2 Methodological Approach

Any critique of the theoretical structure of the work of a social scientist must itself be based upon a ‘philosophical’ interpretation of the nature of scientific theories in general, and of social science theories in particular.

This section will constitute a relatively brief exposition of a particular methodological perspective rather than an argument for its validity. Here it will be sufficient to show that the methodology chosen is coherent and consistent in its internal structure. Given the nature of philosophical concepts, this is indeed the only a priori test that can be applied. However, the subsequent application of the approach to Smith’s writing will, to the extent that it yields an adequate analysis of his work and resolves some of the apparent contradictions within it, provide some sort of empirical test of the position chosen. It will subse­quently be argued that the model to be developed here, although based on contemporary perceptions of science, is in its formal structure very close to Smith’s model of science. It is indeed this possibility of reflexive application that enables the model to generate a set of organising principles which suggest how the various aspects of his work might be related.

The methodological approach taken here is based upon the work of the late Imre Lakatos, although his schematisation will have to be extensively supplemented in order to deal with the complexities of Smith’s ‘dialectical’ approach. Lakatos was responsible for developing Karl Popper’s ‘falsificationist’ position¹ into what he called ‘sophisticated methodological falsification- ism’.² His purpose in doing so was to strengthen Popper’s view of science against the ‘irrationalist’ attacks of Thomas Kuhn and his followers.

Kuhn on the other hand has a socio-psychological view of science. Scientists carry out ‘normal science’ within a ‘paradigm’ which is secured as far as possible against refutation. Inconsis­tencies abound, but are ignored until such time as a ‘crisis’ occurs, during which the old paradigm is swept away, and replaced by a new one. There is no rational cause for such a crisis to appear at any particular time; it is a psychological phenomenon, akin to ‘mass panic’. The transition to a new paradigm is a matter of quasi ‘religious conversion’, and since scientific paradigms are incommensurable, there can be no metaparadigmatic standards by which to judge the adequacy of paradigms, and thus no logic of scientific discovery, but only a psychology thereof.⁴ Whilst Kuhn has consistently denied that such a model does in fact involve an irrationalist perspective, it is difficult to see that he has made out a convincing case for this claim. In particular he is unable to provide any ex-ante criteria which can be used to identify a ‘rational’ change to a new scientific paradigm. In the absence of such criteria the adequacy of a paradigm can only be a function of the confidence which its adherents place in it, and such a test falls well short of the objectivity sought by Popper.

Popper’s view has been (and remains) very influential amongst practitioners of science, partly one might suspect because it is rather more pleasant to consider oneself as engaged in a rational activity, rather than as being a member of an irrational, and potentially panic-stricken, mob.

There can, however, be little doubt that the actual development of science has been much closer to Kuhn’s model than to Popper’s. Scientists in general have in the past clung to theories in the face of considerable ‘refutation’, and ‘crucial experiments’ (such as the famous Michelson-Moreley experiment with light) are seen as such only long after they took place, when a new paradigm has been established. Further, there are, as Lakatos has pointed out, logical objections to Popper’s position, in that there is no clear-cut distinction between ‘theoretical’ propositions, and ‘observational’ propositions — i.e. all scientific observation is ‘theory laden’ rather than ‘pure’ or ‘basic’.

As a standard example of this we can take an observation made in a ‘Wilson Cloud Chamber’ where the condensation traces are ‘seen’ as atoms, in terms of pre-existing theory. This alone makes the refutation of a theory contingent upon the acceptance of other theories, which can themselves be modified to avoid an inconsistency.

Sophisticated methodological falsificationalism seeks to avoid these serious difficulties by postulating the following model:

To be classed as scientific, a theory must have more empirical content than its predecessor, in that it generates additional empirical propositions, as well as including all the unrefuted content of the previous theory. Further, some of this new empirical content must be confirmed by observation, but it is important to recognise that such observation is interpreted within the context of the theory. A theory is only falsified upon this view, if another theory exists which satisfies the above criteria. To save a theory by adding new auxiliary hypotheses is good scientific practice only if new corroborated content is thereby introduced, in which case we speak of a ‘progressive problemshift’.

As an illustration, we can say that the adding of ever more ‘epicycles’ to the Ptolemaic system of astronomical explanation, in order to explain deviations from the model’s predictions, was a degenerative problemshift, whilst if those additions had in fact also generated new (corroborated) predictions about the move­ment of the planets (rather than simply explaining already observed anomalies) the problemshift would have been progressive.

