rs541: Yumi: finance

Keynes's General Theory: Interpreting the Interpretations Author(s): Bill Gerrard Source: The Economic Journal, Vol. 101, No. 405 (Mar., 1991), pp. 276-287 Published by: Wiley on behalf of the Royal Economic Society Stable URL: http://www.jstor.org/stable/2233818 Accessed: 24-04-2016 13:13 UTC

REFERENCES Linked references are available on JSTOR for this article: http://www.jstor.org/stable/2233818?seq=1&cid=pdf-reference#references_tab_contents You may need to log in to JSTOR to access the linked references.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted

digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about

JSTOR, please contact support@jstor.org.

Royal Economic Society, Wiley are collaborating with JSTOR to digitize, preserve and extend access to The Economic Journal

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

The Economic Journal, ioi (March I99I), 276-287

Printed in Great Britain

KEYNES'S GENERAL THEORY: INTERPRETING THE

INTERPRETATIONS*

Bill Gerrard

Keynes's General Theory has given rise to a variety of Keynesian research programmes. The development of these different Keynesian research pro- grammes is well-documented (see, for example, Coddington, I976, Gerrard, I988; Hamouda and Harcourt, I988, for surveys of Keynesian and post- Keynesian economics). However less attention has been paid to explaining a striking feature of this Keynesian diversity, namely, the stress placed on discovering the real meaning of Keynes's General Theory. The legitimacy of any particular Keynesian research programme has been judged with regard to the authenticity of its implied interpretation of Keynes. Inevitably this concern for authenticity has generated much controversy, enveloping Keynesian economics in a 'doctrinal fog' (Blaug, I980, p. 22I). This paper attempts to pierce that Keynesian doctrinal fog. The central thesis is that the causes of the controversy surrounding Keynes's General Theory lie, in part, in the different presuppositions made about the nature of interpretation. It is argued that much light can be shed on the Keynesian debate by drawing on the study of hermeneutics.

The structure of the paper is as follows. Section I discusses the atomistic view

of interpretation which is implicitly presupposed by most contributors to the Keynesian debate. Two variants of the atomistic view are considered: the objectivist/essentialist approach and the relativist approach. Section II provides an alternative presupposition, the organicist view of interpretation, as exemplified by Ricoeur's dialectical approach in hermeneutics. The paper concludes in Section III with a re-examination the Keynesian debate in the 4ight of the organicist view of interpretation.

I. THE ATOMISTIC VIEW OF INTERPRETATION

A principal aim of Keynesian economics has been to give a definitive answer to the question 'What does Keynes's General Theory really mean?'. Much of the resulting controversy arises from the nature of the question itself. In asking the

question an atomistic view of interpretation is presupposed. The author, the text and the reader are treated as individual atomistic entities which are inter- related in a purely external manner: the author produces the text which the reader interprets. There are two variants of the atomistic view of interpretation: the objectivist/essentialist approach and the relativist approach.

* This paper represents a development of themes contained in Gerrard (1 989). I should like to thank Paul Anand, Roger Backhouse, John Brothwell, Meghnad Desai, Athol Fitzgibbons, John Hillard, Brian Hillier and the participants at the RES conference as well as the editors and an anonymous referee for much in the way of helpful criticism. The usual disclaimer applies.

[ 276 ]

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

[MARCH I99I] KEYNES S GENERAL THEORY 277

I. i. The Objectivist/Essentialist Approach

From the perspective of the objectivist/essentialist approach, the aim of

interpretation is the rational reconstruction of the text in order to recover the

author's original meaning. Interpretation is seen to be problematic because the

author's meaning is hidden. The latency of the original meaning creates

confusion and generates the possibility of multiple interpretations. The task of

the interpreter is to resolve this confusion by discovering the 'true' meaning of a text. This presupposes that the true meaning is knowable.

