AN INVESTIGATION OF THE APPLICABILITY or
METHODS OF LOGIC TO THE ANALYSIS OF, ‘
MEG-OWN PROBLEMS

Thesis for the‘Degmg of Ph. 'D.
MICHIGAN STATE UNIVERSITY“ 7
Geraldine F, Dominiak I
1-966

[J LIBRARY ‘9,

l

' O

Michigan but 8 ’

5.1 University

SIS

1'1-..

This is to certifg that the

thesis entitled

AN INVESTIGATION OF THE APPLICABILITY OF
METHODS OF LOGIC TO THE ANALYSIS
OF ACCOUNTING PROBLEMS

presented by

 

Geraldine F. Dominiak

has been accepted towards fulfillment
of the requirements for

Ph 2D , degree in _Bnainess_Administration

WME. fleC/Zr

Major professor

 

Date _ _A__g2§£_1._].9_6§_u

0-169

 

/

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘L
If:

 

 

 

 

ABSTRACT

 

AN INVESTIGATION OF TIE Al PLICABILITY OF
METHODS OF LOGIC TO THE A\ALYSIS
OF ACCOUNTID‘ l‘u PROBLEMS

by Geraldine F. Dominiak

This study was undertaken because of the intense cur-
rent interest in research into the foundations of the account-
ing process. The use of ”logic” has been suggested for
investigations of all kinds of things in accounting—~concepts,
principles, postulates, etc.~-predicting that, with the help
of logical techniques, a general theory of accounting will

be found which would present an interrelated combination

 

of these categories in its structure. Such a theory would
presumably Show what should _b_e_ done in accounting and bring
an end to today's disputes, while at the same time providing
a basis for resolving tomorrow's problems. Current works

by several researchers have been leralded as evidence that
logical studies have been made and can produce such a theory.
The overall purpose of this study was to critically examine
these suggestions, predic toi ns and presumptions regarding

the utilization of logical methods in accounting, in the

plight of the nature of logic and logical methods and the

state of accounting argumentatiot .
The simple suggestion for "logic" in accounting being

far from precise, the early CL9 apters of the study are devoted

i?

to‘showing what logic is avai able by examining the nature

 

.. 1....
..‘ livid“

 

 

I
-~.. .‘
.LI

2:-.3

"' yum
.. “w.

 

 

 

 

‘-. - I
Mr.
I‘M;

em.
“K

 

 

 

 

.C

0.1.

1 o—‘II-

. 2,... .1—7.‘ L - ,- " span-2 -- A -
logic and LS meLLLOL-zs. 1-2.2:. .--...‘L-'-t...«.‘-:_..LOLiS afiu 13.. Obi-Gina" L..'..C

aspects of each method are particularly explored, apart
from specific applicati 101“ s
to possible accounting ut‘ rzatrs .

Since the selection 0
particular case should be dictated by the nature of

0

being investigated, terrain olorical inconsistencies occurri-
in accounting literature were found to seriously hamper th
selection process. Cor seque ntly, a tert inology proposal is
made which allows a distinction between ”theory," “concept,”
'postulate,” "principle,“ and “convention” in such a way

that, though all reta ani applicability generally coinc iden“

I
h—

with current usage, yet specific logical interrelations sir
are established. Methods of logic are suggested for inves-
tigations related to matters in each of the categories

defined. Inductive and statistical generalization nd

9)

analogy. are appropriate methocs for investigations of postu-

J.

lates and concept definitiors. Princi‘ les, as theorems of a

(D

particular theory (informal ax’omatin syst m), result directly

from applications of the deductive rules of inference in that

theory; practices and conventions are not generally subject

up

to investigation by methods of logic. There is still a

CL
place for creativity, ingenuity and inspiration, the accountant
continuing to be confronted with matters requiring exercise

of his judgment .

With explicit meanings attached to these terms, and

uses for logical methods suggested, it was possible to

 

 

 

12213 lie

 

'9. a.
.v -D
c 1-:
_ DV

I. ,‘ -d
“W; ....

 

 

 

 

 

 

 

I
w;
I I

N

 

 

 

 

. Geraldine F. Dominiak

evaluate the predictions concerning what the construction of

a general theory might do for accountants and to assess the
advisability of seeking such a theory.

It is shown that the benefits expected to be gained

should a general theory be constructed utilizing the suggested
logical methods have been greatly OVerestimated. New problems
will arise, and several difficult problems of today will con-
tinue to be problems. .The following are some examples dis—

cussed in the study.

T

D

l. A general theory snowing what should 32g done cannot
be expected, for no method of logic will allow the inferring
of what should be solely or. the basis of-what is. Hence, it
will be necessary to have previously arrived at a notion of
what should be. this decision being Lelated to the proposed
objective. Whether the theory is right or wrong, then, is a
relative matter, depending entirely on the extent to which
it achieves what it was intended to achieve.

2. ._ Construction of "a" general theory does not rule
_out alternative theories, for no rules of logic limit the
number-of theories possible on a given subject. The accept-
ance of a single theory will depend on acceptance of its
pOStulates. Questions of truth or falsity are largely mean-
ingless. Further, since accounting practices involve the
application of'definitions of basic concepts to individual
business facts, elimination of alternative practices depends
uPon acceptance of a single set of definitions of the con-
cepts of the general theory. But acceptance is not a problem

01‘ legic alone. Consequently. today's questions concerning

 

 

 

 

 

 

.
.15“

N“; n-
""V~.V..

 

 

 

 

 

 

 

w

Geraldine F. Dominiak

general acceptance will not disappear with the advent of a
general theory.

3. Acceptance of a single general theory would not
destroy the need for special and less abstract theories in-
volving additional specific postulates and definitions
designed to cover specific problems.

4. After one theory was accepted, new business facts
might necessitate re-examination of previously accepted
postulates and definitions, bringing new conflicts which
cannot be resolved with logical methods alone.

Devoting our primary efforts at this time to the con-

 

struction of the elusive general theory is not advocated,
on the basis of'experiences in other disciplines, in view of
predictable resistance tovthe use of new and complex argu-
ment presentation methods, and for reasons of progress and
practicality.. Because of this conclusion, the final section
of the study includes an analysis of the several studies
pointed to as evidence that a general theory with the help
0? logic is a reality or near reality. Critical examination
of these studies:L fails to give support even to a contention
that the use of logical methods has, or soon can, produce a
general theory in the sense in which the phrase is currently
used.

As evidenced by the several common misunderstandings
pointed out in the study, most accountants are unfamiliar
with many of the most fundamental aSpects of logical infer-

ences. At this stage in the development of formalization

 

 

 

‘ v .
,, . .pmf]

., amm—

 

 

 

 

 

of accounting discussior -,s use of any m----‘

strongly recommended in rather simple, limited, perhaps even

noncontroversial matters.

 

1Studies selected were: narvin Lee Cai lson, "Accounting
Theory as a Logical System,“ 'npublished"J i .D. dissertation
(Commerce, The University of 7H‘sccnsin1964); R. J. Char‘cer
T__9__wards a General Theory of I‘ccomitl’v (Mel bourne, Austr‘lia
The Australian Society of loccaot nts, 1902); .Iaurice lioonitz,
The Basic Postulates of Accounting (No 1' York: American In-
stitute of Certified Public Accountants. 1961 I-Ieurice
Moonitz and Richard T. Sprouse, A. {Foucatw ~‘ of Broad
Accounting Principles for 18153111119353 Enter ori s (16/! i’ork:
American Institute of Certified Public Accou; ants, 1962‘);
Richard Mattessich, Accounting and ./-_r.alytice.l Evie thods (F ome-
wood, Illinois: Richard D. Irwin, Inc., 19614); and S udy
Group at the University of Illinois, A Statement of Basic
Accounting Postulates and Principles (Urbana, Illinois: Cen
ter for International Education and Research in Accounting,

1961+) .

C‘
“9
.

 

.,

  

 

 

 

 

AN INVESTIGATION OF THE APPLICABILITY OF
METHODS OF LOGIC TO THE ANALYSIS

OF ACCOUNTING PROBLEMS

By

Geraldine FEGDominiak

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Accounting and Financial Administration

1966

 

 

 

W
“0

© Copyright by
GERALDINE FLORENCE DOMINIAK

1967

 

 

 

 

   

 

 

 

  

 

 

 

 

 

 

W

 

 

A CKNOWLEDGMENTS

I am grateful to numerous individuals and organi—
zations who have helped to make this thesis possible. My

appreciation is expressed to the members of the dissertation

committee——Professor Herbert E. Miller, Chairman, Professor

Charles Lawrence, and Professor Harold Wein-—not only for
their guidance during the progress of the thesis but also
for their stimulating comments regarding future undertakings.
I am especially indebted to Professor Lawrence, who, in
addition to providing considerable assistance with the
thesis, has, throughout my entire program, enthusiastically
encouraged my interest in the subject of logic as related
to accounting, auditing and teaching. A

I wish to thank Professor Marvin A. Carlson who so
generously supplied me with a copy of his doctoral disser-
tation and offered encouragement and thoughtful comments.
To Professor James Don Edwards, Chairman of the Depart-
ment of Accounting and Financial Administration, I offer my

thanks for his constant encouragement and support during my

stay at Michigan State University. Special thanks are due

to Professor Richard E. Czarnecki for his encouragement,

understanding and friendship during all phases of my program.

ii

 

 

 

 

n

   

 

...n,., r

.wi‘. v

 

 

”nu:
. _\
~- ...t.

n,‘

   

 

.2?

 

 

 

 

 

 

 

I wish to express my appreciation to the Detroit Office

of Lybrand, Ross Bros. 8c Montgomery, for the assistance in
the final preparation of this thesis, with particular thanks

to Richard Berkau for his extensive efforts on my behalf. I

also appreciate the financial support I received during part

of my program under the fellowship program of the Ford Foun—

dation. The conclusions, opinions and other statements in

this thesis are, however, those of the author.
There are no words which could possibly express my
gratitude to my parents for their encouragement, faith,

understanding and sacrifices through every step of my aca-

demic career. But most importantly I thank them for teaching

me to respect knowledge and for instilling in me a desire to

learn.

 

 

 

 

 

 

1T.

 

TABLE OF CONTENTS

ACKNOWLEDGMENTS .
LIST OF ILLUSTRATIONS
LIST OF APPENDICES
Chapter
I. INTRODUCTION TO THE PROBLEM .

The Search for Principles— Problems

 

Discovered . .
The Methodology Problem . . . - 12
A Hierarchy of Accounting Statements . . . 16
Current Studies of Accounting Problems . . 1
I
18
II. THENATUREOFLOGIC. . . . . . . . . . . ..
20
Righting Wrong Impressions . . . . . . 29
Basis for Examination . . . . . . . . . . .
8 .
III. DEDUCTION . . . . . . . . . . . . . . . . . . 3
8 ‘
Methods in General . . . . . . . . . . 34
Informal Axiomatic Method . . . . . 79
Conclusions . . . . . . . . . . . . . . . .
. . . 83
IV. INDUCTION a o 3 a a a a o a a a O l ‘ ‘
88
Argument Based upon Self— evidence
Argument Based upon the Testimony ' 91
of Others . . . . . . : . 96
Argument Based upon Experience .f r -
ASpects of Induction Significant o . . 117

Argumentation in Accounting Theory

iv

 

 
 

 

 

 

 

 

 

  

TABLE OF CONTENTS-—Continued

Chapter

v; TERMINOLOGICAL DIFFICULTIES AND A
”LOGICAL” SUGGESTION . . . . . .

A Terminology Problem

Terms Distinguished-An
Interrelated Structure

Analysis of the Proposal . . .

Comparison with Other Proposals

VI. CURRENT ”LOGICAL” STUDIES EXAMINED .

Accounting Research Studies
Numbers 1 and 3 . . . . . . .
Postulates and Principles of the
Illinois Study Group -
Towards a General Theory . . . .
Accounting and Analytical Methods

Accounting Theory as a Logical System

General Conclusions
VII. RESUME].

”Logic” Clarified . . . . . . .
"Accounting” Clarified, A Proposal
A General Theory . . . . . . . . .
The Beginning . .

APPENDIX .

BIBLIOGRAPHY ,

Page

127
127

131
135
157

I69

181
192
197
204
209

213
214
218
220
226

229

 

 

 

LIST OF ILLUSTRATIONS

Illustration

1. Structure of a Theory .

vi

Page
13A

 

 

 

 

Ah 1 . A V Ad

 

 

 

Appendix
A.

 

 

 

LIST OF APPENDICES

Postulates and Principles of the
Accounting Research Studies

Postulates and Principles of the
Illinois Study Group

The Basic Assumptions of Professor
Mattessich . . . . . . . .

The Basic Assumptions of Professor
Carlson . . . .

Page

229

232

 

 

 

 

In m

‘35 ESSA ‘

 

 

 

 

 

 

 

 

 

 

 

CHAPTER I
INTRODUCTION TO THE PROBLEM

In the accounting literature of the past ten years we
have seen what might be termed a ”return to nature" movement,
wherein academicians and practitioners have called for and
undertaken investigation and research into the very foun—
dations of the accounting process. The reason most repeated by
practitioners and well-informed people outside the pro-
fession for this return to‘fundamentals is dissatisfaction
with the present acceptability in accounting practice of two
or more alternative procedures which produce substantially
different results in the financial statements, an important
output of the accounting process. In a given firm, for a
given transaction, event or situation, alternative procedures
are available. For firms in the same industry, accounting
treatments differ for seemingly similar transactions, events
or situations.