It is this notion of progressive problemshifts that gives a dynamic element to Lakatos’s model. As he says: ‘The idea of growth and the concept of empirical character are soldered into one.’⁵

Clearly, here, no single observation can lead to the falsification of a research programme—only a prolonged struggle with another research programme, which turns out to be better (i.e. more pro­gressive) will lead to its being abandoned.

The central core of a research programme consists of a formal calculus — as, for example, a system of mathematical equations — and a series of epistemological assumptions, implicit or explicit, about the nature of the events with which the theory is concerned. We can say that the epistemology specifies the sorts of things or events that are dealt with by the theory, whilst the calculus specifies the relationships between them.⁶ This core is accepted as irrefutable by those who work with the theory — i.e. it is regarded as unproblematical background knowledge — and is defended when necessary by auxiliary hypotheses in the face of inconsistencies. Lakatos speaks of the ‘negative heuristic’ which is, in effect, an implicit specification of those elements within the core which must not be subjected to the process of refutation. He has been interpreted here as being quite close to Kuhn, in that his model is more tolerant of inconsistency than that of Popper, who takes a sterner line concerning the permissibility of auxiliary hypotheses.

The central core also generates a definition of the sorts of problems to be solved in the peripheral (refutable) area of the research programme, and this he calls the ‘positive heuristic’. As a concrete example of such a core, we could take Newton’s three laws of dynamics, together with his principle of gravitation.

It will be clear from the above discussion that such theoretical cores can fairly be described as ‘metaphysical’, in that they themselves are pre-suppositions with no empirical content. However, given that the word ‘metaphysical’ has distinctly pejorative connotations, it will be better to avoid it and call the protected core ‘metascientific’. Whilst it can be argued that all scientific theories are based upon systems of metaphysical assumptions of the sort outlined above, it is not the case that all such systems serve a function in science; i.e. in the absence of the empirical corroboration implicit in the notion of a progressive problemshift, they could be held to be purely metaphysical, rather than metascientific.

Lakatos’s perspective provides a highly satisfactory explanation of the fact that many refuted scientific theories, when seen from the standpoint of their (more progressive) successors, appear metaphysical, as do for example Cartesian Dynamics, Corpuscu­lar Theory, Phlogiston Theory and (post Einstein) Newtonian Dynamics, whilst the metaphysical (or metascientific) nature of current theories is usually hidden from their adherents. In this connection Ernst Mach⁷ demonstrated the metaphysical nature of many previous theories, but, far from concluding that this was a feature of all science, he went on to urge that all such metaphysics should be purged. Schumpeter also adopted a similar position in his monumental Histoιy of Economic Analysis⁶ where he suggests that an original pre-analytic ‘vision’ is, and should be, purged of metaphysics, by successive refinements of the analytical tech­niques it generates.

Lakatos’s view of scientific theories is, in a sense, a synthesis of Kuhn’s view with that of Popper, in that it emphasises the role of conventional methodological decisions and also makes the refutation of a theory a much more protracted, and less clear-cut process. The synthesis is not unproblematical, in that it is not fully clear how one distinguishes between rational and irrational factors in the ultimate overthrow of a research programme, so that the model seems rather closer to Kuhn’s position than Lakatos appears to have believed it was. To be fully effective as a defence against Kuhn’s relativism it has to be possible to make an ex ante specification of the criteria applied during the progressive/ degenerative test, and it is not at all clear how Lakatos can do this. However, his model does at least leave the question of rationality open, and also has greater generality than either Kuhn’s or Popper’s, both of which can be subsumed under it as special cases. It also adds some, much needed, ‘fine-structure’ to Kuhn’s model and provides the possibility of a non-circular definition of‘paradigm’, and ‘normal science’, thus removing one of the central weaknesses in Kuhn’s approach.

His schematisation is in fact a subtle blend of three major lines of philosophical thought, in that it incorporates elements of the Kantian ‘activist’ conception ofknowledge, ‘conventionalism’ and ‘empiricism’. As well as being logically consistent, it has great explanatory power when applied to the historical, and to the contemporary, practice of science.⁹ Even more importantly, for the present purpose, it is sufficiently general to serve as a vehicle for an exposition of what will be termed a ‘dialectical’ model, and to permit contrasts to be drawn between such a model and the more conventional ‘empiricist’ models of science. At the same time, it is sufficiently restrictive to exclude most of what is generally classified as ‘un-scientific’ activity.

Lakatos tended to confine the application of his model to, and drew his examples from, the physical sciences,¹⁰ but later writers have extended his perspective to include the social sciences¹¹ and in some cases specifically to economics.¹² However, at the time of writing I know of no detailed application of Lakatos’s schemati- sation to a particular economic thinker, although there is no reason in principle why such an application should not be made. Indeed it will be argued below that Smith’s work has a structure which makes it an example par excellence of a Lakatosian research programme.