Within the objectivist/essentialist approach, Keynes's General Theory is

viewed as containing a single essential meaning which is hidden as the result of

the confusion created either by Keynes himself or by the economics profession

in its reading of Keynes. The belief that Keynes's General Theory contains a single essential meaning is shared by most interpreters of Keynes. Leijonhufvud (i 968) claims to have found the 'economics of Keynes' as opposed to

'Keynesian economics', while Shackle (I967, ch. 12) seeks Keynes's 'ultimate

meaning'. Fender (i98I) sets out to find the 'exact nature of the theoretical contribution of Keynes' (pp. I, 2) and, similarly, Chick (I983) attempts to remedy the fact that the 'macroeconomics that has been developed after

Keynes, though claiming inspirations from the General Theory, in my view has not, with some outstanding exceptions, been macroeconomics after the manner

of Keynes - with the method and perspective and insight of Keynes' (p. v). From the objectivist/essentialist perspective, it is necessary to explain why

the essential meaning of the General Theory is hidden. There are three broad

types of explanation:

(i) The confusion is author-generated

A number of writers have suggested that Keynes himself is the cause of the

confusion. There are a number of variants of this 'author-generated confusion' thesis:

(i) Technical incompetence. It is often argued that Keynes had limited analytical abilities. For example, Hahn (i 982, pp. X, Xi) writes that 'I consider that Keynes had no real grasp of formal economic theorising (and also disliked it), and that he consequently left many gaping holes in his theory.' This follows a famous remark by Shove that 'Maynard had never spent the twenty minutes necessary to understand the theory of value' (quoted in Robinson, I964, p. 79).

(ii) The 'vision' thesis. Confusion arises as the inevitable consequences of the difficulties which Keynes faced in trying to formulate his underlying vision in a precise analytical manner. It is a line of argument originating with Schumpeter (1946, p. 501) who distinguishes between Keynes's vision, that is, his 'view about the basic features of society, about what is and what is not important' and Keynes's technique, that is, the 'apparatus by which he conceptualises his vision and which turns the latter into concrete propositions or theories'. According to Schumpeter the General Theory is the final result of a long struggle by Keynes to make his vision analytically operative. Leijonhufvud

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

278 THE ECONOMIC JOURNAL [MARCH

(pp. Io, II) adopts Schumpeter's distinction, arguing that Keynes was not

entirely successful in translating his vision into a logically watertight model.

(iii) Stylistic difficulties. Keynes is often accused of a lack of clarity. This is

suggested by O'Donnell (I989a, p. 6) as a reason for the difficulties in the interpretation of Keynes. O'Donnell approvingly quotes Wittgenstein's maxim

that if' anything can be said, it can be said clearly'. Leijonhufvud (pp. Io, I I)

goes as far as to say that the General Theory is 'a badly written book' and that

the need for 'repairs' has led to confusion because different writers have

corrected Keynes's model in inappropriate ways. It is also argued that Keynes's

style is too loose and vague. This echoes Whitehead's criticism of Keynes's dissertation on probability as using the style of literature, not the style of logic

and philosophy.

(iv) Inconsistencies. Some writers have suggested that Keynes did not have a coherent and consistent vision. It is an argument used by Leijonhufvud but also

by Robinson (1973, p. 3) in her well-known comment that 'there were moments when we had some trouble in getting Maynard to see what the point

of his revolution really was'.

(v) The 'grand mistake' thesis. A recent extreme example of the 'author- generated confusion' thesis is the neo-Ricardian argument that Keynes made a mistake in retaining the neoclassical concept of the marginal efficiency of capital (Milgate, I982). This created the possibility that, if the rate of interest is sufficiently low, there will be sufficient investment to maintain full employment. According to Milgate this undermined the principle of effective demand, the essence of the General Theory. Furthermore it forced Keynes to

develop explanations of interest rate maladjustment; hence Keynes's misguided emphasis on expectations and liquidity preference.

(2) The confusion is reader-generated

An alternative explanation of the confusion surrounding the General Theory is to focus on the actions of the audience. Again there are a number of variants of this explanation:

(i) Inappropriate framing. Readers have interpreted Keynes relative to an inappropriate frame of reference. This results in the development of a variety of 'subjective' interpretations based on personal beliefs and ideological and normative biases. In the process the 'objective' meaning of Keynes becomes lost. This line of argument is epitomised by Leijonhufvud's distinction between 'Keynesian economics' and the 'economics of Keynes'. It is an argument repeated by Fitzgibbons (I988, pp. 1-5) when he points towards the problem of 'systematically biased interpretation'.

