M. J. R. GAFFIKIN
The Methodology of Early Accounting Theorists
If, as Watts and Zimmerman (1986) suggest, a utiique methodological foundation is the hallmark of a mature discipline, accounting fails to qualify.
In this article, methodological arguments of accounting theorists in the first part of this century are examined. 'Pattern Modeling' is suggested as a basis for methodological appraisal. The conclusion reached is that the major theoretical works in accounting were methodologically unsophisticated — particularly in comparison with other disciplines such as economics.
Key words: Accounting theory; Scientific method; Theory.
INTRODUCTION: UNREFINED AND PRIMITIVE METHODOLOGY
An examination of the methodological development of accounting invites the conclusion that the discipline is unimaginative and unintellectual. It seems to have developed no theory of its own, no philosophy. In fact often it has rejected attempts to do just that. Most of its 'theorists' are apologists for methodologies of other disciplines. If the development of a unique methodological foundation is, as those such as Watts and Zimmerman (1986) imply, the hallmark of a mature, intellectual discipline, accounting fails to qualify. Methodologically, accounting is unrefined, even primitive. Such corrosive criticism may well be inaccurate and premature.
It is intended here to examine the developments in the methodological arguments of accounting theorists of the first part of this century and to make some evaluative assessment as to their degree of philosophical sophistication. This approach is consistent with the belief that epistemological enquiry is a prerequisite for eflFective theory appraisal. It is also consistent with efforts in other disciplines. For example, in economics Blaug (1980) and Boland (1982) have undertaken similar studies. The method of approach is necessarily historical. There have been no previous attempts to evaluate the methodologies employed by the authors considered in this study.
Littleton (1933) remains one of the few to have examined the methodologies of past accounting theorists. However, his analysis covered the period to 1900. Those discussed here have made their contributions to the twentieth century accounting literature. Chatfield (1974) has commented on many of the writers discussed here but gave otily a superficial description of their work rather than an in-depth epistemological analysis. But not all twentieth century theories of accounting are included in the analysis.
The primary concern is with theories oi financial accounting written in the English
M. J. R. GAFFKIN is Senior Lecturer in Accounting, The University of Sydney.
language. Therefore, methodologies to develop systems of 'internal', financial reporting
("management' accounting) models are, generally, not considered. Nor is the possibly interesting and methodologically useful work of numerous accounting theorists in
Germany, Italy and other European countries and elsewhere writing in their native languages. Some (e.g., Mattessich, 1984) might argue that such mild restrictions inhibit the analysis. A counter-argument is that most major theoretical accounting expositions in English have been directed to external financial reporting and their methodologies derived from other disciplines rather than those writing in other countries.
Nevertheless, the discussion has been so restricted primarily to make the analysis manageable, more precise and, therefore, more effective.
In attempting to categorize the methodologies employed by the early accounting theorists, the analysis provided by two economists, Wilber and Harrison (1978) may well prove useful. They refer to it as 'pattern modeling' and in this paper some of its major features are described in order to determine whether 'pattern modeling' is an appropriate classification of the methodology employed by