It is encouraging in this connection to find that no less an authority than the late Ronald Meek has, in one of his later works, gone some way towards this position, although working with Kuhn’s schematisation of ‘paradigm’. He suggests that Smith’s political economy

... possessed most of the basic characteristics of a paradigm as Professor Kuhn defines them. A ‘group of adherents’ was certainly attracted to its system from competing modes of scientific activity: its two main rivals — the system of the Physiocrats, and that of Steuart’s Inquiry — did not long survive its publication...'³

and that the system was ‘enduring’ and ‘open ended’ in that it left problems to be solved by later adherents.

Meek, however, shares Kuhn’s doubts about the possibility of using ‘paradigm’, in its strict sense, in the social sciences, and considers that here a ‘community of practitioners’ is not as readily identifiable as it usually is in the physical sciences,¹⁴ but he says

... even if Smithian political economy can’t be described as a paradigm in the strict sense of the word, it surely represented one of the nearest approaches to it in the history of economic thought. (Ibid.)

We shall return shortly to the model so far developed, but first need to establish a definition of what it means for a social-science model to be dialectical. Here a definition is sought that is sufficiently general to include the dialectics, both of Marx and (as we shall see) of Smith, as special cases, but also sufficiently restrictive to exclude not only ‘non-scientific’ dialectical models (such as that of Hegel) but also empiricist models.

It should be recognised at the outset that when we discuss the nature of dialectical (or for that matter, empiricist) models of science, we inevitably raise questions that may well be incapable of being given any definitive answer, for at bottom, we are here concerned with the link between (scientific) thought and reality. It is not intended to pursue these deep, and possibly irresolvable, questions into the realms of metaphysics; it is sufficient for the present purpose to speak of the way in which social science models appear to be constructed (or at least of the way in which they can be formally reconstructed) without enquiring as to the precise manner in which those models are (or are not) adequate reflections of reality. It should also be made clear that this book is concerned only with scientific knowledge, which should not be identified with knowledge in general. The commonly held view, popularised by Huxley that science is ‘nothing but trained and organised common sense’, neglects the role of theory in the sciences and so blurs a crucial distinction. Even if we take the view that ‘common sense’ observation involves interpretation in terms of a complex network of preconceptions, it is clear that an additional, and perhaps different, process of interpretation is involved when observations are made in the context of a scientific theory.

With these qualifications in mind, the claim made here is that a distinction can be drawn between dialectical and empiricist approaches to social science, and that they can be shown to be addressed to fundamentally different sorts of problems. It may be of course that neither approach will often be found in its pure form, and that in practice it is the degree of emphasis that is relevant, but this does not invalidate the argument.

To establish this position, the first distinction that must be made is between what Nicholas Georgescu-Roegen¹⁵ has called an ‘arithmomorphic’ concept and a concept that is designed to reflect qualitative phenomena. A clear example of arithmomorphic conceptualisation is the set of real numbers, each of which is discretely distinct from any other. Qualitative concepts are, for reasons that will become clear, more difficult to specify, but at this stage the notion can be illustrated with Georgescu-Roegen’s examples of‘democracy’, ‘good’, ‘justice’, ‘likelihood’, and ‘want’. These he says

... have no arithmomorphic boundaries; instead they are surrounded by a penumbra within which they overlap with their opposites. (Ibid., p. 145.)

Though they are

... not discretely distinct, dialectical concepts are nevertheless distinct. The difference is this. A penumbra separates a dialectical concept from its opposite. In the case of an arith­momorphic concept, the separation consists of a void... (Ibid.)

He considers that such qualitative concepts violate the logical principle of non-contradiction in that, in certain instances ‘B is both A and not A’ and it is indeed for this reason that he calls them (showing the strong influence of Hegel on his work), ‘dialectical’. Violation of the law of non-contradiction is generally regarded as a somewhat serious business, and could indeed be held to render all coherent discourse impossible, so that Georgescu-Roegen’s presentation of this point is perhaps too strong. But from what he says elsewhere it seems clear that what he means here is, that within the penumbra that separates a qualitative concept from its opposite, we have a ‘moment of becoming’ or a qualitative change — i.e. ‘B’ is in transition from being ‘not A’ to being ‘A’. The law of non-contradiction is, there­fore, perhaps better thought of as being suspended, rather than violated, in that we have a zone of qualitative transition, which cannot be grasped by the purely arithmomorphic, and thus intrinsically static, symbolism of conventional logic.