(ii) Selective reading. A closely related variant to the inappropriate framing argument is the problem created by readers considering only parts of the text. Thus O'Donnell (I989 a, p. 4) sees the main reason for the multiple interpretations of Keynes as the 'tendency to base interpretations on selected parts, rather than the whole of his relevant writings'.

(iii) Reliance on secondary sources. The tendency towards multiple interpret- ations of Keynes has been exacerbated by the tendency to read about Keynes rather than to read Keynes himself.

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

I99I] KEYNES S GENERAL THEORY 279

(3) The confusion is generated by differences in the composition of the stock of relevant textual evidence

Different interpretations may arise because of differences between interpreters with regard to the definition of the text to be analysed. There are two variants of this explanation:

(i) Which text? With an author as productive as Keynes, an inevitable problem is whether to interpret the target text in isolation or in the context of the author's other related writings. This can lead to multiple interpretations if readers find inconsistencies between different texts. The definition of 'other relevant writings' may vary between interpreters, particularly with regard to the relative weights to be attached to earlier and later writings as well as to formal writings and more informal sources such as speeches, unpublished papers and private correspondence. Thus, for example, the recent emergence of the 'new' Keynesian fundamentalism associated with Carabelli (I988),

Fitzgibbons (I988) and O'Donnell (1989 a) amongst others, represents a shift of weight in favour of Keynes's early philosophical papers, a source largely ignored by previous interpreters.

(ii) Changes in availability. Not only can there be 'subjective' differences about the definition of the text, there may also be 'objective' differences over time as the stock of documents available for interpretation changes. Such changes have been particularly important in the interpretation of Keynes's General Theory. The publication of the Collected Writings, especially volumes XIII and XIV, shed considerable light on the development of Keynes's thought immediately before and after the publication of the General Theory. The stock of documents expanded subsequently with the publication of volume XXIX as the result of the discovery of a laundry basket of previously unknown papers by Keynes. These included early drafts of the General Theory focusing on the concept of a monetary production economy.

These various sources of confusion create the possibility of multiple interpretations. Convergence towards the correct interpretation is usually presumed to be ensured by the use of consistency with the textual evidence as the criterion of choice between competing interpretations (for example, Leijonhufvud, I968, p. 8). Interpretation is viewed, therefore, as a scientific problem. The scientific nature of interpretation has been highlighted by Stigler (I965) with particular reference to the problem of multiple interpretations of Ricardo. Stigler argues that hand-picked quotations are insufficient to validate any particular interpretation. He proposes instead the use of two different principles of interpretation: the principle of scientific exegesis and the principle of personal exegesis. Scientific exegesis is interpretation which aims to maximise the value of a text to the science. In this case the text of an interpretation is its consistency with the main analytical conclusions of the author. This type of interpretation is concerned with the 'strong' form of the text, that is, an amended form of the text which removes 'blemishes' such as logical errors and tautologies. Personal exegesis, on the other hand, aims to discover what the author really believed and thus the test of an interpretation becomes consistency with the author's style, that is, what the author actually wrote.

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

280 THE ECONOMIC JOURNAL [MARCH

According to Stigler this latter form of interpretation is of no direct relevance

to scientific progress.

Stigler's separation of scientific and personal exegesis has been the subject of

recent criticism by Aksoy (i 989) and Hollander (i 989). Both question whether scientific exegesis can really be considered as interpretation when it denies any significance to what the author really believed. Hollander criticises scientific exegesis for leading to the distortion of an author's writings since it justifies disregarding those parts of an author's writings deemed by the interpreter to

be 'blemishes'. Hollander describes this as a 'lazy man's procedure' and argues for the need to explain the 'residuals', that is, the differences between the 'strong' form of the text and what the author actually wrote.