Concerned accountants recognized that published financial
Statements, from which they obtained their evidence of the
current problems in accounting, were accompanied by an
auditor's report stating that such statements were in accor-
dance with "generally accepted accounting principles. ” If
these reports were assumed to be made in good faith, the
auditors must have believed that the accounting practices
adopted m the principles mentioned, or were consistent

I

 

 

 

 

 

 

. u
“1
."L': «5‘
n .n _;
.; -- ,7
‘ ~ t.‘
2".»
u...
Yum-r

 

 

 

 

 

r.
t":
..L

 

'u

 

 

 

 

with some unmentioned accounting "principles. ” But, in

either case, it was the principles referred to which seemed
to be in need of careful investigation for the source of the
practice multiplicity problem.

The ideal solution to the question, of course, consisted
in comparing the troublesome practices with an exhaustive
list of ”generally accepted accounting principles. " However,
review of any number of accounting texts and reference works
made it plain that no such list actually existed. That is,
no such list existed prior to the publication of Accounting

Research Study No. 71, and that study does not, as yet, have

 

complete regulative authority. The problem of what caused or
justified the various practices remained unsolved.

The possibility that alternative practices FEE alterna-
tive principles and equally accepted generally was swiftly
eliminated. First, new practices were being adopted every
year, and it was difficult to conceive that "generally
accepted accounting principles” could be multiplying annually.
It seemed equally inconceivable that new practice-principles,
with the stability which the latter word implied, could
receive general acceptance with such rapidity. Literature
Spoke of both practices and principles, and the latter were
presumed to be of a more theoretical or at least a more
general character than the former. Moreover, if practices

were principles, it would seem to follow that the principles

 

 

 

. lPaul Grady, Inventory of Generally Accepted Accounting
Prlnciples for Business Enterprises (New York: American
Institute of Certified Public Accountants, 1965).

 

 

 

 

 

 

 

 

“2r:
........,

 

 

 

 

 

 

 

 

 

 

 

-.:
:
4x
' ».\

 

 

 

were inconsistent. It was seen that there were cases where

substantially different results could be reported in terms
of the income of one entity by selecting one or another of
the alternative practices. If it is granted that there is
such a thing as income, then, while several numbers (stated
in dollars, common dollars, units of product, or whatever)
might all be approximations of that income, it was not
possible that (I) all could be income, or that (2) none of
the numbers of the set possessed the property of being a
closer approximation. Thus, if practices were principles,
the principles referred to in reports were immediately
lacking in an important feature and could be called incon—
sistent. Practices so conceived could not be the ”principles”
so carefully considered in determining the appropriateness
of accounting information. It must be the case that prac—
tices are not principles, and the latter are to be found
elsewhere.

The Search for Principles—Problems Uncovered

 

Alternative practices seemed to receive justification

.in accounting literature. Professional journals published

articles about specific accounting procedures which articles

’ made use of many general statements concerning accounting.

Some honest men were convinced by these articles and alter—
native practices were approved as could be seen by the
reports of the auditors. It was not unreasonable to assume
that if the practices themselves were not principles, the

Justificatory articles must have appealed to whatever the

 

 

 

 

 

 

 

 

 

 

 

an.

 

 

 

We

 

 

u,

'.

 

 

 

 

 

prflmiples were and shown that the practices were consistent
wiflithose principles. The general statements concerning
accmnwing which appeared as parts of the justifications
eiflwr were or might lead to the principles being adhered to
acmnding to the auditor's report—or, it was possible that
muleiflm arguments might appear convincing, the reasoning
flwolved therein was somehow in error.

A brief examination of the literature justifying
Mtemmtive practices produced several interesting obser—
vations.

l) Justifications of practices in the same
situations sometimes turned partially on disagreements
regarding common terminology. For example, advocates
of flow—through investment credit treatment often deny
the expense nature of income taxes and the resulting
applicability of matching of costs and revenues.

2) Justifications of practices in the same
situations did not always refer to the same general
statements about accounting. For example, a statement
of the necessity of conservatism and objectivity
generally appears in support of some practices relating
to the immediate write-off of certain expenditures,
while references are made to matching of costs and
revenues and fairness of financial position presen-
tation in support of deferral practices.

3) Justifications of practices in the same

situations sometimes inferred a hierarchy of general

 

 

 

S‘C‘f."
“Anni.
1

 

 

 

”w.

 

 

 

 

 

 

 

 

 

,statements about accounting. For example, the necessity

of adherence to acquisition costs was considered subor—
dinate to the need to provide for foreseeable losses

in the reporting of inventory amounts at the lower of
cost or market; and, matching of costs with revenues,
among other things, seemed to justify the departure from
the continued historical cost reporting of plant,
property and equipment investments.

A) The general statements concerning accounting
which were appealed to in the justificatory articles
were neither succinctly stated nor uniquely titled or
worded. No simple statements could be recognized as an
exact formulation of one of the well—known accounting
”principles.” Title confusion is exemplified when
references are made to accounting adherence to _
historical costs as the cost concept or the cost
principle or the cost convention. Arguments sometimes
mentioned the entity theory, the entity concept or the
entity postulate.

5) -Various discussions of alternative practices,
appearing in the form of prose arguments which purported
to show that the practices followed readily from certain
general statements about accounting, were, while con—
vincing to some, lacking in the rigor or certainty which
accountants attributed to mathematical and scientific
argumentation.

Disagreements of the type noted in (1) above indicate

 

 

absem

tar-hole;

 

 

 

 

 

 

 

 

the absence of a uniform interpretation of basic accounting

terminology. Arguments of the type noted in (2) seem to
indicate that there might be some inconsistency between. the

' general statements about accounting. which, when considered
alone, are quite acceptable. The third observation introduces

the possibility that perhaps some general accounting state-

 

ments are more general than others-—an idea of basic and
subsidiary principles. The situation found in cases of
type (A) both reinforced the possibility of a hierarchy, or
series of levels of accounting generalizations, and at the
same time pointed out that terminology disagreements among
accountants extended beyond basic terms of the discipline

itself. Clearly, attempts to identify generally accepted

 

' revealed a Pandora's box of related

accounting ”principles'
problems.

The final observation is, perhaps, the most far-reaching
in its implications, for it relates not to specific accounting
problems but to the way in which all accounting problems
are approached—~the method of investigation, argumentation
and justification. If problems exist in accounting, as the
first four observations indicate they do, attempts to solve
them must adopt some method of investigation and presen-
tation of results which will be accepted as appropriate and
convincing so as not to create further problems. If past
investigations of accounting problems did not make impossible
the inconsistencies found in present day accounting, the

methods used in developing present day accounting must not

 
 
 
 
 
 
 
 
 

 

 

 

 

 

 

have been altogether appropriate. Along with the problems
pertaining to the subject matter of accounting, it appeared
that the topic of investigative methodology deserved consi—

deration.

The Methodology Problem

 

Current literature contains numerous articles on
research methodology.1 There are references to ”the scien—
tific method,” ”rigorous reasoning,” statistical research,
empirical studies, etc. The common thread which runs
through these writings of both accounting practitioners
and accounting academicians relates to the use of ”logical"
methods. Professor Storey said, in his recent monograph,
The Search for Accounting Principles—Today‘s Problems in

PersQective:

The decision to rely on "distillation of practice”
in the early days for the development of principles
was probably a sound one. Accounting has now
outgrown the stage in which a process of this type
is sufficient. . . . Accounting has developed about

 

 

 

 

H 1For example, Norton M. Bedford and Nicholas Dopuch,
Research Methodology and Accounting Theory—Another
Perspective,” The Accounting Review, XXXVI (April, 1961),

pp. 351-361; Carl Thomas Devine, ”Research Methodology and
Accounting Theory Formation, ” The Accounting Review, XXXV
(July, 1960), pp. 387—399; Myron J. Gordon, ”Scope and Method
Of Theory and Research in the Measurement of Income and
Wealth, " The Accounting Review, XXXV (October, 1960),

pp, 6014—618; Brother LaSalle, ”Basic Research in Accounting, ”
The Accountin Review, XXXIV (October, 1959), pp. 603—608;
Olin Park, Thought Processes in Creative ACCOunting,” The
Accounting Review, XXXIII (July, 1958), pp. 11141—4114; Mil_t6n
H. Spencer, Axiomatic Method and Accounting 303181109,” Egg
Accounting Review, XXXVIII (April, 1963), ML 310-315;

1- Kleerekoper, The Economic Approach to Accounting,"

Journal of Accountancy, CXV (March, 1963), pp. 36-40.

 

 

as far as it can without more reliance on the use of
logic in the formulation of principles, and logical

 

 

methods will ecome increasingly important in future
developments. (Underscore added.)

A strong display of confidence in logical methods and their
applicability to accounting is evidenced in this comment of
Professor Spencer: ”Accounting, of course, may also be

viewed as a logical 'game,’ and hence an axiomatic structuring
2

 

is certainly possible. ” Confidence and hopefulness in the
beneficial use of logic is expressed by John W. Queenan,
former president of the American Institute of Certified Public

Accountants, in his comment: "The process of developing the

postulates and principles envisaged by the research program

 

is one in which abstract reasoning and logic can play an

important part. ”3
The suggestion that "logical techniques” be utilized

by accountants is not entirely new. Professor Leo A. Schmidt

wrote, in the Journal of Accountancy in 1949, "What we

really need in accounting is the practical application of

 

1Reed K. Storey, The Search for Accounting Principles—
Today's Problems in Perspective (New York: American Institute
of Certified Public Accountants, 196A), p. 62.

2Spencer, 0 . cit., p. 316.

3John W. Queenan, "Postulates: Their Place in Accounting
Research, " Journal of Accountancy, CXIV (August, 1962), p. 32.

 

 

:he very

 

 

 

 

 

 

 

the very simplest principles of logic.”1 Professor Littleton

suggested, in 1939, that accountants engage in ”syllogistic
mmlysis" which permits ”conclusions to be deduced from
perceiving relationships between obviously true propositions”
or puts ”principles to the test of being logically linked
to premises. . . ."2 But, Professor Storey's comments indicate
that accountants have not usually adopted the techniques of
loSic, or at least not the proper ones,in developing present
day accounting. The continued pleas for ”more logic” in
amounting seem to bear out Professor Storey's conclusion.

The suggestions for ”more logic” have related to the
investigations of all kinds of things in accounting:
principles, postulates, axioms, practices, concepts,
theories, problems. There has been, however, no precise
exPlanation of exactly what this logic is that is supposed
to be used nor in which accounting investigations. Some
Of the recent research in accounting fundamentals has
apparently been considered more ”logical” than that of the
past Professor Storey remarks:

The work of the American Accounting Association,
that of several individuals, and the early Accountlng
Research Studies are ample evidence that logical

Studies Of accounting are not only possible but can
be lmpOrtant contributions.

__~“___________Ai

l -

Leo A. Schmidt, ”Practical Uses of the Dev10e of
Ammal LOEiC in Accountants‘ Daily Work,” Journal Of
LEEEHEEEE_, LXXXVIII (November, 1949): p-_§787—n———

. Illi-
. . Littleton Essays on Accountancy (Urbana, _
3018: University of Illinois PreSS, 1951 9 pi 374' This
Olufe 13 a collection of comments by A- 0' Littleton l
earl1er years.

 

 

3 _
Storey, 0p. cit., p. 62.

 

 

 

 

A}. Tm

 

 

 

 

:mu-l
-\

—In.u-‘

 

 

 

Tm;

~n...1

 

 

 

 

 

 

 

1,, __

lO

Mom5authors consider the current works of Professor Richard I
MMfiessichl and Professor R. J. Chambers2 to be more

rigorous than the accounting investigations of the past. But
all these comments about rigor and logic exhibit a certain
ammnw of vagueness. It is no more sufficient or appropriate
U>suggest a "logical analysis” or "deductive" treatment of
accmufiing problems, or theories, or whatever, without
Ewestigating whether such problems are capable of being
treated in this manner than to discuss the depreciation of an
asset without investigating the nature of the asset. And,
carrying the analogy further, the selection of a particular
nethod of depreciation is made only after carefully reviewing
thenmthods available and the characteristics of each, and
trying to match these characteristics to the nature of the

depreciation problem at hand.

 

>Some of the suggestions regarding the use of logical
techniques in the discussing of accounting problems have
Umluded reference to specific subject areas of logic such
as'fleduction,” ”induction” and ”axiomatization.” However,

Hmse terms, when used, convey what might be termed a

 

1For example, "Towards a General and Axiomatic Foun—
dation of Accountancy,” Accounting Research, VIII (October,
1957), pp. 328—355, and Accounting and Analytical Methods
lbmewood, Illinois: Richard D. Irwin, Inc., 1964).

2For example, ”Blueprint for a Theory of Accounting,"
Accounting Research, VI (January, 1955), pp. 17—25, and
Towards a General Theory of Accounting (Melbourne, Australia:
The Australian Society of Accountants, 1962).

Vi

«.1qu "w
.I'uu .. ‘ J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

r__7

laymmfls understanding of the techniques suggested. As a A
remflt, the advantages (and disadvantages) of using ”logic”

lmve sometimes been, to some extent, misrepresented. Many

wnds from other fields of knowledge or activity have found
tmfir way into the everyday vocabulary of the accountant;

accmnmants have appropriated words from economics, mathe-

 

mMfics, statistics, banking, law, engineering, etc. These

wnds may have had quite specific meanings in their original

fields, and, in most cases, such meanings were preserved

whmiadopted into the language of the accountant. Just as

Hm accountant has adopted the terms used in other fields, so
hastw selected, for his own use, many of the tools of other
dimflplines. Auditors are utilizing the statistician's

tools3nmnagement accountants are investigating new ways to

 

mmlyiflm tools of higher mathematics. Now accountants are

adwmating the employment of the terms and tools of logic.

The profession has in the past consisted of some
able men armed with a kit of practical rules; and
conceivably it could move in this more or less
haphazard manner to meet the future needs of society,
but the systematic structure of logic will certainly
become an increasingly important influence. . .
Such stock intellectual cliches as "it is logical to
assume,” etc. are now on occasion examined to see
whether they are or are not in fact ”logical.”