This interpretation would seem to be confirmed by his later statement that one of the essential features of a qualitative concept is that as a ^l... penumbra surrounded by another penumbra it confronts us with an infinite regress...’ (ibid., p. 54), whilst claiming that:

Far from being a deadly sin, the infinite regress of the dialectical penumbra constitutes the salient merit of the dialectical concept; as we shall see it reflects the most essential aspect of change. (Ibid., p. 47.)

One of his central arguments in this context, is that the application of arithmomorphic concepts to some aspects of social reality leaves ‘qualitative residuals’. He sees the ‘indifference’ curves of ‘standard economics’ as being residuals of this type projected on to a cardinally quantifiable mapping of commodity combinations, and holds that much the same is true of conven­tional economic presentations of the production function.

The distinction that Georgescu-Roegen wishes to establish is far from unproblematical, as he himself would be the first to admit. There is in fact a powerful school of thought which rejects the belief that any meaningful distinction of this sort can be drawn, and insists that what have been here called qualitative concepts are in fact pre-scientific first approximations, which must be analysed into quantifiable and empirically operational components.¹⁶

However, this assertion is not self-evidently true, and indeed, the progress of social science in general, if measured by such a standard appears exceedingly limited.¹⁷ It can indeed be argued that it is precisely the empiricist emphasis upon quantification, together with its implications for verification that has led to frequent misinterpretations of Smith. Much of what he was attempting to do is no longer regarded as ‘scientific’ because of changes in the methodology and perspectives of science itself. An empiricist interpretation of Smith tends to focus on limited and selective aspects of his overall model, which happen to conform to empiricist preconceptions, but this leaves out key elements of his work which in fact give it unification. As a result, his problems and objectives, in so far as they are noticed at all, are not seen as legitimate science, and thus the major part of his work, which is qualitative in its orientation, is ignored.

Underlying this dispute about the status of qualitative concepts, is a fundamental difference of perspective concerning the ultimate structure of reality. On the one hand we have a view of the world which sees it as being composed of discrete, atomistic, ‘simples’ of some sort, whilst on the other hand it can be seen as a continuum.¹⁸ In the former case our concepts are, or can be, representations of pre-existing discrete entities and the application of discrete symbolism to the world is theoretically unproblematical. In the latter case, we ourselves must somehow ‘cut up’ a continuous reality into conceptually manageable ‘lumps’, and such a conceptual compartmentalisation is presum­ably a precondition of being able to think at all.

This dichotomy seems to be as old as philosophy itself, being present in the contrast between Plato’s view of‘the real’ as being composed of discrete ‘forms’¹⁹ and Aristotle’s notion of a qualitative continuum.²⁰ It has reappeared in numerous guises; for example, in the perennial dispute between ‘nominalism’ and ‘realism’; in the early classical empiricist emphasis on the distinction between primary and secondary qualities,²¹ together with its associated doctrine of the a posteriori nature of general concepts, and in the continuing debate between ‘methodological individualism’ and ‘methodological holism’.²² It perhaps finds its most concise expression in the work of Ludwig Wittgenstein, whose early attempts to show how language ‘hooks on to the world’ by being the formal analogue of the logically discrete, atomistic, ‘simples’ that form the ultimate structure of reality²³ became, in his own opinion, untenable, as the result of its inability to handle purely qualitative, and, therefore, logically inter­dependent, phenomena, such as colour.²⁴

For the present purpose it manifests itself most relevantly in the dichotomy between the empiricist view of verification, and what can be called the dialectical view. The empiricist position can be characterised as seeking the refutation/confirmation of propositions or hypotheses by referring them to an atomistic ‘given’ which is static in the sense that only quantitative change is postulated. The discrete ‘objects’ or ‘facts’ to which verification is linked are not conceived of as being in themselves capable of qualitative transition: it is for precisely this reason that in all the numerous variants of empiricism, ‘meaning’ is identified with ‘verification’. This remains true despite the fact that the somewhat naive tenets of ‘logical positivism’ have now been more or less abandoned in favour of more sophisticated models which allow for the role of theory.

Interestingly, the distinction we are seeking to pin down finds clear expression in the literature of Samuel Beckett, for this is surely what he has in mind when, in his play, Watt, he says:

The change. Of what did it consist? It is hard to say. Some­thing slipped... There is a great alp of sand, one hundred metres high, between the pines and the ocean, and there in the warm moonless night, when no one is looking, no one listening, in tiny packets of two or three millions the grains slip, all together, a little slip of one or two lines maybe, and then stop. All together, notone missing, and that is all...²⁵

Watt is generally interpreted as being a ‘logical positivist’ and as such is obsessed with quantification; the passage appears to postulate the existence of change which is purely qualitative, and thus outside the scope of quantitative comprehension.