The controvery surrounding Stigler's scientific approach to interpretation is

suggestive of an important, but often overlooked, characteristic of interpret- ation. 'Consistency with the textual evidence' is inadequate as an objective criterion for assessing competing interpretations. The definition of both 'consistency' and 'textual evidence' is open to debate. The textual evidence may be defined as the single target text only or it may include other relevant texts. Likewise 'consistency' is open to multiple interpretation. Stigler suggests two alternative definitions: consistency with the author's main conclusions (i.e.

scientific exegesis) and consistency with the author's beliefs (i.e. personal

exegesis). Recently O'Donnell (i989.b) has proposed a three stage consistency test for any interpretation of Keynes. According to O'Donnell, interpretations

should be: (i) internally consistent; (ii) consistent with quotations taken in context; and (iii) consistent with all of an author's writings. This definition of

consistency appears to be unnecessarily restrictive in at least two ways. First, it seems to impose on an author's writings a degree of integration and continuity over time which may be unwarranted. Secondly, it seems to exclude evidence drawn from sources other than the author's own writings.

Thus the choice between competing interpretations cannot be purely

objective. Such choices must always be made on the basis of certain presuppositions. Any particular interpretation has a conventional foundation, a set of presuppositions which are treated as beyond doubt. This conventional foundation includes presuppositions about the aims of interpretation, the consistency criterion and the relevant evidence. Different writers may base their interpretations on different conventional foundations. Furthermore the very presupposition that the author's meaning is hidden provides the means by which any particular interpretation can be rendered consistent with the text. Any apparent inconsistency between an interpretation and the text can be

explained away as the result of author-generated confusion. The objectivist/ essentialist approach has a built-in immunising stratagem with which to protect the validity of any particular interpretation. The parallel with recent developments in the philosophy of science is clear. The Duhem-Quine thesis on the underdeterminacy of empirical testing has led to the recognition that science consists of theoretical structures with conventional foundations as

exemplified by Kuhn's theory of paradigms and Lakatos's notion of scientific research programmes with hard cores.

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

I99I] KEYNES S GENERAL THEORY 28I

1.2. The Relativist Approach

The contradictions within the objectivist/essentialist approach to interpret- ation have led some to adopt the relativist approach. Rather than viewing interpretation as the discovery of the author's meaning hidden within the text, the relativist approach treats interpretation as the product of the reader imposed on the text. Thus the analysis of interpretation moves from being text-

centred to being reader-centred. There is no single essential meaning within a

text, only a variety of reader-determined meanings. Interpretations are always relative to the frame of reference within which the reader is operating. This

implies that there is no objective standard for a rational evaluation of alternative interpretations. The choice of interpretation is made by the

individual reader on the basis of a subjectively determined frame of reference. Any interpretation can only ever be treated as consistent with its own

underlying frame of reference. In this sense, one interpretation is as good as another.

From the relativist perspective, to ask the question 'What does Keynes's

General Theory really mean?' is really to ask 'What does Keynes's General Theory

mean to me, given my frame of reference?'. Readers choose that interpretation which makes sense within their own world-view. For example, since the neo- Keynesians adopt the choice-theoretic and market-theoretic perspective of mainstream economic theory, it follows that they interpret Keynes's General Theory to be dealing with the effects of imperfections such as non-atomistic

market structures or imperfect information which prevent the price mechanism

from operating effectively. The relativist approach, however, also suffers from self-contradictions. To

adopt the relativist approach is to accept that the interpretive process is indeterminate. Relativism implies that 'anything goes'; consistency with the text is always relative to a subjectively determined frame of reference. This

brings the validity of the process of interpretation itself into question. The relativist approach denies that interpretation can be the pursuit of under- standing beyond that which is relative to the individual's own frame of reference. But this runs counter to the explicitly stated aims of those who engage in interpretation. Interpretations are advanced on the basis of intellectual and cognitive properties which are deemed to transcend subjectivist concerns. In order to justify the public presentation of an interpretation, interpreters adopt an objectivist/essentialist approach when advocating their own particular interpretations. Relativism cannot be followed consistently since it would deny the intellectual properties which interpreters always claim for their own interpretations. Relativism is a position that can be entertained but never occupied. The relativist approach, through the force of its own logic, becomes self-contradictory. Relativism always reverts to a form of' back-door' objectivism in which writers use the relativist approach to criticise the interpretations of others but claim authenticity for their own interpretations. Practised in this form, relativism is but a variant of the 'reader-generated confusion' thesis.