Bum if logic is to become an implement in the accountant's
toolldt, it is the logician's understanding of terms and
mmflwds that the accountant must know—just as he is now

cmMng to realize that it is the statistician's and not

1
Devine, op. cit., p. 389.

 

 

 

 

 

.,..,.-
.. 1....1...

 

 

 

 

 

 

 

 

"4w

 

 

 

r__7

l2

thelayman's concept of probability that is to be under- A
stood and utilized. ;
It is the first objective of this paper to give an
emflanation of what ”logic" is available to the
accmnmant-investigator should he desire to incorporate

loguml tools into his investigation of problems in account-

 

ing. The methods of analysis and argumentation previously
{med hiaccounting literature will be compared with the
avaflable logical methods, and it will be demonstrated that

Hm advocates of logical techniques in accounting must be

 

mumorting the use of formal logic. The suggested advantages
ofadopting formal methods will be presented and evaluated.
hithis connection, some of the serious difficulties
awountered by the researchers in other fields when

attamming to adopt the methods of the logician will be pre—
samed and parallels with accounting will be drawn which
shmfld suggest the parameters of application of logical

teduflques to accounting investigations.

A Hierarchy of Accounting Statements
The possibility that there are various levels of
accmnming generalizations, as mentioned earlier, might be
cmmidered further justification for finding out in which
accmnfiing investigations logical techniques would prove

fmutful investigative tools. The very term which accountants

 

 

 

 

. J .
‘v :d v
..v- x

 

.. . has

 

 

ZMMV
~-_...c1\

 

 

an.
- x-w

 

i
A.
“A”

 

 

 

 

 

  

 

 

 

13

use to identify certain generalizations, i.e., ”principles,"
suggests an analogy with the sciences and the existence of

different levels of abstraction or importance. If there are,

in fact, principles, their source may be more general

statements to which the principles are related in some

discernable way. These more general statements might be

called postulates or assumptions, again, not unlike the
terminology in the sciences.
This position was firmly adopted in connection with

the present research programs of the AICPA. In the report

of the special committee on research programs of the Council

of that organization is found the statement: "Postulates are

few in number and are the basic assumptions on which

principles rest. ”1 The titles of Accounting Research

Studies Numbers 1 and 3,2 are prime examples of the adoption

of this position. Very early in modern accounting literature

we find Professor Paton referring to certain general state—
3

ments of accounting as basic ”postulates” of accounting.

Some type of structured whole consisting of assumptions,

1Journal of Accountancy, CX (December, 1958), p. 63.

2Maurice Moonitz, The Basic Postulates of Accounting
(New York, 1962), Study No. l, and Maurice Moonitz and Robert
Sprouse, A Tentative Set of Broad Accountig Principles for

Business Enterprises (New York, 1962), Study No. 3, both
Dahlished by the American Institute of Certified Public

Accountants.
3William A. Paton, Accounting Theory (Ann Arbor,
Michigan, 1922).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.21

 

 

.. in-

prnuflples and practices is implicit in the writings of

Pmflessor Littleton when he states:

Fundamental truths of accounting (principles) may
either be generalized out of practical experience,
or deduced from stated premises which are accepted as

true in themselves.
That there may be several levels of accounting
may help to

 

gamrafizations, and an investigation of them,

mwwer other questions now being posed by the members of the

accmnming profession. There is a growing unrest concerning

ceflmin individual general statements about accounting which
In

 

appear to have strongly influenced accounting practice.

paflflcular, the recording and reporting of historical cost

ammnms almost exclusively has received severe criticism.

Em problems related to this criticism are associated with

 

Hm problem of variety in practice, for the question is

raised as to the status of such general statements. How

Mmortant is cost recording to the smooth functioning of

Wm accounting process? What other areas of accounting are

related to this statement—what areas would be affected by

Hm discontinuance of carrying on the accounting process

filaccordance with historical costs? What sort of research,

argmwnt or evidence would, or could produce a change in

HUS part of the accounting process? If the levels of
accmnwing generalizations could be distinguished and

eXiflained, some of these questions might also be answered.

1 .
,Littleton, op. cit., p. 391»

 

 

 

 

new
.. w.

 

 

 

 

 

 

 

 

 

 

. x.
w:

 

 

 

 

 

 

 

 

 

 

15

A very serious impediment to an explicit and complete
Mntement of a hierarchy of accounting generalizations has
bemithe terminological inconsistency of accounting authors.
Wmt were considered by Professor Paton and some others as

posmflates were designated by others as concepts, by some

asprinciples, by others as theories, and by still others
To give just a few examples, Professor Windal

as practices.
Professor Paton speaks

mfltes of the realization ”concept,”

ofa realization ”postulate” and the Study Group of the

Unversity of Illinois proposes some realization "prin-
3

cnfles."2 Dr. Moonitz suggests entity ”postulates,”

Dr.Iprig discusses the problems of the entity ”theory,"

A

andlhn Li explores the effects of adopting the entity
l'concept.”5 Whether the accountant's interest in ”con—
servatism" is called a ”principle” or a ”convention” de—

pmnm on the individual author. All these terms——concept,

lFloyd W. Windal, ”Legal Background of the Accounting
Cmmept of Realization,” The Accounting Review, XXXVIII

(Jmnmry, 1963), pp. 29—36.

2A Statement of Basic Accounting Postulates and
Prhufiples (Urbana, Illinois: Center for International
Emumtion and Research in Accounting, 196 .

3Moonitz, Basic Postulates.

AArthur N. Lorig, ”Some Basic Concepts of Accounting
and Their Implications, ” The Accounting Review, XXXIX
(July, 1964), pp. 563—573.

5David H. Li, ”Alternative Accounting Procedures and
the Entity Concept,” The Accounting Review, XXXVIII (January,
1963L pp- 52-55, andnghe Funds Statement Under the Entity
Concept, ’ The Accounting Review, XXXVIII (October, 1963),

Pp. 771-775.

 

 

 

 

 

 

 

. pnhmiple, postulate, theory, practice, even convention-—
tmve been used so often there seems no doubt that accountants
consider them worthy of retaining in accounting terminology.
Dififinctions between them have been implicit in many
matements about accounting, including those of the AICPA
commttee and Professor Littleton, quoted earlier. If these
terms are to be retained, it is these which should be
stnumured into a hierarchy in which accounting ideas and
matements can be placed--a structure such that individual
1amwn methods of logic can be seen to be applicable at
specific levels and the interrelationships and interdepen—
dency can be established.

It is the second objective of this paper to suggest a
stmnmure of accounting knowledge. The structure will
anew for the distinction of theories, concepts, postulates,
prhuflples, practices and conventions. The proposed
relationships within the structure will be such that the
logical tools for investigation of the levels of the
stnkmure can be suggested, wherever some tool is appro-
priate. The problems which remain after the applications of
logical techniques will be explained, and suggestions will
benmde as to those areas of accounting interest in which

finmal logic seems completely inappropriate.

Current Studies of Accounting Problems
Since the accounting profession has expressed its con—

cemiover the state of the body of accounting knowledge, many

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.1.

 

 

 

l7

mflflications have been forthcoming relating to fundamental
accmnwing matters. Some of these investigations, it has
hemisuggested, are the result of response to pleas for
logmzin accounting and represent instances of logical
meflxfls being utilized to obtain more rigor in analysis and
reasoning. Investigations generally represent full scale
attacks on the whole of accounting. Especially well-known
mmmg the recent works are the following:
1) American Institute of Certified Public
Accountants, Accounting Research Studies
Numbers 1 and 3

2) Towards a General Theory of Accounting,
by R. J. Chambers

3) A Statement of Basic Accounting Postulates
and Principles, Study Group of the Univer—
sity of Illinois

4) Accounting and Analytical Methods, by
Richard Mattessich

Miequally ambitious project was undertaken by Dr. Marvin
Cmflson, in his "Accounting Theory as a Logical System," an
as yet unpublished doctoral dissertation submitted at the
Whyersity of Wisconsin.

The third objective of this paper is to examine each of
Hm studies mentioned above, identify the logical methods or
techniques utilized, and determine whether the methods used
cohwide with those proposed as most advantageous for that
area of accounting (concept, postulate, etc.) in the

suggested interrelated structure of accounting knowledge.

 

 

‘ r

0' 2

 

 

 

 

 

 

 

 

 

 

 

 

 

CHAPTER II
THE NATURE OF LOGIC

In general usage, the words ”logic" and ”logical" are

 

takaias synonymous with ”reasonable," with the ”logical"
being the ”reasonable" and the "illogical," the "unreason-

ableN' The terms are often, but not always used to

dnracterize arguments, i.e.,justifications for the holding
This usage is basically in

for the

of some position or other.
kemfing with the meaning intended by the logician;
logician is concerned, not with the actual physiological or

psmflwlogical processes of reasoning but with the accept—
More

 

ability of the final, completed process as a whole.

emflmtically, only this usage coincides with the logician's.

Acceptability may be characterized by an affirmative answer

to the question: In an argument, does the conclusion

(position held) really follow from the premissesl (as-

mmmtions, stated or otherwise)?

Whereas the connection between premisses and
conclusions is thus grounded in logic, ordinarily
the premisses and conclusions themselves are not;
and herein precisely lies thg application of logic

to fields other than itself.

1This spelling follows the use of C. S. Peirce, A.
Gmnmh and others in referring to the logical term and
distinguishing it from ”premise" in other senses in—
duding the legal application of the plural.
2Willard Van Orman Quine, Methods of Logic, Revised
(New York: Holt, Rinehart and Winston, 1959), p. xiv.

18

 

 

 

19

 

Thelogician is not concerned with the truth or falsity,
acceptability or reality of the premisses or conclusions,

mm with the form of the argument comprised of premisses

and conclusions.

The distinction between correct and incorrect
reasoning is the central problem with which logic
deals. The logician's methods and techniques have
been developed primarily for the purpose of making
this distinction clear. The logician is interested
in all reasoning, regardless of its subjept matter,
but only from this special point of view.

The logician will be interested then, in all areas of

knowledge where reasoning is purported to play a part in the

hwestigations concerning that knowledge. Insofar as

accmnming discussions rely on reasoning (argumentation) to

mmport a stated position, the logician does have an interest

hithem. But, as Copi says, "only from this special point
ofxflew." The logician has developed certain tools with

much some problems can be attacked. Students of other

dimflplines who would use them are admonished:

To appreciate the value of logical tools, it is
important to have realistic expectations about their
use. If you expect a hammer to do the job of a
screwdriver you are bound to be disappointed, but
if you understand its function you can see its
usefulness. Logic deals with justification, not

with discovery. Logic provides tools for the
analysis of discourse; such analysis is indigpensable
g

to intelligent expression and understandin .

lIrving M. Copi, Introduction to Logic (New York: The
Malilan Company, 19517: p. 6-

2Wesley C. Salmon, Logic (Foundations of Philosophy
Series; Englewood Cliffs, New Jersey: Prentice-Hall, Inc.,
A.

1963), p. 1

 

 

 

 

 

 

 

.2”: -,‘
.....E w

 

.el

-\’_‘
NH.

 

 

 

 

 

  

 

 

   

 

 

20

A reading of the nontechnical and popular literature
wilogic and the literature of most other subjects, as well,
tends to leave the reader with three somewhat erroneous
impressions.

l. The word ”logic" is often modified by the ad—
jective "formal," giving the distinct impression
that logic may be formal or informal.

2. There is a division of logic into inductive and
deductive which is mutually exclusive and jointly
exhaustive.

3. While many researchers in the sciences using
logical tools may still find themselves having
disagreements, the logicians themselves know
(a) what the tools are, and (b) how they are to
be used, and are in agreement in their knowledge
of these two matters.

The objectives of this chapter will be to correct

ste erroneous impressions and to establish a basis for
examining the characteristics of the methods of logic, pay—

ingtmed to Professor Salmon's warning to "appreciate the

value" of these tools.
Righting Wrong Impressions

Formal and Informal Logic

The logician deals formally with both formal and
hubrmal arguments. Logic is a subject of individual study,
investigating forms of argumentation and codifying those
mfles which allow the recognition of correct and incorrect
arguments. Logic cannot deal informally with arguments, for

it is the essence of logic to deal formally with them. Nor

 

 

 

 

 

21

shmfld informal arguments be construed as arguments which
lnve no form, for logic deals only with forms, and hence
cmfld not be relevant in such a case. In fact, argument
is a term of the logician, and an argument without some
fonnis Simply not an argument at all.