A further implication of the empiricist view is that, in so far as qualitative concepts have any role at all to play, it is in the heuristic context and not in the context of verification, but since meaning is so tightly linked to verification²⁶ such concepts are left without any formal status.²⁷ By contrast, in the dialectical view, verification is to be found in the ‘confirmation’ of genetic accounts of ‘becoming’ — i.e. of qualitative development. The ‘real’ is the totality of a continuum of dynamic development, not an atomistic structure existing at any given moment in time. This continuum is conceived of as being both spatial and temporal, and the concepts we devise to reflect it are thus seen as a ‘cutting up’ of an essentially continuous reality. Dialectical models are thus inextricably linked to historical context. They are designed to ‘reflect’ or ‘mirror’ the dynamic principles which are believed to govern reality. Borrowing Hegelian terminology, we might say that they have a dual status being conceived of as both concept and mode of being.

A word of warning is perhaps appropriate at this stage. As a direct consequence of the present broadening of Lakatos’s definition of science, to include historical accounts of qualitative change, terms such as ‘model’ and ‘theory’ have taken on a wider range of application than is normally found in empiricist inter­pretations of science. The notion of‘reflecting’ dynamic principles of change, which was discussed above, falls outside the rigorous definitions of scientific activity found in such interpretations, and the position argued for here is that these standards must be relaxed and broadened in order to see that Smith’s historical approach is scientific in intention. Indeed if we work with the standard empiricist definitions of model and theory, then, as numerous writers from Shumpeter onwards have pointed out, there is little that qualifies as scientific in Smith’s work outside of the first two books of the Wealth of Nations. Given such an interpretation the ‘Theory of Moral Sentiments is in fact nothing of the kind, being at best a system of pre-scientific first approximations.

A further (closely related) aspect of dialectical models of human behaviour, is that the empiricist notion of causality, where cause and effect are discrete entities in a deterministic chain, is replaced by the notion that cause and effect are contained in the same concept or principle. It is this dialectical tension that gives rise to what Georgescu-Roegen calls ‘the infinite regress of the dialectical penumbra’ (as discussed above). The social world is not seen as a causally linked network of discrete objects, but as a complex of self-generating processes in which qualitative changes can occur. The ‘given’ empirical reality at any ‘moment’ is the dynamic process of ‘becoming’.

There are of course many ambiguities in the notion of an ‘internal’ cause, not least of which is that change appears to be both caused and uncaused. But in its defence it must be said that all notions of causality contain paradoxes. It is not, however, intended here to enter into metaphysical discussions of causality, since it is adequate for the present purpose to identify the implications of a dialectical approach to methodology.

In brief summary of the above, it can be said that in the empiricist conception of science, it is concerned with the quantitative extrapolation of ‘the given’, and in as much as there is change which is irreducibly qualitative, it must of necessity be beyond the scope of scientific enquiry. This is not to say that empiricist science cannot take account of qualitative change, but it can only do so ex post, by constructing a model which incorporates the new parameters. At the formal, or analytical level (as opposed to the purely descriptive), it proceeds, as it were, by discrete jumps from one qualitative nexus to the next.

Dialectical science on the other hand seeks to comprehend the moment of qualitative transition itself — its ‘verification’ thus lies in the ability of its principles to reflect such transitions. To do this, a scientific dialectical model must isolate features that are, in some sense, essential principles of movement in society, and give these principles speculative development. The question must of course arise as to how we know that the principles chosen are the essential ones. It is here that the generality of Lakatos’s schematisation of science is of value, for by seeing dialectical models as Lakatosian ‘research programmes’, we can at least in principle apply a modified form of the progressive/degenerative test. A dialectical research programme will then be classed as progressive, in so far as it generates concepts and hypotheses which have what will here be called ‘genetic’ empirical content, in that they make meaningful references to, and provide coherent explanations of, aspects of a concrete, historical, society which is undergoing (or has undergone) qualitative change.

Seen from this perspective ‘verification’ has a very different meaning from that found in contemporary science, being linked with plausibility, coherence and explanatory power rather than prediction. A strict empiricist would, of course, argue that predictive power is the ‘touchstone’ of science and that the other factors mentioned simply do not constitute verification. For the present purpose, however, it is sufficient to show that Smith believed that they do, and this will be done in due course, when it will be shown that his position is fully consistent with the present interpretation.