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

282 THE ECONOMIC JOURNAL [MARCH

To summarise: Much of the confusion surrounding the interpretation of

Keynes's General Theory arises from the presupposed atomistic view of interpretation. The atomistic view, in both its objectivist/essentialist and

relativist forms, is self-contradictory. These contradictions have been carried

over into the interpretation of Keynes. Thus the task of piercing the Keynesian doctrinal fog requires, as a first step, moving beyond the atomistic view of interpretation.

II. THE HERMENEUTIC APPROACH TO INTERPRETATION

An alternative view of interpretation is the organicist view which stresses the

importance of context. The writing and reading of texts are processes which are

socially and historically contingent. The author, the text and the reader are not atomistic entities but form a dynamic whole in which the nature of any part is defined by its interrelationship with the other parts. It is a general vision of the process of interpretation which finds its expression primarily in the study of hermeneutics.

Hermeneutics is the study of interpretation. It originated as the study of the principles of biblical exegesis, deriving its name from Hermes, the messenger of the Gods. Gradually hermeneutics extended its scope beyond the confines of biblical exegesis, becoming the study of textual interpretation in general. It was systematised in the nineteenth century in the writings of Schleiermacher who set out the basic principles of hermeneutics. Schleiermacher stressed the importance of context, arguing that the historical conditions of the author must be taken into account in recovering the original meaning of the text. In particular, it was crucial to understand the original audience to whom the text was directed. Following Schleiermacher, hermeneutics became closely associated with the German historicist school, especially Dilthey, who came to regard the hermeneutic approach as a general methodology for the human sciences. During the twentieth century hermeneutics has undergone at least two marked changes in orientation. In the writings of Heidegger and Gadamer, hermeneutics emerged as a general philosophical position, becoming an integral part of the work of the Frankfurt School. Later hermeneutics took a more linguistic turn, becoming synonymous with the structuralist approach to

the study of language and symbols. The ever-changing nature of hermeneutics means that there is no single

coherent set of hermeneutic principles. Instead there are different conflicting traditions. Hirsch (1976) identifies three such traditions:

(i) The biblical/intuitionist tradition, in which the concern is for the 'spirit' of the text. Meaning is seen as not fully expressible in words, implying that the understanding of a text requires getting behind the text. Schumpeter's notion of vision is suggestive of the intuitionist tradition. Interpretation is more than logical deduction. It requires intuition in order to achieve 'communion' with the spirit of the text.

(ii) The legal/positivist tradition, in which the concern is for the 'letter of the law', that is, what is actually written. Meaning is seen as wholly contained

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

I99I] KEYNES S GENERAL THEORY 283

within the text and recoverable by means of logical deduction. It is a tradition

which has its origins in part in Jewish Talmudic scholarship. The latter is

explicitly invoked by Patinkin (1978) in support of his contention that the

students of Keynes's thought should concern themselves only with what Keynes

actually said.

(iii) The metaphysical tradition, in which the original meaning of the text is seen as unrecoverable, and, hence, unknowable. Given the seemingly nihilistic

implications of the metaphysical tradition for the interpretive process, it is a

tradition that has little parallel in the Keynesian debate.

Given these very different traditions in hermeneutics it is important from the outset that hermeneutics should not be seen as a set of unquestionable principles which can settle the debates over Keynes's General Theory once and

for all..Hermeneutics can be a double-edged sword.

A key concept in hermeneutics is the hermeneutic circle which arises from a

general paradox within any organicist approach. To know the whole, one needs to know the parts. But the parts can only be known in the context of their

interdependencies within the whole. Thus to know the whole requires that the

whole be pre-known. This circularity has a 'narrow' and 'wider' sense. In the narrow sense the text itself is treated as a whole while in the wider sense the text

is treated as part of the historical context. Within hermeneutics, the hermeneutic circle has been subject to two very

different interpretations (Hirsch, 1976). The 'old' or 'romanticist' her- meneutics considers the aim of interpretation to be the recovery of the author's original meaning. The hermeneutic circle is the recognition of the need to understand the historical context of the text. In contrast, the 'new' hermeneutics considers the hermeneutic circle as the rejection of objectivity. Interpretation is always relative to the historical context of the reader. Thus, just as with the atomistic approach to interpretation, the hermeneutic approach is caught in an objectivist-versus-relativist controversy.

