terms of God's agency and purpose are logically compatible with scientific explanations of its origins in a Big Bang or whatever. The same applies to explanations of the origins of humankind.

The associated idea that scientific explanations are the best explanations contains a partial truth, provided the question is first asked, 'best for what?' If the task is that of developing a new medicine to cure people of a certain disease, then a scientific explanation of a human being would be likely to be the 'best' explanation. If, however, the aim is to woo a partner, it definitely would not.

To give one type of explanation, when one of a different function and type is required, is to commit an explanatory type-error. A whimsical comment by a colleague who lectured in zoology will illustrate the point. A student asked why a dogfish had the particular structure that it did. The lecturer replied 'Because God made it that way - but I don't suppose that was the answer you wanted!' Of course he knew very well that the student wanted a biological explanation and he was committing a type-error in offering an explanation of a different function and type. It was not that the two accounts were incompatible, just that an explanation of the existence of the dogfish in terms of divine agency was an inappropriate answer to the student's question.

In the two sections that follow we will consider two classic confusions of types and levels of explanation. In one (ontological reductionism) some atheists have attempted to replace and deny theological explanation by extending some aspects of scientific explanation beyond their proper level, and in the other (the "god-of-the-gaps theology) some theologians have mistakenly tried to put theological explanation in a situation where scientific explanation was appropriate.

"Explaining Away": Reductionism and Emergence

It is a standard and successful practice in science, one which is particularly favoured by physicists and chemists, to give descriptive explanations (6) of solids, liquids and gases in terms of atoms and molecules. This reduction of matter to its constituent parts facilitates understanding, and promotes reason-giving explanations of such phenomena as the behaviour of solids, gases and liquids, as well as the formation of chemical compounds. All matter can be looked at in this way. But it has sometimes been claimed that the atom-and-molecule account of matter is all that there is to be said. The belief which is expressed in such a claim is loosely termed 'reductionism', but this is a little too imprecise; it fails to distinguish between two very different forms of reductionism:

(i) methodological reductionism
(ii) ontological reductionism.(7)

(i) Methodological reductionism, such as treating the large-scale behaviour of gases as the incessant, elastic collisions of numerous molecules, is just one of the standard procedures, or methods, of science - particularly of the physical sciences - and is without metaphysical(8) significance. It is the legitimate reduction of macro-phenomena to their molecular components, which is a successful technique in certain areas of science.

But even with methodological reductionism, care is needed, for the emergence of new properties can take place as a result of the way the individual constituents fit together. The so-called phenomenon of emergence(9) signals the fact that there is more to be said than appears at the analytic level. The synthetic level also needs to be considered, something which is true within the physical sciences as well as in the life sciences.

For example, an electron and a proton in combination become an atom of hydrogen, and, with a large number of hydrogen molecules, each consisting of two atoms, a new emergent property - 'gaseous' - arises which is not possessed by the electrons and protons which constitute hydrogen. The same is the case for oxygen but with a different combination of constituent particles.

Then if the hydrogen is burnt in the oxygen, water results, having the emergent property of 'wetness', a property which is not possessed by either hydrogen or oxygen. The phenomenon of emergence acts as a caution that significant information may be lost if a system is scrutinized only at the component level.

Biologists, perhaps, are less prone than physicists and chemists to reduce everything to atoms and molecules. Their discipline reminds them constantly of the importance of the organisation of the individual constituents. But biologists also need to recognise other higher-order categories of concepts than biological ones. Pair-bonding is a biological concept; love is not. To attempt to reduce the richness of the human experience of love to 'nothing but' pair-bonding is to leave a lot unsaid.

(ii) Ontological reductionism, which gets its name from ontology, the study of being, of what is, surfaces in assertions like, 'human beings are nothing but complex chemical mechanisms'. Of course, human beings are complex chemical mechanisms, and the assertion is trivially true in the restricted sense that if you take all the atoms and molecules away, there will be nothing left! But the addition of the words 'nothing but' signals the making of a particular claim, namely that the scientific account of a human being is the only valid one. By implication, matter is all that there is and any spiritual dimension is denied. Because of the cumbersome nature of the expression ontological reductionism, it is customary to abbreviate it to 'reductionism'. This practice does not usually generate confusion because methodological reductionism is not a metaphysical issue.

It is the give-away words 'nothing but' - or similar words like 'just', 'only' and 'simply' - which move the statement, 'human beings are nothing but complex chemical mechanisms', out of the scientific universe of discourse into metaphysics. This is because such a statement asserts, not simply that matter can be explored by attending to atoms and molecules, but that the scientific account is all that there is to be said about being. Given that science involves the study of the natural world, if, as the logical positivists claimed, 'Whatever knowledge is attainable, must be attained by scientific methods; and what science cannot discover, mankind cannot know',(10) then the natural world is all that can be known. But that conclusion simply follows deductively from the premisses; its truth or falsity depends on the truth or falsity of the premisses themselves. But as subsequent philosophical analysis has indicated, the second premiss, 'what science cannot discover, mankind cannot know', is inadequate as an expression of the scope of science.