The characterisation of the dialectical model as so far developed is not sufficiently restrictive; the ‘objective idealism’ of Hegel, and possibly even the ‘subjective idealism’ of Fichte, would qualify as candidates for the status of dialectical models as they have been depicted here, although, since neither model could lay serious claim to genetic empirical content, we should not have too much trouble in eliminating them. The definition can be made more restrictive, however (thus excluding them at the outset), if we introduce the notion of dialectical materialism. The issues here are complex, and at this stage only a partial characterisation will be provided; this will be further developed during the application of the whole schematisation to Smith’s work.

At the most general level we can say that the ‘materialistic’ element in dialectical materialism resides in the view that the economic organisation of a society is the fundamental factor in shaping its institutions and its intellectual activity. Such a perspective is not of course confined to dialectical models, and many ‘empiricist’ economists, and some (but by no means all) historians and political theorists would accept it as non- controversial. The dialectical materialist view is however distinctive, both in the stress that it places on the interaction of economic organisation and socially generated individual con­sciousness, and in its emphasis on the dynamic, historical, nature of that interaction. Here, the economic ‘base’ of society is both cause and effect of people’s conscious activity. It generates and constrains the boundaries within which conscious activity is possible, but at the same time it is the product of that activity.

Dialectical materialism is thus to be contrasted not only with the ‘idealism’ of Hegel²⁸ or Collingwood, but also with the deterministic materialism of, say, Hobbes or Helvetius.

One of the basic strengths of dialectical materialism is that it does attempt to occupy this middle ground, and thus formally denies the duality of thought and matter implicit in most philosophy down to Hegel. One of its basic weaknesses is that it fails to provide an adequate account of the precise nature of the interaction which it postulates, so that the position taken is in constant danger of being conflated with the doubtless true, but trivial, proposition that ‘everything depends upon everything else’. Some of the best work in the Hegelian-Marxist tradition has been addressed to this issue but it can hardly be said to have been satisfactorily resolved. It is however beyond the range of this book²⁹ to pursue this question further as far as Hegel and Marx are concerned, although further reference will be made to it in connection with Smith.

In summary of the above characterisation it can be said that the dialectical materialist perspective has, as the ‘subject’ of history, people creating their social world by conscious productive activity, in concrete systems of economic organisation, and not any extra-human subject such as ‘Spirit’, ‘Collective Mind’ or ‘Reason’.

This leads me to a not uncontroversial claim which must be justified as the text unfolds. The ‘labour theory of value’ at one level is nothing more, or less, than a technical, economic, assertion that the labour embodied in the production of a commodity deter­mines its exchange value; this indeed is, to a large extent, how the theory functions in the work of Ricardo. But, on a broader definition it becomes something more fundamental for it is then the assertion that people through their productive activity create their own social reality. This, it will be argued, is how the theory functions in the work of Adam Smith; it is the fundamental metho­dological perspective that shapes and guides virtually everything that he has to say about society.³⁰

It is necessary here to enter a disclaimer. If we work with the narrower, technical definition of the labour theory of value, we are almost inevitably led into a discussion of what has become known as the ‘transformation problem’ and into an economic analysis of Ricardo’s and Marx’s criticisms and extension of Smith’s discussions, of ‘labour embodied’ and ‘labour commanded’. Such a treatment is the legitimate territory of the economist, Neo-classical or Marxian, but not of the present work which has chosen other, and less conventional, ground. I have nothing to contribute to the debate and will refer to it again only in passing.

Before concluding this chapter it might be appropriate to provide a brief summary of the methodological equipment developed so far, and then to indicate how it will be used in the remainder of the b∞k.

It has been argued that the notion of Research Programmes, as presented and developed by Imre Lakatos, gives a schematisa- tion that is both logically consistent, and plausible in that it provides a reasonable explanation of the actual progress of scien­tific endeavour. The model provides us with a coherent explanation of the interaction of ‘metascience’ and empirical testing, and challenges many established beliefs as to the nature of falsification and/or verification in the context of scientific theories. Further, because of its vision of science as a process, and thus intrinsically dynamic in nature, it can be extended to cover socio-historical models of qualitative change, which fall outside the traditional (empiricist) definitions of scientific activity.

It is the function of Chapter 3 to show that the methodological position arrived at in this process is fully in accord with Smith’s own explicit discussions of the nature of science, and it will there be suggested that the present chapter has in fact used modern philosophical concepts to reconstruct and formalise the perspec­tive found in Smith’s own ‘philosophical’ writings.

Later chapters will seek to demonstrate that much of his work is amenable to penetration when approached from this perspec­tive, and that some of Smith’s key concepts then undergo subtle, but not unimportant, changes of meaning.