An attempt to transcend the objectivist-relativist duality in hermeneutics is to be found in the writings of Ricoeur (I 976, 198I). Following Hirsch, Ricoeur views interpretation as a process of guess and validation in which interpreters marshall evidence to show a particular interpretation to be more or less probably true. Interpretation involves the logic of qualitative probability and uncertainty. However, unlike Hirsch, he does not view the aim of interpretation as the recovery of the author's original meaning which he considers to be 'a lost psychical event'. But the denial-of this objective standard does not entail that anything goes in interpretation. The text contains potential meaning to be actualised by readers but it constrains the possibilities of that actualisation. For Ricoeur interpretation is the appropriation of the reference power of a text, that is, the actualisation of a text's ability to disclose possible ways of looking at the world. Interpretation involves what Gadamer termed the 'fusion of horizons'. The world horizon of the reader becomes fused with the world horizon of the author with the text acting as the mediating link. The resultant interpretation is not the function of the text or the reader alone. Rather it is the outcome of a dialectical process.

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

284 THE ECONOMIC JOURNAL [MARCH

From the writings of Ricoeur and others it is possible to develop a hermeneutic framework for the analysis of interpretation using the following definitions:

(i) understanding: the construction of a text's meaning in its own terms; (ii) explanation: the presentation of the understood meaning in terms

accessible to a particular audience;

(iii) interpretation: the process of dynamic interaction between under- standing and explanation;

(iv) meaning: the outcome of the process of interpretation;

(v) judgement: the construction of a relation between the text and something external to it;

(vi) criticism: the explanation and evaluation of the judgement; (vii) application: the process of dynamic interaction between judgement and

criticism;

(viii) significance: the outcome of the process of application. For Ricoeur these concepts are abstract poles within a concrete whole.

Interpretation is a dialectic of understanding and explanation in which the understanding of a text affects its explanation and vice versa. Likewise application is a dialectic of judgement and criticism. At a higher level there is also a dialectical interaction between interpretation and application.

The dialectical approach of Ricoeur and other related organicist analyses of interpretation are able to provide an alternative set of presuppositions about the nature of interpretation to that associated with the atomistic view. This alternative set of presuppositions can be formulated in terms of four propositions.

PROPOSITION I. Interpretation is a multi-dimensional process with multiple objectives.

Interpretation involves understanding and explanation as well as judgement and criticism. It aims not only to provide meaning to a text but also to evaluate the significance of the text. The complexity of interpretation needs to be recognised by writers in order to overcome the confusion. This conclusion is

drawn by Aksoy (I989, p. 744) in his application of the hermeneutic perspective to the debates surrounding the interpretation of Ricardo: '... there is a great deal of confusion about the meaning and the purpose of the act of interpretation. In most cases, interpreters are found to be careless and inconsistent about the differences between interpretation, judgement, evalu- ation and application.'

PROPOSITION 2. Interpretation is not an objective process but this does not imply that anything goes.

There is no objective criterion of consistency with the textual evidence on which to base the rational evaluation of alternative interpretations. But anything does not go. The interpretive process is historically and socially contingent. Fish (I980) argues that interpreters are social beings, constrained by the set of characteristic beliefs of the particular interpretive community to which the interpreters belong. The interpretive community acts as the

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

I99I] KEYNES S GENERAL THEORY 285

intellectual arbiter, setting the cognitive standards by which new interpret-

ations are judged. The acceptance of a new interpretation involves both demonstration and persuasion. The interpretive community must be persuaded

to change its paradigm, that is, the conventional foundations underlying the currently-accepted interpretation. It is only after the interpretive community has been persuaded to change its paradigm that the explanatory power of the

new interpretation can be demonstrated.