Sir Arthur Eddington once compared science to casting a net into the sea. But a three-centimetre net says nothing about whether there are creatures in the sea which are smaller than three centimetres. Science systematically leaves out anything which does belong to the natural world. 'Nothing-buttery', as ontological reductionism has been dubbed(11) on account of the give-away words 'nothing but', fails to give an adequate account of the totality of human experience.

Theological and Scientific Explanation: 'god-of-the-gaps'

Many years ago certain theologians became apprehensive over the success of science in explaining natural phenomena. Their anxieties could be expressed in some such statement as, 'we thought it was God who kept the planets in their orbits, but now we are told it's gravity'. Failing to recognise that the two accounts are not incompatible, they then looked for a place for God in the gaps in scientific explanations - 'you scientists may have explained X, but you haven't explained Y - that's God!' As an apologetic strategy it is clearly counterproductive since the more that science explains, the less 'room' there is for God! Professor C A Coulson dubbed it the 'God-of-the-gaps', saying

It has always been one of our major temptations to try to divide our experience into two (or more) parts and grant science control of the one part, while allowing religion to maintain its authority in the other. This is a fatal step to take. For it is to assert that you can plant some sort of hedge in the country of the mind to mark the boundary where a transfer of authority takes place. Its error is twofold. First it presupposes a dichotomy of existence which would be tolerable if no scientist were ever a Christian, and no Christian ever a scientist, but which becomes intolerable while there is one single person owning both allegiances. And second it invites "science" to discover new things and thence gradually to take possession of that which "religion" once held. In some respects I believe this to be the most serious and wasteful of all our errors.(12)

The limitations of science are methodological, not territorial.

C A Coulson held the Chair of Theoretical Physics at King's College London. Knowing the theological traditions of the College and Coulson's own Christian commitment, one reporter referred to him as 'Professor of Theological Physics'!

The consequences of believing in a 'God-of-the-gaps' are two-fold. For the Christian it can result in an unnecessary fear of scientific discoveries, lest God gets 'pushed further out of the picture'. It can also foster the belief that certain aspects of the physical world will never yield up their secrets because they belong to God's domain. Roger Pilkington pointed out how, in countless ways, such beliefs have been shown to be wrong:

"They won't split the atom. You can't unmake what God has made" - but to-day transmuted atoms if not actually ten a penny are at least ten for a thousand pounds. "It's not gravity, it's God that keeps the planets in their courses. Man will never be able to conquer space. . ."(13)

For the non-Christian the belief that the Christian God is a 'God-of-the-gaps' may result in an enthusiastic pursuit of science prompted by the motive that God will thereby get pushed out of the picture. As Peter Atkins declares in the first edition of his book The Creation,

My aim is to argue that the universe can come into existence without intervention, and that there is no need to invoke the idea of a Supreme Being.(14)

Atkin's use of the word 'need' is interesting and calls to mind a similar statement by Sir Julian Huxley that

in the evolutionary pattern of thought there is no longer either need or room for the supernatural.(15)

But what does it mean to talk about 'need' or 'room'? Although in Newton's time references were made to God in works of science, it is now a convention that scientific books omit references to First Causes. The claim that there is no 'room' for God seems to spring out of 'God-of-the-gaps' thinking, which sees divine activity as plugging up gaps in our current knowledge. It is an inappropriate view since the biblical picture of God's activity is one of total involvement in the creation, sustaining as well as redeeming the world.(16)

The caution with which the idea of 'need' for God should be treated can be illustrated in the following way. A detective working on a murder enquiry may be preoccupied by reason-giving (motive) explanations. A pathologist on the other hand will be engrossed with reason-giving (scientific) explanations for the death of the victim. In this sense, qua scientist, the pathologist has no 'need' of explanations of purpose and intention. But it would be nonsense to deny the validity of these other types of explanation just because in his professional capacity as a scientist, the pathologist had no need of them. It is in this restricted sense only that it is appropriate for Atkins to say that 'there is no need to invoke the idea of a Supreme Being' to do science. Certainly there is no need to mention God in a scientific account; but it does not follow from this that God doesn't exist.

'God-of-the-gaps' is a classic example of a type-error in explanation, as it substitutes an explanation in terms of God's agency and purpose in places where a scientific one is required. As C A Coulson puts it:

When we come to the scientifically unknown, our correct policy is not to rejoice because we have found God; it is to become better scientists.(17)

Book Reference

I have developed these ideas in my Beliefs and Values in Science Education, Buckingham: Open University Press, ISBN 0-335-15645-2 (pbk)

References