Notes

1. K. Popper, The Logic of Scientific Discovery (Hutchinson, London, 1959).

2. I. Lakatos, ‘Falsification and the methodology of scientific research programmes’ in I. Lakatos and A. Musgrave (eds), Criticism and the Growth of Knowledge (Cambridge University Press, London, 1970) and also I. Lakatos, ‘The methodology of scientific research programmes’ in Philosophical Papers I (Cambridge University Press, Cambridge, 1978).

3. It is rather ironic that Popper, who has been a severe (if unconvincing) critic of Hegel, has himself postulated a schematisation of the process of the growth of knowledge which is formally (although not perhaps epistemologically) very close to Hegel’s position. For criticism of Hegel (and Marx) see K. Popper, The Open Society and Its Enemies (2 vols., Routledge & Kegan Paul, London, 1945), vol. 2, and The Poverty of Historicism (Routledge & Kegan Paul, London, 1957). Popper’s own schematisation is to be found in Objective Knowledge (Oxford University Press, Oxford, 1972).

4. Cf. S. Kuhn, The Structure of Scientific Revolutions (Chicago University Press, Chicago, Ill., 1970).

5. I. Lakatos, ‘Falsification and the methodology of scientific research programmes’ in I. Lakatos and A. Musgrave (eds), Criticism and the Growth of Knowledge (Cambridge University Press, London, 1970), p. 119.

6. Lakatos does not normally work with the traditional tripartite division of a scientific theory into ‘calculus’, ‘model’ and ‘linking terms’, but for the purpose of illustration we can do so, identifying the calculus with the formal relationships explicit within the theory, and the model with its implicit or explicit epistemology. The ‘linking terms’ would then be analogous to the implicit ‘interpretations’ made within the perspective of the model. On the role of models in science, see N.R. Campbell, Foundations of Science (Dover, New York, N.Y., 1957) and M.B. Hesse, Models and Analogies in Science (Sheed & Ward, New York, 1963).

7. E. Mach, The Science of Mechanics (Open Court, LaSalle, Ill., 1942).

8. J.A. Schumpeter, History of Economic Analysis (Allen & Unwin, London, 1972). Cf. for example, p. 41 et seq.

9. There is a substantial secondary literature on Lakatos. Cf. for example, S.J. Latsis, ‘A research programme in economics’ and A. Leijonhufvud, ‘Research programmes in economic theory’, both in S.J. Latsis (ed.), Method and Appraisal in Economics (Cambridge University Press, Cambridge, 1976).

10. Lakatos himself provided a superb application of his model in I. Lakatos, Proofs and Refutations: The Logic of Mathematic Discovery (Cambridge University Press, Cambridge, 1976). Cf. also D. Bloor, ‘Two paradigms for scientific knowledge’, Science Studies (1971); and C. Howson (ed.), Method and Appraisal in the Physical Sciences (Cambridge University Press, Cambridge, 1976).

11. Cf. P. Urbach, ‘Progress and degeneration in the I.Q. debate’, British Journal for the Philosophy of Science (1974).

12. Notably by M. Blaug, The Methodology of Economics (Cambridge University Press, Cambridge, 1980). Cf. also S.J. Latsis, ‘A research programme in economics’ in S.J. Latsis, Method and Appraisal in Economics (Cambridge University Press, Cambridge).

13. R.L. Meek, Precursors of Adam Smith (1750-1775) (Dent, London, 1973), p. viii.

14. The existence of ‘paradigms’ in economic thought is debated in GJ. Stigler, ‘Does economics have a useful past?’, History of Political Economy (1969); and A.W. Coats, ‘Is there a structure of scientific revolutions in economics?’, Kylos (1969), and in D.F. Gordon, ‘The role of the history of economic thought in the understanding of modern economic theory’, American Economic Review (1965). Cf. also H. Katouzion, Ideology and Method in Economics (Macmillan, London, 1980).

15. N. Georgescu-Roegen, The Entropy Law and the Economic Purpose (Harvard University Press, Cambridge, Mass.). Georgescu-Roegen’s published work has been of considerable influence in shaping the general perspective of this book.

16. The literature on this topic is considerable, and indeed, obscure; however, lucid discussions are to be found in R. Dubin, Theory Building (Free Press, New York, N.Y., 1978) and A. Kaplan, The Conduct of Inquiry; Methodology for Behavioural Science (Crowell, New York, N.Y., 1964).