By informal argument is generally meant something quite
like Suppes' description of an informal proof, ”enough of
wiargument is stated to permit anyone conversant with the
subject to follow the line of thought with a relatively
lugh degree of clarity and ease, ." obvious premisses
areleft unstated and only "the essential, unfamiliar,
unobvious steps” are given while omitting the "trivial and
routine inferences."l From this description it is clear
Hat an informal argument should be capable of being trans-
Ibrmed into a formal one. The advantage of formal proof
or argument as opposed to an informal formulation seems
apparent from the use of so many words in the definition
Muse meanings are imprecise, such as ”essential,” ”obvious”
and'Trivial." It is imperative to understand that the
logician dealing with informal arguments will apply the
same rules as when he deals with a formal argument, but in
the former case, these rules must be applied to both the
stated and the unstated premisses and inferences. And since

Hm same rules are to be applied, it is usually more con—

 

lPatrick Suppes, Introduction to Logic (The University
Series in Undergraduate Mathematics; Princeton, New Jersey:
D.Van Nostrand Company, Inc., 1957): p- 122'

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i‘v

 

 

 

 

 

 

22

venient for the logician to transform informal arguments I
into formal ones.
There is another sense in which logicians have used
the term ”informal” in connection with logic. This usage is I
common in discussions of so—called ”informal fallacies,”
but this is not to be understood as some kind of informal
logic. Some arguments appear to possess, by the nature of
the individual words or whole premisses used, qualities
which cannot be attributed to the form of the argument

itself. Yet, such qualities may seem to affect an

 

individual's acceptance of the argument. It is customary
among logicians for the term "fallacy” to be used in
connection with an argument which, although not correct, is
in some manner I'psychologically persuasive. " Examples of
the type of problems involved are the use of words such as

"fair, H H

good" and ”patriotic" which carry with them certain
emotive force that may have an influence on the effective-
ness of an argument in which they appear. But the effective—
ness of an argument is not to be equated with the correctness
of that argument. The feelings which certain words may
produce are not, in any strict sense, logical. Nor are they
a part of the logician's realm of study, but rather that of

the psychologist. However, when the logician investigates

such an argument, he may, and often does, find that the use

 

of certain words may be considered a means of asserting

certain premisses which can, and in fact do, play some part

 

 

 

23

fliestablishing the correctness of that argument. The
statement "Since Jones is a communist he won't make a good
certified public accountant" is a sample of a brief, informal
argmwnt. Perhaps its proponent fully expects the designation
ofJones as a communist to support his conclusion. If so,

the argument is not deductively correct, for nothing has

been said about any relationship between communists and

CPAs. The logician, upon examining the argument, might

suggest that it is a shortening of the more formal argument:

Jones is a communist. (Premiss 1)
No communist is able to report

impartially. (Premiss 2)
Jones is not able to report

impartially. (Intermediate inference)
Good CPAs must be able to report

impartially. (Premiss 3)
Jones cannot be a good CPA. (Conclusion)

This longer argument, with premisses 2 and 3 added to the
original statement, is correct in terms of its logical form
The expansion of the original statement shows that not one,
mm at least three premisses are involved in the argument.
The logician, in his analysis, has not changed the reasoning
of the argument at all, but has simply brought out that an
acceptance of more than the formal correctness of the
argument requires the acceptance of at least three assumptions.
Most of the so—called informal fallacies deal not with
the form of an argument itself, but with the manner in which
the premisses are stated (or unstated), the attitude of the
érguer, or the psychological effects of specific terminology

or attitude. What is to be emphasized here is that in any

 

 

 

 

 

 

 

 

 

 

 

.-...1

. :‘r

 

 

 

 

 

 

11 ,1

2A

argmmnt, the truth of the premisses, or the acceptance of 5
thmm is a matter apart from the logical correctness of the
argument itself. It is not strictly appropriate to refer
to dealing with such problems as an application of ”informal )
logic" but rather as the investigation of the logical
problems of dealing with informality.
In the later discussion of deductive logic, reference
wffl.be made to informal axiomatic systems. Here, the
impression that there are formal axiomatic systems and

hfibrmal ones is a correct one. It is important to under-

 

stand that both of these types of system are part of formal
logic; i.e., introducing the adjective "informal" at that
time is not to be construed as in contradiction to what has

been stated above.

Division and Disagreement
The second and third general impressions listed earlier
may conveniently be discussed together. In a statement
relevant to both impressions, H. S. Leonard comments:
Some earlier authors have thought of deduction
and induction as only two among many types of proof.
But other authors have used the terms in . .
a jointly exhaustive sense. . . . Many logicians in—
sist that perfect induction is a form of deduction.l
Em much talked-about dichotomy is here seen to be not

above question—and by logicians! In a more explicit

cmmwnt, likewise indicative of disagreement, but more

 

 

1Henry S. Leonard, Principles of Right Reason (New
York: Henry Holt and Company, 1957), pp. 434—435.

 

 
 
 
 
 

:05. .r «I 2.. \ . . . .
_ at C. ”h

 
 
 
 
 
 
 
 
 
 

 

 

 

 

25

 

basically concerned with the subject matter, A. P. Ushenko
suggests:
It would contribute to clarity and order of

thought to have inductive logic completely disasso— .

ciated with logic, to discard its misleading name,

and to reorganize its contents by their incorporation

into a more comprehensive study, the ”Methodology

and Philosophy of Science.”1
(Net such a statement could be made by a logician becomes
nwre understandable when the nature of inductive logic is
explored and the tendency in much of the popular literature
M>equate ”logical” with ”certain” is recognized. Admitting
Hmt some disagreement does exist, for the sake of simplicity
0f eXplanation, an effective method of dichotomizing logic
Duo inductive and deductive will be employed later in this
paper.

DisaSreements among logicians exist not just in con—
rmction with discussing the organization of the subject, but
also regarding the subject matter itself. Although there
are rules of inference and some terms which, unlike account—
ing Hprinciples,” are universally accepted, there are like—
The existence

Wlse some few rules which are controvers1al.

and lengthy discussions of logical paradoxes bears witness

Ushenko, The Theory of Logic (New York: Harper

A. P.
8c Brothers, Publishers, 1936). p« 139-

r exam 1e r of b contradiction and proofs 1n-
VOlVinS the usg of Eogn's Iemma (a well—known formulat1on
IanrdinS a maximal element as considered in set theorY) are
Congidered by many as controversial. For those who question
the validity Of this reasoning, a conclusion not acceptable
§° them and whose argument includes the use of these rules,
18 Simply not proved. ,

 

 

 

 

 

 

 

 

 

‘

 

 

 

 

\

F ..

to the fact that some logical tools and suppositions are not 1

 

altogether without hazards. Lewis and Langford have defined

a paradox to arise "whenever we seem to have two incompatible
I

 

propositions true on logical grounds or for logical reasons.”
I This definition is more optimistic than the later comments
3 of these authors suggest is warranted, for after classifying
Hm paradoxes according to the nature of the difficulty in—
volved, the authors stated:
We shall then try to get some notion of the

theories which can b§_advanced by way of explain—

ing their occurrences, as well as the technique

that has been used for the purpose of avoiding

them in practice.2 (Underscore added.
Elfact, it might be suggested that incompatibility is, in
a sense, the result of the logical suppositions relating
to what are called the laws of contradiction and excluded
[Mdd1e. (These say, briefly, that no proposition is both

true and false and every proposition must be either true

or false.) Although technically the law of excluded middle

 

does not require the use of a two—valued (true—false) logic

Hithe sense that one need not deny the law in order to

accept a three—valued or n—valued logical SYStem: the
tWO-Valued logic is, to a great extent, essential to the
WhOle conceptual scheme. Logical laws: and hence appli-

0ations of them, should not be considered beyond revision

lclarence Irving Lewis and Cooper Harold Langford, 2)
S mbolic LO ic (New York: Dover Publications: Inc., 193 ’
p~ 3 .

2Ibid.

 

 

 

 

 

 

as

 

H »«

 

 

 

and expansion.

There have been suggestions, stimulated largely
by quandaries of modern physics, that we revise the
true-false dichotomy of current logic in favor of
some sort of tri— or n—chotomy.

There has, in fact, been some research done in connection

wiU1nmny-valued logical systems, and use is often made of
Huee—valued systems in proofs of the independence of the

axioms of some axiomatic systems.

There are many other areas in which logicians have
disagreements or questions about their own field of know—
ledge, the resolutions of which may very well significantly
affect the methodology of those other subjects choosing
to utilize the tools of the logician. The above discussion
should be sufficient to indicate that in the study of logic
as in most other fields, complete harmony is not to be found.
Itis appropriate to note here also, that the foundations
of what will be termed later ”mathematical or symbolic
logic" were laid relatively recently (the nineteenth cen—
tury), and as an indication of the state of development of
that discipline, Lewis and Langford stated, as late as
1932:

We find ourselves in a lively period of new
discoveries; an old subject, which has been com—
paratively stagnant for centuries, has taken on
new life. We stand today, with respect to logic,
where the age of Leibnitz and Newton stood with
respect to what can be accomplished in terms of

number; or where Riemann and Lobatchevsky stood
with respect to geometry. A wealth of new facts

 

l . .
Quine, o . cit., p. x1v.

 

 

 

 
 
 

 

 

 

28

dawn on us, the significance of which we are
only beginning to explore.

Preliminary Considerations
for the Accountant

From this preliminary discussion of logic in general,
accountants who would wish to utilize the methods and
techniques of the logician must be aware of four important

considerations.

1. Only arguments in the logician's sense of the
word (premisses, intermediate inferences, if
any, and conclusions) are the proper subject
matter for logical methods. For this reason,
discussions of accounting topics must be
separated into formal and informal arguments,
and simple discussions.

2. Only the form of those arguments will be on
trial, and not the value, reliability or
acceptability of a series of assumptions or
premisses.

3. It should be seen immediately that if an ob—
jection is raised to a particular assumption
in some argument under consideration, logical
analysis may offer further assistance only if
that assumption can be formulated as the con—
clusion of still another argument whose form
can be analyzed in accordance with the rules
of logic.

4. The use of logic, as opposed to the study of
it, must be coupled with an understanding
that there are what might be termed con—
ventions employed, that methods and the con—
ventions are subject to change, and that the
field of logic is still developing.

 

 

 

How these matters may limit the use of logical tech—

 

1Lewis and Langford, op. cit., p. A.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

29

niques in accounting and also help to isolate problem

areas will be further discussed in later sections of this
paper. What can be stated, unequivocally, now, is that
whatever assistance the accountant may receive from the
utilization of logical techniques,it will come from methods
of formal logic. Hence, it is these methods which will be

examined here.

Basis for Examination

 

Much insight can be gained into the application of
formal logical methods through an examination of their
characteristics. To simplify this examination, it is
desirable to categorize these methods in some way, and
despite what has been said thus far, the inductive-deductive
division will be used. As was stated earlier, the division
of logic into inductive and deductive is a common but not
universally accepted one. It is not desirable to preju—
dice the case for use of methods of either type, so an
explanation will be given for rejecting certain distinctions
commonly proposed. The basis of division to be used in this

Paper will be explicitly stated.

Logic and Discovery

A distinction is sometimes made between induction and
deduction which may give the impression that induction is

the logic of discovery. Such a distinction characterizes

 

 

 

 

 

- .~
15.11. .lV‘

 

 

 

 

.tiiu

“pH“

"‘flrn
. n...

..-,,.,
u

 

 

 

 

 

 

 

3O

deductive arguments as those in which
all of the information or factual content
in the conclusion was already contained, at
least implicitly, in the premisses .

mule in inductive arguments,

the conclusion contains information not
present, even implicitly in the premisses.

Sudia distinction may produce an immediate prejudice in
favor of using inductive methods. This distinction sug-
gests that deductive techniques will not extend knowledge
mule inductive techniques will. This suggestion is most

distressing, and has, I believe, influenced some members

 

of the accounting profession2 to conclude immediately that
attempts to utilize deductive techniques would make no con-
tribution to the organization and content of the body of
accounting knowledge and ought to be discontinued (or,
perhaps, not started at all would be more appropriate).

Even assuming that no new general statements about

 

accounting could be ”deduced” from already known ones,
would it be considered constructive and valuable to know
that such deduction is possible? To the writer, it seems

that the value of such a contribution to the organization

 

1Salmon, op. cit., p. 14.

 

2See, for example, the comments of William J Vatter,
”Postulates and Principles, ” Journal of Accountancy,
CXVIII (July, 1964), pp 59- 6H__LaSalle, op. cit. ; Malcoln
L Pye, "Reasons, Probabilities, and Accounting Principles,’
The Accounting Review, XXXV (July, 1950) Pp- A37- 443; and
to some extent, Spencer, op it~

 

.AJ
0
.rD
n3.
.w“

RE
Ah

1

 

w 4
'1" rv

.,. ‘.
Zhu

 

 

31
of the body of accounting knowledge would be substantial.
Moreover, the use of deductive techniques might be bene-

ficial in another, perhaps negative sense.

Not only does it rule out impossibilities but
it reveals the possibilities of hypotheses other
than those usually taken for granted; and in this
respect it frees the mind and contributes not only
to the fixed form but to the living growth of
science.1

Such a pessimistic view is not necessary, however. It
would have to be a very broad sense of containment that
would make all the information in conclusions "contained" in

the premisses. ”All the possible games of chess that can be

played can be deduced from the few rules of that game,”2

yet one would not want to say that the games are contained
hithose rules. In discussing the applicability of deductive

reasoning, Hempel comments:

Thus, in the establishment of empirical know—
ledge, mathematics (as well as logic) has, so to
speak, the function of a theoretical juice extrac—
tor; the techniques of mathematical and logical
theory can produce no more juice of factual infor—
mation than is contained in the assumptions to
which they are applied; but they may produce a
great deal more juice of this kind than might have
been anticipated upon a first intuitive inspection
of those assumptions which form the raw material
for the extractor.

 

It seems impossible to effectively distinguish between

 

lMorris R. Cohen, A Preface to Logic (Cleveland: The
World Publishing Company, 1944), p. 21.

2Ibid., p. 27.
3Carl G. Hempel, ”On the Nature of Mathematical Truth,”

The Azerican Mathematical Monthly, LII (December, l945),~
p- 55 .

 

 

 

 

 

 

V
E.
[l

 

. A: g
,.
.
22'

 

   

 

n...

 

 

 

32
induction and deduction on the basis of the type of infor—
mation (new or old) each will produce, and this distinction
is rejected for the purposes of this paper.
It is appropriate at this time, however, to consider
one sense in which the above comments on deductive logic are

extremely important for those who would advocate its use.

Whatever form deductive argument may take, it is true that

 

the terms appearing in its conclusion(s) can be those and

only those which appear in the premisses of that argument.