PROPOSITION 3. The significance of an interpretation is not determined by correspondence with the author's original meaning.

The significance of an interpretation is something external to the text. It

depends on the successful application of the insights gained from the text. In the

case of an economics text, its significance can be seen as its ability to explain economic behaviour. The question of whether or not this is what the author really meant is largely irrelevant in this context. From this perspective, Stigler's distinction between personal and scientific exegesis can be justified to the extent that it represents the hermeneutic distinction between meaning and signific- ance.

PROPOSITION 4. The existence of multiple interpretations is not a problem.

Multiple interpretations of a single text arise partly because of the inescapable vagueness of the written word. It is this vagueness which gives the text its reference power, its ability to disclose different possible ways of looking at the world. The text constrains the range of potential meanings but cannot determine the actualisation of any specific meaning by the reader. The continued existence of multiple interpretations need not, therefore, be treated as a problem to the extent that it is indicative of the high reference power of the text and the heterogeneity of the readership. It only becomes a problem if interpreted as such from an objectivist/essentialist perspective.

III. CONCLUSION: HOW TO PIERCE THE KEYNESIAN DOCTRINAL FOG

Much of the doctrinal fog surrounding the interpretation of Keynes has been generated by the confusion of interpretation and application. Interpreting Keynes's General Theory has involved two questions, the question of in- terpretation, 'What does Keynes's General Theory mean?', and the question of application 'What is the significance of Keynes's General Theory for explaining how the economy actually works?'. The two questions are interdependent but they have tended to become interwoven in a very confusing manner. For example, Leijonhufvud (I968, ch. I) differentiates between what he calls the 'doctrinal-historical' question of understanding Keynes's General Theory and the task of discovering fresh perspectives with which to understand the economy and with which to assess economic theory. Leijonhufvud considers the latter to be the primary objective, the doctrinal-historical question being 'strictly secondary'. Yet the validity of the fresh perspective discovered by

11 ECS 101

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

286 THE ECONOMIC JOURNAL [MARCH

Leijonhufvud is assessed on the basis of it having 'a much firmer foundation in Keynes's writing than can be claimed for the [income-expenditure] in- terpretation' (p. iI). Thus Leijonhufvud considers his interpretation as the 'economics of Keynes'. Leijonhufvud does not assess his interpretation with regard to its relevancy to the understanding of the economy and economic theory but in terms of its authenticity as an exegesis of the real meaning of Keynes's General Theory.

Piercing the Keynesian doctrinal fog requires disentangling the question of interpretation and the question of application. The significance of Keynesian economics depends on its ability to provide an understanding of how the economy actually works. The significance of Keynesian economics does not depend on being the economics of Keynes. What Keynes himself believed is a question for the historians of economic thought, not for macroeconomists. This is not to say that interpretation is unimportant; quite the opposite. Rather the

point is that the usefulness of an interpretation depends on its ability to generate a better understanding of economic behaviour. The interpretation of Keynes's General Theory has relevance for macroeconomics if and only if it can provide access to new understandings of the macro economy. Distinguishing between meaning as the aim of interpretation and significance as the aim of application will do much to overcome the confusion surrounding Keynes's General Theory.

The continuing achievement of Keynes's General Theory is its ability to

generate a diversity of research programmes. This diversity has resulted from the fusion of Keynes's horizon as expressed in the General Theory with a series of very different horizons due to his interpreters and is evidence of the high reference power of Keynes's General Theory. It is a text which continues to disclose a number of different possible ways of looking at the macro economy. The recognition of this will clear the Keynesian doctrinal fog. The multiple interpretations of Keynes's General Theory need no longer be interpreted as a problem yet to be solved.

This conclusion that the economics profession should stop worrying about

the multiple interpretations of Keynes's General Theory is an example of the therapeutic role that hermeneutics can have in economics. Hermeneutics can help economists become more self-aware about some of the causes of controversies in the subject-field, thereby helping to resolve these controversies. This is the pragmatic justification for introducing hermeneutics into economics.