17. Cf. H. Woolf (ed.), Quantification: A History of the Meaning of Measurement in the Natural and Social Sciences (Irvington, New York, N.Y., 1961) and P. Doreian, Mathematics and the Study of Social Relations (Schocken, New York, N.Y., 1971). An interesting historical treatment of probability concepts is to be found in I. Hacking, The Emergence of Probability (Cambridge University Press, Cambridge, 1975).

18. A.O. Lovejoy, in The Great Chain of Being (Harvard University Press, Cambridge, Mass., 1936), makes some perceptive comments upon this topic.

19. For example, in the Phadros, the Meno and the Republic. Also cited here can be the Pythagorean conception of the world, which was unambiguously atomistic and arithmomorphic; very similar perspectives are also to be found in Kepler’s work, and that of Galileo.

20. As presented in the Physics. For Aristotle all forms of being were seen as forms and types of movement.

21. Cf. J. Bennett, Locke, Berkley and Hume (Clarendon Press, Oxford, 1971), for a detailed discussion and also H.H. Price, Hume’s Theory of the External World (Clarendon Press, Oxford, 1948). Von Leyden has said, on this topic, that ‘as the result of the contemporary mechanistic view of nature it was natural for (Locke) as it had been for Galileo and Descartes, to single out as real the so-called primary qualities of bodies such as extension figure and motion: they were quantitative and hence mathematically definable characteristics and could also be regarded as both necessary and sufficient conditions of sensory perception.’ W. Von Leyden, Seventeenth-Century Metaphysics (Duckworth, London, 1968), p. 46.

22. Cf. K. Popper, The Poverty of Historicism (Routledge & Kegan Paul, London, 1957) and J.N.N. Watkins, ‘Historical explanation in the social sciences’, British Journal for the Philosophy of Science (1952). The whole debate is well reproduced in P. O’Neill (ed.), Modes of Individualism and Collectivism (Heinemann, London, 1973). Cf. also I. Krimerman (ed.), The Nature and Scope of Social Science: A Critical Anthology (Meredith, New York, N.Y., 1969).

23. L. Wittgenstein, Tractatus Logico-Philosophicus (Kegan Paul, Trench, Trubner, London, 1933), p. 37.

‘203 In the atomic fact objects hang one in another like members of a chain.

2031 In the atomic fact the objects are combined in a definite way.

2032The way in which the objects hang together is the structure of the atomic fact.

204 The totality of existing atomic facts is the world.’

24. Cf. L. Wittgenstein, Philosophical Investigations (Blackwell & Mott, Oxford, 1972). There is a good discussion of this issue in A. Kenny, Wittgenstein (Penguin, Harmondsworth, 1975). See also articles by A. Kenny, R.M. White and R. Rhees in Understanding Wittgenstein, Royal Institute of Philosophical Lectures, Vol. 7, 1972/73.

25. S. Beckett, Watt (Calder, London, 1976), p. 41.

26. The literature on meaning in empiricist perspectives is vast; good representative treatments are G. Ryle, The Theory of Meaning in British Philosophy in the Mid-Century (Allen & Unwin, London, 1957), R. Carnap, Meaning and Necessity (Chicago University Press, Chicago, Ill., 1956) and W. Quine, ‘Designation and existence’, Journal of Philosophy (1939).

27. Cf. G. Radnitzky, Contemporary Schools of Meta-Science (Akademi Forlaget, Gotenborg, 1970) and P. Feyerabend, Against Method (New Left Books, London, 1975). Feyerabend’s book is, in part, a critique of Lakatos’s work from the standpoint of ‘epistemological anarchy’.

28. Hegel is notoriously difficult to categorise, precisely because he is a dialectician, and thus looks at the interaction of ‘spirit’ and matter. Despite this (and despite some claims to the contrary) his orientation is such that I have little hesitation in classing him as an ‘Idealist’. The true subject of his metaphysics is ‘Spirit’, and it could well be argued that he regarded ‘Matter’ as merely an aspect of ‘Spirit’.

29. Cf. G. Lukacs, History and Class Consciousness (Merlin Press, London, 1971), K. Korsch, Marxism and Philosophy (N.L.B., London, 1970) and J. Habermas, Knowledge and Human Interest (Heinemann, London, 1972). Lukacs later renounced this work under pressure from the Soviet state, and at the time of his death he still held that it was fundamentally mistaken — arguably however, it remains the most coherent attempt at elucidation.

30. This interpretation is close to the view held by Meek, who has said that the essence of the labour theory of value is that the ^ζ... fundamental relationships into which men enter with one another in the field of production ultimately determine the relationships into which they enter in the field of exchange.’ R.L. Meek, Studies in the Labour Theory of Value (Lawrence & Wishart, London, 1973), p. 73.