Some authors have suggested that accounting ”principles”

are normative in nature, based on and derived from account— ;

ing "postulates" which are said to be of an absolute nature. 3

Fbr example, Professor Moonitz reports that after recognizing

and defining the problems to be solved, one moves ”to their

solution by careful attention to what 'ought' to be the case,

not what 'is' the case,” and that ”relatively heavy reliance

must be placed on deductive reasoning in the development

of accounting postulates and principles.”1
If the interpretation of "normative” intended by

accountants is in the sense of the ”should” or the ”ought,”

as I believe it is, there will be an immediate limit on the

 

use of any method of deduction. Although particular state—
ments may be derivable from general ones, normative state-
ments must be deduced from normative statements. One can

reason from ”All assets ought to be recorded at their

 

lMoonitz, Postulates, p. 6.

 

 

 

 

i
i

«Wren

fishing»

 

 

 

 

 

w

I 33

exchange value” to ”This asset ought to be recorded at its 3

exchange value.”

One cannot reason to such a conclusion from A
the assumption "All assets have exchange value." It is

normal and acceptable to find the words ”should be” in a

conclusion only in the sense indicated by a premiss such

as ”All assets app recorded at exchange value” and a con—

clusion stated as the one above; i.e., if you start with the

assumption that all assets are recorded in a certain manner,

it does follow that whatever asset you look at, you should

find that it is recorded in that manner. To deduce norms,

it is necessary to hypothesize norms.

 

The situation is not unlike that in the statement of I
a theory in economics. A simple statement of identities,
such as, for example, income equals consumption plus invest-
ment (Y=C+I), or the money supply times the velocity of its
circulation equals the price level times income (MV=PY),
is not sufficient to reflect a theory. It is necessary to
include at least one behavioral equation in order to contri—
bute empirical content to the concluding equations. Thus,
continuing with the examples, to (Y=C+I) one might add
(C=a+bY) and (I=I), or to (MV=PY) one might add (V2V); and,
what results are theories which can be investigated. The

behavior of the economy can be compared with the theory

 

because the theory hypothesizes behavior. Hypothesizing a
state of affairs allows the exploration of the consequences

of that state; without an hypothesis you can deduce nothing

 

 

 

"4-;- "m
.,- .r
.... . «

.1

1?
.. ,,
m-wm

13114-1.
.——

- rm
‘ ‘w’lC

 

 

 

"L no
Lu
*3..."
.....-.v.
A .1
A.

 

V» .
_~'A\
.N

 

. 4

 

34

save that which is true by definition in the system.

If one proposes, then, that accounting principles are
deducible from postulates, either both groups of general
statements must be, in some part, normative, or neither
group will be. For, if the principles are normative, their
normative aspect must have been present in the postulates. If
the postulates are not normative, whatever principles are

deducible from them have no such additional characteristic.

Logic and Data Gathering

Under a commonly used distinction, appearing often
in accounting literature,l those methods of reasoning from
the particular to the universal or the less general to the
more general have been called inductive, the reverse
reasoning being true of deduction. Reflection on this
distinction seems to suggest that the process of data
gathering and compilation is reserved for those who would
use inductive reasoning. The users of deductive reasoning,
on the other hand, are less ”empirical,” and certain pre—
judices are again provoked.

Such a characterization will not be used in this
paper; for, since it leaves some arguments as neither induc~
tive nor deductive, it does not effectively dichotomize all

arguments. Examples may make this point clear.

 

1See, for example, the comments of William J Schrader,
"An Inductive Approach to Accounting Theory,’ The Accounting
Review, XXXVII (October, 1962). pp 645 6493 13833119) L010-

 

 

 

 

 

 

 

 

 

 

 

 

35

This bank account is either free or 1

restricted. Premiss
This bank account is not restricted. Premiss '
This bank account is free. (Conclusion

This argument moves from two statements about a particular
bank account to another statement about this particular
bank account. Under the suggested distinction above, it
would be neither inductive nor deductive. Consider also

the following argument.

Machines A and B are both screw machines. (Premiss)
Both machines were produced by the

same manufacturer. Premiss)
Both were purchased in the same month. Premiss)
Both produce at the same rate. Premiss
Both are used to produce the same product. Premiss)
Machine A required a complete overhaul

after 10,000 hours of operation. (Premiss)
Machine B required a complete overhaul

after 10,000 hours of operation. (Conclusion)

An argument of the above form has often been called an
argument by analogy, characterized as neither inductive nor
deductive. It also moves from the particular to the parti-
cular, and under the distinction suggested above would be
neither inductive nor deductive.
The argument form called perfect induction is
exemplified below.
Unprocessed material on hand is inventory. (Premiss)
Partially processed materials on hand
are inventory. (Premiss
Completed materials on hand are inventory. Premiss
Raw, partially processed and completed
are the only stages in which you find
materials on hand. (Premiss
All materials on hand are inventory. (Conclusion

In an argument of this form, the premisses exhaustively

report the term generalized in the conclusion. It is

 

 

 

 

vrr» "
......zal

r1)

 

 

 

l ..

reasoning from the particular to the general. However, one
would want to attribute to the conclusion a degree of
acceptability that reasoning in that direction does not
normally receive.

Since the distinction based on the quantity (universal
or particular) of premisses and conclusions leaves at least
three general types of arguments (those shown in the examples)

belonging with neither inductive nor deductive, it will be

rejected as not an effective characterization. The exa-

mination of the general characteristics of logical methods

 

can, nevertheless, be expedited by an inductive-deductive

division. When confronted with an argument, the recognition

of its inclusion in the group of inductive or deductive

arguments will better enable one to understand its strengths

 

 

and its limitations. Hence, what is wanted is a distinction

which could be applied to effectively categorize all

arguments.

Eggposed Distinction
The distinguishing characteristic used in this paper

relates to the nature of the argument. The distinction will

be made for valid arguments (correct arguments), only,

since as Leonard says:

On the whole, there is not much point in trying
to classify invalid arguments as deductive or induc-
tive; they are invalid, and that should usually be

an end of the matter.

lLeonard, op. cit., p. 433.

 

 

37

A'deductive argument shall be one such that no other
statements, if added to that argument as premisses, can '
make that argument invalid. An inductive argument shall
be one such that at least one statement, if added to that
argument as premiss, can make the argument invalid.

Stated another way, if an argument is deductive, then if all

the premisses are true, the conclusion must be true. If an
argument is inductive, then if all the premisses are true,
the conclusion is to some extent probably true but not
necessarily true.2

With this distinction, we can safely call the earlier

 

argument about the bank account deductive; i.e., if the two
premisses are true, there is no additional statement which
will cause the conclusion to be false. The distinction
allows the classification of the machinery argument as
inductive, since a single statement about Machine B having
surpassed 10,000 hours of operation without need of overhaul
will cause the conclusion to be false. The inventory argu—
ment is recognized as deductive (despite its common title
of perfect induction).

The next two chapters will be devoted to the logical

methods which can be categorized as deduction and induction.

 

 

1The substance of the distinction is due to the
comments of Leonard, in Principles of Right Reason.

2Use of the word ”necessary" at this point is not
intended to convey the meaning characteristic of necessity
in modal logic where something is necessary if and only if
its negation is not self—consistent.

 

 

 

 

 

 

M,
m

u

 

 

 

CHAPTER III

DEDUCTION

Methods in General

 

Discussions of deductive logic often differentiate
between methods of traditional and methods of symbolic or
mathematical logic. The differentiation is misleading for
atleast two reasons. First, although the methods of modern
logic make more extensive use of symbolization, Aristotle
lflmself, the recognized founder of logic, also utilized
symbols to implement his work. Secondly, with one minor
exception,1 the inferences which were accepted in tradi~
tional logic are likewise accepted in.modern logic. Thus,
Hm valid arguments of traditional logic form a proper
subset of the valid arguments of symbolic logic.

The difference between the old and the new
logic is one of degree rather than of kind, but

 

 

lAs generally presented, the logic of Aristotle
accepted certain inferences as legitimate which would be
precluded by the acceptance of the null class. Thus, from
the statement ”All F5 are Gs” it could be asserted that
Some Fs are Gs" logically followed. The more modern inter-
pretation of the original statement is that it would be
true if all Fs are Gs and also if there are no Fs. Under
this interpretation, one could not assert that some Fs
are Gs legitimately. The modern interpretation, which
does not preclude working with empty classes (or, more
Specifically, the null class), has proved more effective
than the other, older interpretation.

38

 

 

 

 

the C

 
 
 
 
 
 
 
 

./\
A y "me. a l. A» 0v
yr... .1. F... Al
F u AA. . . S HM
_ u 2.. r .. in m: . w 3w .0. Mum n‘
w a .. .. .ml .1. ... , .. q, ... 4.... :, .. .mu: \ H\.Q

 

 

 

 

39

the difference in degree is tremendous.1

One argument should be sufficient to exemplify the
difference in scope. We should want to be able to say
‘Umt the following argument is valid without the addition
of any more information:

All marketable securities owned are assets.

Therefore, the cost of a marketable security

owned is the cost of an asset.
hm argument is admittedly simple (even trivial), and it
would be difficult if not impossible to find someone who
Infused to accept it as correct reasoning. Nevertheless,
it has been established that utilization of all of
Aristotelian logic will not allow the inference to be drawn.
Mme modern logic would allow for proof of such an argument
by using terms expressing relations.

Unfortunately, several accountants who have discussed
the use of formal deductive logic do not seem to be aware
that the science of logic has advanced at all since the
days of Aristotle. For example, Mr. Malcoln L. Pye treats
formal logic as equivalent to syllogistic reasoning (another
name for traditional logic).2 The same misunderstanding
appears in Professor Schmidt's article, ”Practical Use of

the Device of Formal Logic in Accountants' Daily Work,”3

1Irving M. Copi, S mbolic Lo ic (New York: The
Macmillan Company, 1965): p- 7.

2Pye, o . cit.

 

3Schmidt, op. cit.

 

 

 

 

 

 

40

and some writings of Professor Littleton.1 This misunder—
standing has caused these writers to underestimate the
power and the usefulness of formal logic. However, given
flat the essence of traditional logic is included in the
scope of symbolic logic and that the latter is an immense—
ly more powerful tool, the present study of deduction will

pertain to the techniques of the modern logician.

The S stems A roach

When argumentation is analyzed in such a manner
that the smallest unanalyzed unit may be smaller than a
complete sentence, such argumentation in modern logic is
generally presented in the form of systems which are
either formal or informal. A system consists of:

1' Signs or symbols and an effective method
for determining whether a sign or symbol
belongs to the s stem (usually accomplished
by making a list 3

2- sequences of symbols called formulas, which
may be either meaningless or meaningful
(called well-formed formulas);

3. rules for constructing sequences of well—
formed formulas (creative rules, commonly
called axioms, and transformation rules,
commonly called rules of inference).

The difference between formal and informal systemS, like
that between formal and informal arguments in ordinary
language, lies in what is left unsaid. Technically, in

a formal system, each initial sequence of a proof Withln

 

lSee, for example. W51-

 

"O

 

i! . T. .9 Tn..‘
2w
.3.

 
 
 
 

(N .Hv M. A .‘l NAM I
.5 .mxd ... ....
7.. x. , .

 

W.
M
.

 

 

 

 

 

Lu

the system is also a proof within that system while in
aninformal system, initial sequences of a proof are not
always also a proof. In more general terms, a formal
system operates with no presupposed knowledge, such as
meledge of accepted logical inferences or meanings of

logical symbols.

Axiomatic Methods

 

"Axiomatic” is the name given by logicians and mathe—
maticians to the method of formalization which

. . . begins with a list of undefined terms and a
list of assumptions, or postulates involving these
terms, and the theorems are to be derived from the
postulates by the methods of formal logic. If the
last phrase is left unanalyzed, formal logic being
presupposed as already known . . . the development
is by the informal axiomatic method. And in the
Opposite case we shall speak of the formal
axiomatic method.

ere is a technical difference between a formal system, as
discussed above, and a formal axiomatic system. It is
muficient, for the purposes of this exposition, to explain
that the difference again relates to what is left unsaid,
and formal systems are generally said to ”have deduction
itself as its subject matter.”2 In either SYStem type,
fmwever, it is not legitimate to omit any premisses in any
Mpof, or the resulting sequence of formulas is not a

pPOOf. A formal system, equated in this interpretation With

1 - ' 1 Lo ic

Alonzo Church Introduction to Mathematica . :
¥fl£fl§.l_(Princeton: New Jersey: Princeton Univers1ty
regs, 1956), p_ 57_

2
Copi, Symbolic Logic, p. 184.

 

 

 

 

 

 

 

 

 

 

L
‘v

...A '(7

,
f
.

 

 

 

 

 

 

..,,.

 

41

the system is also a proof within that system while in
mainformal system, initial sequences of a proof are not
always also a proof. In more general terms, a formal
system operates with no presupposed knowledge, such as
meledge of accepted logical inferences or meanings of

logical symbols.

Axiomatic Methods

"Axiomatic” is the name given by logicians and mathe—
maticians to the method of formalization which

. . begins with a list of undefined terms and a
list of assumptions, or postulates involving these
terms, and the theorems are to be derived from the
postulates by the methods of formal logic. If the
last phrase is left unanalyzed, formal logic being
presupposed as already known . . . the development
is by the informal axiomatic method. And in the
opposite case we shall speak of the formal
axiomatic method.

 

flmre is a technical difference between a formal system, as
discussed above, and a formal axiomatic system. It is
mnficient, for the purposes of this exposition, to explain
that the difference again relates to what is left unsaid,
and formal systems are generally said to ”have deduction
itself as its subject matter.”2 In either system type,
meever, it is not legitimate to omit any premisses in any

proof, or the resulting sequence of formulas is not a

proof. A formal system, equated in this interpretation with

 

1Alonzo Church, Introduction to Mathematical Logic,
Volume I (Princeton, New Jersey: Princeton University

Press. 1956) p 57

Cori, Symbolic Logic, p. 184.