But hermeneutics can be a double-edged sword. Hermeneutics provides an attitude of mind, not a set of ready-made answers. Hermeneutic understanding can only emerge through a dialectical process of application to specific problems, the interpretation of Keynes's General Theory being one such

application. If hermeneutics is introduced into economics as an abstract methodology, it will generate more heat than light. Hermeneutics is a study of

controversy but it is also a study in controversy. The emergence of the rhetorical approach in economics exemplifies the possible dangers of replaying

the old controversies of other subject-fields. Hermeneutics can be an effective medicine but should be marked 'handle with care'.

University of York

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms

[MARCH I99I] KEYNES S GENERAL THEORY 287

REFERENCES

Aksoy, E. G. (I989). 'Problems of interpretation: hermeneutic circle, objectivity and validity in the

interpretations of Ricardo.' Proceedings of the I6th Annual Meeting of the History of Economics Society,

Richmond, Va., vol. 2, pp. 725-47. Blaug, M. (I980). The Methodology of Economics: Or How Economists Explain. Cambridge: Cambridge

University Press.

Carabelli, A. (I988). On Keynes's Method. London: Macmillan.

Chick, V. (I983). Macroeconomics after Keynes: A Reconsideration of the General Theory. Oxford: Philip Allan. Coddington, A. (I 976). 'Keynesian economics: the search for first principles.' Journal of Economic Literature,

vol. I4, pp. I258-73-

Fender, J. (I98I). Understanding Keynes. Brighton: Wheatsheaf. Fish, S. (I980). Is There a Text in This Class? The Authority of Interpretive Communities. Cambridge, Mass.:

Harvard University Press.

Fitzgibbons, A. (I988). Keynes's Vision. Oxford: Clarendon Press. Gerrard, B. (I988). 'Keynesian economics: the road to nowhere?' In J. M. Keynes in Retrospect: The Legacy

of the Keynesian Revolution J. V. Hillard, ed.) Aldershot: Edward Elgar.

(I989). 'Some notes on interpreting Keynes' General Theory.' IRISS, University of York Discussion

Paper, no. I40.

Hahn, F. H. (I982). Money and Inflation. Oxford: Basil Blackwell. Hamouda, 0. F. and Harcourt, G. C. (I988). 'Post-Keynesianism: from criticism to coherence?' Bulletin of

Economic Research, vol. 40, pp. I-33. Hirsch, E. D. (1976). The Aims of Interpretation. Chicago: Chicago University Press. Hollander, S. (I989). 'Principles of textual interpretation: illustrated by Ricardian growth theory.'

Proceedings of the I6th Annual Meeting of the History of Economics Society, Richmond, Va., vol. 2, pp. 763-74. Leijonhufvud, A. (1 968). On Keynesian Economics and the Economics of Keynes. Oxford: Oxford University Press. Milgate, M. (I982). Capital and Employment. London: Academic Press.

O'Donnell, R. (I989a). Keynes: Philosophy, Economics and Politics. London: Macmillan. (I989b). 'Keynes on probability, expectations and uncertainty.' Paper presented at the gth Keynes

Seminar, University of Kent, Canterbury. Patinkin, D. (1978). 'Keynes' aggregate supply function: a plea for common sense.' History of Political

Economy, vol. IO, pp. 577-96. Ricoeur, P. (1976). Interpretation Theory: Discourse and the Surplus of Meaning. Forth Worth, Texas: Texas

Christian University Press. (I98I). Hermeneutics and the Human Sciences. Cambridge: Cambridge University Press.

Robinson, J. (I964). Economic Philosophy. Harmondsworth: Penguin.

(I973). 'What has become of the Keynesian revolution?' In After Keynes. Oxford: Basil Blackwell. Schumpeter, J. A. (1946). 'John Maynard Keynes I883-I946.' American Economic Review, vol. 36, pp.

495-5I8. Shackle, G. L. S. (I967). The Years of High Theory. Cambridge: Cambridge University Press. Stigler, G. (I965). 'Textual exegesis as a scientific problem.' Economica, vol. 32, pp. 447-50.

This content downloaded from 192.102.230.248 on Sun, 24 Apr 2016 13:13:41 UTC All use subject to http://about.jstor.org/terms