 

 

 

42

logistic system, is the most rigorous form of deduction.
&mh rigor is appropriate for the study of deduction, but
exceeds that required for the study of most other subjects,
including accounting.

There are, then, two methods of deductive logic to be

dealt with here-formal and informal axiomatic methods.l

 

Whmia system is to be set up to formalize the discussion of
some area of a particular subject, the system builder is
said to be using the axiomatic method. The absence of a
(fistinction between logical and subject matter primitive

symbols, or of explicit statement of effective rules for

 

recognizing meaningful formulas, or of some of the rules of
proof-construction, or some combination of these charac—
terizes an informal system, and the builder of such a system
is said to be applying the infonnal axiomatic method.

In Church's definition quoted above, in a formal
axiomatic system the formal or underlying logic is not
presupposed. A formal axiomatic system is only slightly
less rigorous than a logistic system, since the primary
distinction lies in explicit separation of the primitive

symbols and axioms into those pertaining to logic and those

 

 

 

The use of natural deduction systems with no stated
postulates is considered here as a special case of the infor-
nml axiomatic method. Such systems can be used to ex-
plore the consequences of individual premisses which should
be stated as part of each deduction in which they are assumed.
But, where more than two or three assumptions are dealt with,
the establishment of them as axioms is more convenient and
the system with these axioms added will be a case of the
informal axiomatic systems used here for discussion purposes,

 

 

 

1""!
l“ V"
Tl
..
:07:
.4. H
..;_., .;

 

2*:‘71:
«mu

"W.“

will

: W
W“

1;“

 

 

 

43

pertaining to the individual branch of knowledge under
study.

For the purposes of researchers in fields other than
logic, the formal axiomatic system requires unnecessary
precision. The formal axiomatic method has been used on
occasion by mathematicians, However, most mathematical
presentations, even when dealing with highly complex and
abstract problems, involve the informal axiomatic method.

Accountants perform mathematical operations such as
addition, subtraction, differentiation, etc. without stating
at the outset of each discussion the rules by which these
operations are performed and the justification for their
allowability. If accountants find a use for the axiomatic
method, whatever systems are constructed will accept at
least one (and probably several) symbol (viz., the equality
or identity sign), and the rules for its use without expli—
cit incorporation into the formulation of the system. If
accountants were to explicitly include all suppositions
within their system, it seems doubtful that the system-
builder would include a complete set of effective for—
mation rules which stipulate which symbol sequences are
meaningful—persons conversant with accounting and logic
will be presumed to have knowledge of what sentences are
meaningful. If some accountants become more familiar with
logical inferences, it is likely that whatever systems are

constructed using the axiomatic method will presuppose the

 

 

 

 

 

 
 
 
 
 
 
 
 
 
 

 

 

 

 

44

tmflmlogies of the propositional calculus.l Since any

axflmmtic system constructed with such presuppositions

wmfld not qualify as formal, accountants using the axiomatic

meflmd will, for good and practical reasons, undoubtedly

mmfloy the informal axiomatic method. The degree to which

rign’is lost in return for practicality will depend on the
(kgree of informality in the system constructed. Since it
islikely that the informal axiomatic method will be the

one useful in accounting, the remainder of this chapter will
be devoted to that method. Most of these comments apply

emwlly well, however, to formal axiomatic systems.

Informal Axiomatic Method
As explained earlier, informal axiomatics is a deduc—
tive method of dealing with a subject by constructing an

hmormal deductive system. At a minimum, the basis of the

1In simple terms, tautologies are sentences that are
true for every combination of truth—values (truth and
falsity) of the variables in the sentence. One of the many

mmsible translations of several of the more familiar tau—
tologies of the propositional calculus are given below as
examples.

p v —p) A proposition is either true or false. (The
law of excluded middle.)

—(p & —p) A proposition is not both true and false.
The law of contradiction.)

(p:>-—p) If a proposition is true, then it is not
true that it is false. (Converse law of
double negation.)

(pTDp) If a proposition is true, then it is true.
(The reflexive law of implication.)
(qub (—q3 -p)

If, whenever one proposition is true a second
is true also, then, whenever that second
proposition is false, the first is false
also. (The law of contraposition.)

 

 

 

 

45

system must include

1. a list of undefined terms relating to
the subject matter;

2. a list of axioms;

3. some reference to at least some of the
rules of inference to be allowed; and

A. an intended interpretation of the un—
defined terms.

figgr and the Importance of Rules

The rigor of reasoning reflected by a given system
inits proofs is a function of the nature of the allowable
transformation rules and the extent to which the system—
WUlder has explicitly stated them. The following comment
appeared in an editorial in the Journal of Accountancy
nmmdiately after the publication of Accounting Research
Study No. 3:

Dr. Sprouse and Dr. Moonitz have given great
weight to the word co—ordinated, and have accordingly
constructed their set of principles in accordance
with rigorous rules gflogic.l (Underscore added.)

Mm Moonitz presents fourteen general statements about
accounting and other things which, he suggests, form a
ibundation for accounting principles. However, the princi—
ples suggested in Accounting Research Study No. 3 have not,
Ulany way, been shown as having been the logical conse~

mmnces of the postulates, inferrable from them through

 

 

1Editorial, ”The Approach to Accounting Principles,”
CXIII (May, 1962), p. 37.

 

 

 

 

 

 

 

 

 

 

Li: J1 .

imam:
,‘ VLIUVM

 

 

 

 

 

 

 

 

m.

~21“
a...

 

 

 

 

 

 

1L6

 

the use of rules of formal logic.l No rules for allowable
inferences were given.' Professor Devine, as quoted earlier,
referred to ”stock intellectual cliches” utilized in account—
ing literature and discussions. I believe he would agree
that ”it follows logically that . . ." falls in this class.
The specification of rules of inference in an axiomatic

system i_s a precise statement of what the individual author

actually means when he says ”it logically follows. ” The

 

reference to an explicit rule of inference stated at the
outset of an argument i_s the support for the use of the

cliche on a particular occasion. Without such a statement,

 

any reader of such argument (proof, justification) may

-..‘__

legitimately ask "How does it follow?”
The more explicitly stated the rules, the less open
to criticism are the system—builder—arguer's proofs

(assuming, of course, that the rules are followed). In

 

the way of more general example, one system may require, by
its more explicit statement of the allowable inferences and
the types of inferences allowed, six steps to construct a
proof of a given theorem, while another, composed of the
same undefined terms and axioms, might allow the proof of
that theorem in one step. Assuming the systems are of

equal quality (characteristics of quality to be discussed

 

shortly), the proof in the first system is said to be more

 

1More complete discussion of these studies can be
found in Chapter VI.

 

 

 

h

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

47

 

rigorous than that in the second. Less is left to the

':Mmgination. Inferences which are obvious to some are
rmt so apparent to others; rules of inference which are
familiar to some are unfamiliar to others.

Scope of the Problem
To Be Solved

 

The use of the informal axiomatic method does not by
itself limit the scope of the problems to which it can be
applied. It is often noted that completeness is a require-
nmnt of an axiomatic system, and this requirement is inter—
preted by some as dictating the type and scope of problems

to be attacked with this method. But, completeness is

‘.-. _

misunderstood if interpreted in this manner. A system is
complete if its theorems are all those which the system-

builder intended to find as theorems of the system. As

 

sudn the requirement suggests that the system is suited

to the problem at hand, and in no way demands a certain
scope or complexity of the problem to be handled. G. Peano
(1858—l932) devised a system to derive the entire arithmetic
of natural numbers. There have been several formulations of

Euclidean geometry.l There have likewise been proposals, at

 

 

lFormalization was valuable since, despite literary
references to Euclidean geometry as an outstanding example
0f the axiomatic method, Euclid's formulation was extremely
informal by modern standards of rigor. No rules of infer—
ence were stated and, in fact, some of the proofs required
assumptions other than those stated as axioms and postu—
lates. Later references in this paper to Euclidean geometry
are to be understood as pertaining not to the system as proposed
by Euclid, but as formalized by later geometers.

 

 

 

4‘
\V.

 
 

y,:.“u A

 

 
 
 
   
 

 

 

 

 

 

48

leamb for outlining and formalizing a general ”theory”lof
amounting as well as suggestions that certain areas of
accounting be'axiomatized. DR Scott, in 1949, views the
theory of accounting and suggests:

As the system of accounts has become more complex
and the variety of services rendered by it has
increased, the general concepts and rules governing
its operation have become correspondingly broader.

The logical outcome of this evolution would be a
consistent hierarchy of rules and principles pro-
ceeding from the specific and detailed to the more

and more general until the broadest accOunting
principles merged into still broader principles of

social organization. Throughout its modern history
accounting has been moving towards such a perfected
system of theory.
hlwhat way can the experience of others be related to the i
accountants' proposed use of the axiomatic method?
Though the scope of those formalizations mentioned
above is immense, they do have some limits, and there is no
basis for concluding that all those who resort to the axio—

nntic method did have, or need have, such grand—scale objec-

tives. The systems constructed by mathematicians3 have not

 

1”Theory,” "concept,” ”postulate” are generally not
presented in quotation marks in this paper in two contexts:
U) when the usage is common in a field other than account-
ing, as in relativity theory, and (2) when the usage coin—
cides with that suggested in Chapter V.

. 2DR Scott, ”The Influence of Statistics upon Account—
ing Technique and Theory,” The Accounting Review, XXIV
(January. 1949), p. 85.

 

3Historical facts stated in this paper relating to
mathematics, physics and other sciences were synthesized
from various sources, but especially from the following:
R.Blanche, Axiomatics, translated by G. B. Keene (New
york: The Free Press of Glencoe, 1962); Albert Einstein,
Fundmmnts of Theoretical Physics,” published originally

 

 

 

n NU. a
.l. 0 at S LIV .\
. n. rub l n (x
.1 cc 0 F1. .P. 2. cm 3.» it i. \
\. H1— .M.... r lulu Kiln A. A»... W. ..\\V «DI». ‘ . A «U . n
w .. .. .. .. ..I i. 1.. C. T». v.. .. ..v.\.vr. A/~;\uv\xvh..\.. .T‘ \

 

 

 

Ff — wti

49

all started out to axiomatize the whole of mathematics. The }
Peano system was concerned with only the natural numbers.

One of Tarski's interests was the algebra of real numbers.
Woodgerl has utilized the axiomatic method in his formulation
of certain parts of biology, primarily genetics. It is true
that in some ways the Newtonian and earlier theories are

made obsolete by the presentations of relativity theory; but,

 

hlone sense at least, the theories are discussing different
things. The Newtonian theory coincides with the Einsteinian
theory when the velocities of bodies are small as compared
wraithe velocity of light. This fact suggests the notion
of a theory holding ”over a relevant range,” a phrase used

often also with economic theories. A related example of this

 

same notion is found in the field of geometry where axiomatic
systems have been constructed with interpretations that

reflect Euclidean and non—Euclidean ideas of space. The

 

in Science, XCI (1950); Philipp Frank, "Philosophical Inter-
pretations and Misinterpretations of the Theory of Relativ—
ity,” published originally in Interpretations and Misinter-
pretations of Modern Physics (Paris, 1938); and Eugene P.
Wigner, ”The Limits of Science,” published originally in
Proceedings of the American Philosophical Society, XCIV
(I950); the last three references reprinted in The Philosophy
Of Science, edited by Herbert Fiegl and May Brodbeck (New
York: Appleton-Century—Crofts, Inc., 1953), pp. 2l2—23l,
253—261 and 757—765, respectively; also Carl C. Hempel and
Paul Oppenheim, ”Studies in the Logic of Explanation,"
Philosophy of Science, XV, 2 (1948), reprinted in the Struc—
ture of Scientific Thought, edited by Edward H. Madden
(Boston: Houghton Mifflin Company, 1960).

 

 

 

 

 

1J. J. Woodger, The Technique of Theory Construction
(International Encyclopedia of Unified Science, Foundations
0f the Unity of Science, velume II, No. 5; Chicago: The
University of Chicago Press, 1939).

 

 

~_ W
F “—..—-

50

fiworems of the first have been interpreted (had specific (
memfings given to the symbols used in them) for use in most
ofthe geometrical operations with objects of our more
cmmwn experience. The theorems of the latter have been
hmerpreted for use in some geometrical operations per—
tahflng to astronomy and certain branches of modern physics.

Hmre is, in fact, still no assurance that physical space

 

corresponds to the postulates of Euclidean or non—Euclidean
geometries.

From the very beginnings of the sciences as we know
thmntoday, there have been hopes that a basis would be
found to unify all the various branches of study in each, and
perhaps even an underlying theory unifying all the physical

sciences. Research in this direction is continuing. Yet,

 

fbr the highly developed and co—ordinated science of physics,
fbr example, there is as yet no general theoretical basis.
The hope of formulating a system for a general theory

of accounting, of which the systems of, for example, financial

 

accounting, cost accounting and governmental accounting are,
for all practical purposes, special cases or subsystems, is

rwt beyond the realm of possibility. To the extent that

all accounting operations in these fields have something(s)
hlcommon it may be suggested that these things may be the
consequences of some general statements concerning the
accounting process. But other disciplines have not limited
their research in formalization to the search for a unifying

general theory. One might say that without the formalized

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

r' w
-—-‘.

51

lomwledge acquired in the special branches of each science, ,
Umre is nothing to unify. For the most part, the theories
hlaccounting are at a stage of development and formalization
much strongly discriminates against the success of such a
search. The axiomatic approach to the formulation of a gen-
eral theory of accounting is an ambitious objective at the
present time. The use of the axiomatic method in the pre—
sentation of some special area of accounting knowledge may

be a more limited objective, but an admirable one, for it
cmucontribute to the understanding of what is involved in
that area, and in the use of the method, and can pave the

way for research on a less limited scale.

Primitive Terms

 

 

Just as the scope of problems to be attacked is not
limited by the use of the informal axiomatic method, the

undefined terms in an informal axiomatic system are not

 

subject to strict limitations. In the Peano axiomatization

of the arithmetic of natural numbers there were three
primitive notions. After the initial formulation (which

was not intended primarily to exhibit frugality in the
primitive base), successive, successful attempts to construct
informal axiomatic systems of Euclidean geometry reduced the
primitive terms from four to two.1 Economy per se in the

size of the primitive basis of a system is not a virtue,

 

lPasch's system (1882) required four terms and the
System of Pieri (1899) and Padoa (1900) required only two.

 

maps eve
E; interpré
1'5. system
{fairy cc

'.‘. studem

 

 

 

 

ll
. successor." Axioms are, SO to S

52

but it offers certain advantages.

For an informal axiomatic system to be of interest to
anyone beyond its originator, the intended interpretation
of the primitive notions must relate the system to some
mmject matter. For example, ”N” or "*” may be manipulated
within the framework of a system without any reference or
perhaps even any interest in what one might mean by them.
Byinterpreting ”N” as "number" and ”*” as "successor,"
the system becomes related to arithmetic. Most fields of
inquiry contain more than a few words of especial interest
to students of that field. If the number of primitive notions
is limited, additional terms must be defined by means of
those notions, and the problem of selecting such notions
as will be appropriate and sufficient for defining the
remaining terms is a major one. If the number of primitive
notions is not kept small, the problem of definitions is
alleviated to some degree but a new problem takes it place—
an increase in the number of axioms. Axioms assert that
certain relationships hold between the terms used in their

Statement and may be very simple or very complex. For

example, one of the Peano axioms of arithmetic can be stated

' n -
Simply as ”Every number has a successor, which states a

H
relationShip between the primitive notions ”number and

peak, together with the

definitions of a system, the only source of information
o 't

in the SVstem. If a primitive term is to be useful, 1
‘ ‘ ' terms
muSt appear in some axiom; a large number of primitive ,

 

n'li‘f‘f
"rm; FEW-

' ”I‘m

 

 

 

 

mull

~u .

 

 

 

53

then, requires either axioms made complex by the inclusion
of more terms, or a large number of axioms. In the presen—
tation of the results of an axiomatization as support for
one's argument, there is, at least initially, the question
of the acceptability of axioms (which will be discussed
later in more detail). If the number of axioms is large,
the question claims more time for resolution. If, instead,
axioms are made more complex, the deductive simplicity of the
system is decreased. Thus, though logic does not require
economy in the adoption of primitive notions, practicality
would seem to dictate that such terms be kept to a minimum.
An additional advantage is to be gained from using few
primitive terms. Again, remembering that the system may be
used in conjunction with the presentation of some argument,
even before the question of axioms’ acceptability arises, the
participants in the presentation must be aware of the
intended interpretation of the terms used in the argument and
its SLlpporting proof within a system. The fewer terms
involved, the fewer interpretations to be understood and
agreed upon as suitable at the beginning of the argument.
Finally, the ability to define a large number of terms of
the SUbJeCt matter by using combinations of only a few may
in itself be of value in adding precision to the pre—system-
atic understandings of those defined terms.
An example of the problems produced by SeleCtion Of
primitive terms, and the degree to which precision might be

reQUired in formalization of a general statement of

 

 

 

 
 
 
 
 
 

 

 

 

 
 
 
 
 
   
 

 

 

 

54

accounting, may be helpful to illustrate how logical presen-
tMflcn can differ from ordinary careful writing. Consider
Posmflates A—l and A—5 in Accounting Research Study No. l.1
Ad.states, in part, ”most of the goods and services that
we produced are distributed . . ." This statement may be
unwnstood to assert one of two different things.

1. All goods and services are produced, and
most are distributed, or

2. Of the produced goods and services, most
are distributed.

Rm difference between the two interpretations may appear
Might when the postulate is considered in isolation. How—
evmg consider Postulate A—5, which states that money is the
cmmwn denominator for the measurement of ”goods and ser—
ifices, including labor, natural resources, and capital.”
Nmm the first interpretation, above, of Postulate A—l can—
rmt be accepted; for, natural resources,which are part of
'goods and services” according to A-5, are not produced, and
Hm first interpretation of A—l asserts that fill ”goods
and services” are produced. Whether labor can be produced
is questionable and may depend on the definition offered
for the term ”produced.”

On the other hand,adoption of the second interpretation
of A—l may be more appropriate but brings up the question of

what can be said about goods and services that are not

Ithe result of production. Do the same observations about

 

1For a complete statement of the postulates discussed
here see Appendix A.

 

 

 

 

55

thethstribution through exchange apply? If conclusions
arechawn from the postulate as rephrased under the

second interpretation (together with other statements, of
course), it must be recognized that they can apply only to
produced things.

The addition of the explanatory phrase "including
labor, natural resources and capital” in A—5, though
possibly intended by the author only as a matter of example
br even as an afterthought), causes further problems in
the attempt to present a more formal and precise statement
of the postulate. If these additional terms are to be
retained in the more precise statement, the postulate may
remflre restatement in a form something like the following:

Money is the common denominator in terms
of which measurement is made of
1 things that are goods and labor, or
2 things that are services and labor, or
3 things that are goods and natural
resources, or
4) things that are services and natural
resources, or
5 things that are goods and capital, or
6 things that are services and capital, or
7 things that are goods, but neither labor,
nor capital, nor natural resources, or
8) things that are services, but neither
labor, nor capital, nor natural
resources.
Ems example indicates that even the extreme care which
Professor Moonitz must have taken in his writing does not
produce the precision which may be required for logical
analysis of the written words. The case for holding down

the number of terms, and especially primitive terms, is

very strong.

L‘-

 

flq

 
 
 
 
 
 
 

ct .. .. . . F. 1 .

3 Au . . m... i. . r... “.1 A.» 5 .
"V. :i .1. .C x. .. .. . , .. .
S k .. _ _ .2. y. . ..
.Z M... u. ..

7... 4 n .

 

 

 

 

56

The historically earlier attempts at the axiomatizing
of any specific field are less thrifty with primitive
rmtions for at least two reasons. First, the initial for—
unlization can be studied for the very purpose of accomplish—
ing a reduction in primitive terms. This can be seen in the
axiomatizations of plane geometry. Secondly, the initial
formalization usually generates renewed interest in the
pre—systematic, empirical definition of those terms taken as
prhmtive. The interest draws attention to the details of
mmh definitions, often suggesting new, so to speak, more
prhmtive notions which may be fruitful and less complex.
Geometry is a good example here, too, for an analysis of
the early primitive notions such as ”plane,” and ”segment"
led to their discard in faVOr of, in one system ”movement,”
and in another, "distance."

Attention to definition and interpretation as the
result of a systematization of a theory has produced other
effects. With the theory of relativity, the idea of simul—
taneity has undergone a change so that its complete expla-
nation is deemed to require reference to an observer. In
connection with the developments of mathematical systems, the
older definitions of ”number” have been reconsidered, and

substantial work has been done to clarify this notion.1

 

1Notably the words of Frege and Russell; Gottlob Frege,
The Foundations of Arithmetic, translated by J. L. Austin
(New York: Harper & Brothers, 1950) and Bertrand Russell,
fggroduction to Mathematical Philosophy, Second Edition,
0.

 

'mlopmem
tires: it
1:3."1 in

this an

:15 notior

 

 

,
~S
u

g

"/i'

 

 

 

 

57

Developments in the field of physics have renewed the
interest in an adequately explicated concept of ”probabil-
ity."1 While the actual use of the primitive notions
within an informal axiomatic system does not require that
the notions be precisely understood, the use of the sys—
tem as an adjunct to the oral or written justification of
a position held mediates in favor of such precision.

There is another problem which has confronted the
users of the informal axiomatic method which should be
briefly considered here.2 Most of the physical sciences
have now, in their vocabularies, both abstract and elemen-
tary terms (often called observation terms) .3 Which kinds

of terms should be primitive has been something of a prob—

lem. There has been a trend toward choosing those at a high

 

 

One of the classic investigations of the concept is
found in Rudolf Carnap, Logical Foundations of Probability
(Chicago: The University of Chicago Press, 1950).

2The problem of abstract terms has an interesting
Sidelight unrelated to the present discussion. Such
terms as molecule, rigid body, number and hydrogen ion
have been used as the starting point of many theories,
and the resulting formalized theories have been excep—
tionally valuable. Whether in fact there are such things
as these, however, is a question which has not been an~
swered. This writer adopts a kind of instrumentalist View
that if such abstract, theoretical constructions yield
beneficial results, the ontological questions will be left
to the philosophers. .

3A particularly helpful discussion on the use of
abstract terms is to be found in Rudolf Carnap, Foun—
dations of Logic and Mathematics (International Encyclo—
pedia of Unified Science, Volume I, No. 3; Chicago: Uni—
versity of Chicago Press, 1939) .

 

 

ad<

nu
. .4

 

Lem:

 
 

.01

n
n.
n

 

 

 
 
 
 
 

 
 

 

 

58

level of abstraction, but most of the sciences have not been
able to adequately define the wealth of elementary terms by
reference to these abstractions, and some of the former are
still to be found as primitive.

For accountants who would utilize the axiomatic method,
Hm problem of primitive notions may be a formidable one
but not incapable of solution. In some of the studies more
concerned with the nature of accounting, efforts have been
made to determine what Professor Littleton called ”the
center of gravity,”1 or the underlying concepts of the
accounting process. In these works, though, there is not
complete unanimity that, as Professor Littleton suggests,
'anome” is that "basic concept that makes accountancy
different from all other methods of quantitative analysis.”2
Dr. Marple is more inclined to "capital” as the basic

3

notion. In the recent American Accounting Association
nwnograph, Professor Goldberg proposes that characterization
of accounting will include ”basic premisses” or primitive

notions relating to activity, outlook, measurement and

record, and suggests that the notions selected might be

 

1A. C. Littleton, Structure of Accounting Theory

(American Accounting Association, 1953).
21bid., p. 18.

3Raymond P. Marple, Toward a Basic Accounting Phil—
oso h (New York: National Association of Accountants,
1954).

 

 

2m “ "com

m...”

"SEAS.
Ar ’
.m In

 

s

o

 

r

'pvent," "commander," "resource" and "reddito" (the last

59

being a word coined specifically for this purpose).1

Hucughout most of accounting literature, reference is made

to such terms as "cost," ”value,” ”economic benefits.” Even
less abstract terms such as ”liability,” "asset” and "equity"
lmve received considerable attention, though the variety of
hfierpretations of these terms and the problems which appear
hlpractice relating to their application in a given situ—
ation challenge the designation of them as less abstract than
Um others mentioned. Collections of a number of these
terms might form a primitive vocabulary for several axiomatic
systems. As was noted above, pioneering works in axiomati—
zation in most fields have not been particularly frugal in
adopting primitive terms. That these accounting terms are

not precisely defined or that there is no universal accept—

 

ance of a definition would not preclude their use for some

 

particular problems. For example, it is possible to say

several things about assets or recorded amounts without

I
I
i reference to cost. Or to make several comments about
i
; expenses without reference to liabilities or even equities.
l

One might adopt the axiomatic method for an exploration

' of those possibilities.
‘ At any rate, the use of the axiomatic method, with

its need for undefined terms, to more rigorously approach

 

lLouis Goldberg, An Inquiry into the Nature of
flflfifigfiipg_(American Accounting Association, 1965).

 

 

1.1. if . P-\ Q V .
L1. .. . T. nth A“ 1.. v p.” :v in «v .7 .
3 2.. Ac .1. .0 1.» . .- . a... .. . . .r.» a... ...i..
by Ca w . w . a. . r .. ...x.. . . .2 . t
.t n. it It .... n: L". ._.. r .
r. . . p . .. . .. n .
M... n.“ a . ... m

 
 
 
 
 

 

 

mpblems in accounting, would undoubtedly generate the same

60

hfiense interest in the adequate defining of basic concepts
aslms resulted in other fields of study. Following the
analogy further, accountants may find that concepts more
abstract than asset, income and equity will be necessary
or more fruitful as more research is done with the use of

this method.

Axioms

The selection of axioms poses at least as many problems
as the choice of primitive terms. The problems divide into
two basic categories, one related to the number of axioms and
the other connected with their selection. As stated earlier,
axioms can be said to "govern” the primitive terms; hence,
the number of such terms has some effect on the number, or
at least, the complexity of the axioms. In the Peano axio—

matization of the arithmetic of natural numbers, for

 

instance, the three primitive terms are governed by five
axioms. Further, in the discussion of primitive terms it was
noted that when the axiomatic system is used as an adjunct

to the presentation of an argument or a position, the accept—

ability of axioms is subject to question. Hence, the number

 

Of axioms will affect the amount of time devoted to estab—
lishing a preliminary "meeting of the minds” on the subject
in question.

The distinguishing characteristic of deductive argu—

ments has been stated as the fact that no additional

 

:fazisse
...:, IRE
- ace

 
 
 

.“w

 

 

c
c

,\
u

in
"(I

y.
“j

 

 

T

premisses can render the conclusion invalid; that is, if a

61

deductive argument is valid, then if all the premisses are
true, the conclusion must be true. This characteristic is,
of course, true of axiomatic systems. The truth insured by
the use of the axiomatic method consists in the fact that l:
Hm axioms are granted as true, the theorems must also be
granted as true. Thus, the famed incontrovertibility which
amp accountants have attributed to the conclusions derived
wiflflh a given deductive system rests solely within that
system——they are necessarily true and their contradiction
false only if one adds the further qualification ”within
that system” or ”if the axioms are true.” If the rules

of inference are not in question when an accounting argument
is supported by an axiomatic presentation, the arguer's
problem consists precisely in having his axioms granted as
true. His first desire would be, of course, to find axioms

which were obvious to everyone-—the often made request of

 

accountants to establish their arguments on ”self-evident”
truths.

The suggestion for self—evident axioms is generally
made after reflection on Euclid’s presentation of plane

geometry which separates axioms and postulates. Those state—

 

ments termed postulates were so—called because Euclid
believed them not as self-evident as his axioms, though
perhaps equally well accepted by fellow geometers. Despite
the seeming difference, it is essential to understand that

the status of those postulates ip_Euclid‘s system is exactly

 

 

62

the same as that of the axioms—statements in both groups
are appealed to in proofs of theorems and all the proofs
are considered to have equal validity. In the system,
than all must be accepted equally.

Looking at the experiences of others we find that the
establishment of theories in the physical sciences would
have been severely hampered by the requirement that all
postulated relationships be self—evident. As the scientist
has investigated matters more and more general in nature,
the desire for self-evidence in working hypotheses has
declined, and been supplanted in many cases by a desire for
merely an understandable hypothesis. And, as the scientist
has adopted methods of theory formalization even this cri—
terion has met with only limited success.

In consequence it became more and more possible
to forego an "intuitive understanding” of the ab-
stract terms and axioms and theorems accumulated
with their help. . . . The demand for Leven] an
intuitive understanding of the axioms was less and
less fulfilled when the development Eof complex
formalizations through the use of the axiomatic
method] led to the general theory of relativity and

then to qpantum mechanics, involving the wave
function.

 

The axiomatic method has become a standard practice in the
highly developed science of physics. That self—evidence
or an ”intuitive understanding” of axioms is not a formal
requirement in the presentation of a system in physical

theory should not preclude an interest in such a quality on

 

lCarnap, Foundations of Logic, pp. 209—210.

 

’18 Dart of S
..

axiomatic met

accounting en

mess is r

idem woul

Pt)

([1
rr
:5”
(D

and to 1)
mid be k
Lave mt g2
atsempts t
:3 accepti
The x
1.19. Euc
i‘lidence
late, am
tom the
amen,

Mm

 

 

 

63

the part of scientists in fields only now adopting the
axiomatic method. Such statements as ”The assets of an
accounting entity equal the equities in that accounting en-
tity," or, as Professor Moonitz suggests, "Every accounting
process is related to some entity or other,” if not self—
evident would at least gain universal acceptance. Neverthe—
less, the fact that such a requirement has, on the whole,
proved to be forbiddingly restrictive in advanced sciences
mwuld be kept in mind. That Professor Moonitz's postulates
have not gained universal acceptance should not discourage
attempts to, for example, determine the logical consequences
of accepting them.

The whole idea of self—evidence is, in fact, a relative
one. Euclid, himself, was not wholly convinced of the self—
evidence of his postulates, especially the parallel postu—
late, and strived to deduce it as a theorem of his system
from the remaining axioms and postulates. The history of
geometry shows that others were equally unconvinced, and the
non—Euclidean geometries appeared with some form of denial
or omission of this postulate as an axiom. For anyone
not trained in mathematics or geometry, the axioms of Euclid
cannot be said to be self—evident-familiar, perhaps, but
not patently obviousn And history shows that what may
at one time be termed self—evident can be, at a later time,
considered false. That the sun revolved around the earth

seemed a self—evident proposition to early astronomers.

 

 

 

hit, me use
that the axiom

'13.? 'cc accept

 

liaise ]

milling

:7“ "c.

 

 

 

64

Agahu the use of an axiomatic system does not require .
thmsthe axioms be self—evident or obvious but only that
Hwy be accepted for the purposes of the particular argument.

In accounting literature dealing with the use of
methods of deductive logic to approach accounting problems,
a desire for self—evident axioms is commonly expressed.
Riview of the vagueness of such a notion, it seems unwise,
ifrmt impossible to make such a stipulation.

If self—evident axioms are not required, it might be
suggested that the accountant-arguer turn to statements
Hw.truth of which has been established in previous deduc-
tive argumentation. But as every other scientific investi—
gator, the accountant should recognize that infinite re—
gress will result, for the truth of those other statements
likewise rests upon the truth of their premisses or axioms.
Accepting the fact that any axiomatic presentation will
bring, at some point, the question of acceptability of ax—
ioms, the advantage of initially attacking problems of
mmller scope than the whole of accounting can be seen——
the less sweeping the assumptions, the more likely initial
agreement. Likewise, the fewer the assumptions, the less
upon which to disagree.

Still, of what advantage to accountants is the use of
the axiomatic method in argumentation, if, at the starting
point of the argument the axioms must be tentatively

accepted. One of the most important benefits of using the

 

 

 

 

 

"“1th methyl _‘
:v N w. ‘-

ir emhrat ion
:.'.::25 in Accoc
inflated in l<
mired. The
general stateme
1% presented 11
ire mimit'lve
3:3: is the ac
i.e.asming umi“

a-,,:.,

...._1\ Led if

E m) 9 any

"mm

mm
Accept
W is no
we“? dc
3? the amg
is 609% no
E515 not

the Dhmsi

 

65

axiomatic method is that it enables the system—builder to
explore the consequences of his assumptions regardless of
Umir truth or their wisdom. The system—builder can find the
answer to the question "What am I committed to i£_I accept
Hus assumption?” And where are these accounting assumptions
for exploration to be found? Professor Moonitz‘s pos—
mflates in Accounting Research Study No. 1, if they can be
fonmflated in logical terms, offer some assumptions to be
explored. The whole of accounting literature is filled with
general statements about accounting which, if they could

be presented in suitable logical form, could be used as

the primitive basis of individual axiomatic systems. To

what is the accountant committed if he assumes that his
measuring unit is uniform? If it is not? To what is he
committed if he equates acquisition cost with market value
without reservation? The answers to questions such as these
nay enable the accountant to decide whether or not he wishes
to make any such assumption regardless of its absolute

truth.

Accepting statements regardless of their demonstrated
truth is not nearly so heretical as it sounds at first. The
geometer does not, after all, believe it true that the sum
of the angles of all physical triangles is 180 degrees.

He does not expect to find even one line between two points.
He is not sure if any two lines are ever parallel. Nor does

the physicist claim or observe the truth of statements about

 

 

 

 

 

 

first come:

:i‘ieziiczive arg
iEiLCii‘ie logic.
true in the pea;

:._:1 within Ch

 

The writei

system-builder

(:v
("r
(1’)
o;
)——l
S

r #-

:i‘rei considei

T153 Of any

in addi‘
iciieeiiori W
thgmm
itiepehdeht
Independenc
Trim the hi
'«izhout wh
ierhe sew
‘0 an im

is“319510

 

66

rigid bodies in free fall. Yet the axiom systems of these
scientists rest on assumptions about such things as points
andljnes, parallels and rigid bodies. Returning to the
very first comments on deduction, validity or correctness
of deductive arguments can be determined by the rules of
deductive logic. Whether the axioms of any system are

true in the real world is simply not a matter of consider—

atflniwithin that system. It is a question outside the

system entirely.

The writer has suggested only where the accountant—
system-builder might get his axioms and why they might be
selected. In the later chapter dealing with inductive logic,
other considerations will be given with respect to the

truth of any given statement.

Other Conditions of a System

 

In addition to completeness which was discussed in
connection with the scope of the problem being investigated,
it is generally required that axiomatic systems have
independent and consistent axioms and rules of inference.
Independence means, briefly, that no axioms be deducible
from the others and that no rule of inference be included
without which all theorems are still deducible. Indepen-
dence serves the purpose of reducing the number of axioms
to an irreducible minimum hence decreasing the number of

assumptions which must be granted. It is likewise helpful

 

 

 

 

"an inesiigai l.“
ZTiE‘I‘ to develop
'izieiei, ii is
“13.35.33" wish to
1'? fine indeoehd~
allies to ihte
rich for all C
The centrz
the consiste
3.15 even moi
Vile account:
iii good Peas
1 inference
LES of the a:
1;? logicians
the aXiOms E
1Gelllcians fr
he WStem
Tamils he
Obilio

T‘Equiremen

 

a “Wiper

““8 with

 

67

when investigating the system apart from its contents in .
order to develop metatheorems, or theorems about the system.

ibwever, it is not an indispensable quality. For accountants

 

’ who may wish to utilize the axiomatic method, the importance

of the independence requirement should not be exaggerated or

allowed to interfere with the construction of a system
mnch for all other purposes appears adequate.

The central requirement of any axiomatic system is that
it be consistent. This word is very familiar to accountants.
It is even more important in logic than in accounting, for
while accountants may now and then countenance inconsistency
for good reason, the logician cannot do so. Since the rules
of inference which would be preferred by accountants in any
use of the axiomatic method would certainly be those accepted
by.1ogicians, the consistency of their systems will depend on
the axioms alone. There are several descriptions given by
logicians for the consistency of a system, but in general
the system is consistent if it is not possible to prove as

theorems both a statement and its negation.

 

Obviously, accountants would want to abide by this
requirement. But it must be emphasized that consistency is
a requirement of a_giv§n system, and that it is not synony-
mous with deducibility. We might consider here the sug—
gestions of some writers that the acceptable accounting
practices (presumably after accountants utilize logic) will

be thoSe practices which can be shown to logically follow

 

 

 

 

in acceptable W

at only those 9

:szent that an E
".1150?on accou
I: is entirely
statement-in i
granite, if 0
inn-to be in
iespite the f2
ion. to foil
nan ie, the
ii?" that are
ient (perhap
hanged and
tilize cert
three (and
their logic
the usefuir
Smile sta‘
tilt) Statem

It‘ll: the la

 

68

from acceptable principles which were determined logically,
and only those practices, since those will be the only ones
consistent with the acceptable principles. Assuming for a
moment that an axiomatic system is constructed for a general
theory of accounting, this expectation is still in error.

It is entirely possible, logically speaking, for a given
statement—-in this case, the statement of an accounting
practice, if one could be given in an appropriate logical
form-—to be logically consistent with several others

despite the fact that the statement in question cannot be

wwwn to follow logically from the others. Consider, for

 

example, the practice of carrying as an asset items of inven—
tory that are damaged but salable, and a general state-

ment (perhaps called a principle) that merchandise which is
damanged and unsalable is not an asset. It is possible to
utilize certain rather complex techniques to compare two or
three (and perhaps a few more) simple statements to determine
their logical consistency outside any particular system. But,

the usefulness of these techniques is limited to relatively

 

simple statements in small groups. In the example given, the
two statements about merchandise are logically consistent,
but the latter (Damaged, unsalable merchandise is not an

asset) does not imply the former.1 The area for exercise

 

 

1The statement regarding COnsistency and implication is
supported by actual tests made by this writer which are some-
what complex to be presented here, but may be of interest to
the reader relatively familiar with uniform quantification
theory. The statements about accounting were first translated

 

alternative prac
griicipies, the
exec after the
accounting.
0n the of:
axiomatic systt
amen the nee
accounting i f
1316 of "prim.
twister} it W
is acceDtEd;
meD'iance o:
”.9 establis

iiinciples c
\

53h iWiCa

it it 5 .

~i M

ii mangle

thump :
y‘1

 

.3 not an

Mable I

 

69

of a judgment factor by accountants would remain. For, if
alternative practices are logically consistent with accepted
principles, the individual accountant must select a practice—
even after the logician had axiomatized a general theory of
accounting.

On the other hand, the development of gne_consistent
axiomatic system for a general theory of accounting wgglg
answer the need for consistency in the principles of
accounting if the theorems of that system were given the
name of ”principles.” For this development to be of value,
however, it would still be necessary that only the one system
be accepted; and the steps to its acceptance include the
acceptance of its axioms. Again, then, it can be seen that
the establishment of a complete set of consistent accounting

principles on the basis of a deductive presentation of the

 

into simple closed schemata and then tested for consistency
and implication in accordance with the techniques suggested
in Methods of Logic, Revised, by Quine (Sections 20 and 21).
The translations used are given below, with the following
dictionary.
" " is intended to mean ”merchandise”

”D” is intended to mean ”damaged”

”S" is intended to mean ”salable”

”A” is intended to mean ”asset”

”&” is intended to mean and

”D" is intended to mean only if”

"3” is intended to mean 'some

"~ " is intended to mean not”
The general statement that damaged and unsalable merchandise
is not an asset is then transcribed as follows:

)EMX & Dx & —Sx)D —Ax]

And, the statement describing the practice that damaged but
salable merchandise is an asset has been transcribed as follows;

(3x) (Mx & Dx & Sx & Ax)

 

 

 

 

 

actions pr03€<

Qtoiization
The const‘:

M
.h w a“
EEY‘LLCY‘, .eqm

323 an intende

teccne interpi
interpreted t
:i‘ the snhjec
Thinning is
(a

Tris schema
:icns’nip of
“and” ceiat
tatever tt

n

Wand:
One
1"tent at
aCQQ‘mtai
fieIns 00
EMODS

in the

 

 

70

general theory of accounting is, as stated earlier, a truly

ambitious project.

Symbolization

 

The construction of an axiomatic system, it was said
earlier, requires the selection of certain primitive notions
and an intended interpretation of them. Since the axioms
{mstulate relationships between these notions, the axioms
become interpreted as postulating relationships between
interpreted terms; i.e., interpreted, they become sentences
of the subject matter being axiomatized. For example, the
following is a schema of uniform quantification theory:

(-E’>x) (Mx & DX 80 Sx & Ax)
This schema postulates (asserts) a relationship (the rela~

tionship of being conjoined to, or in simpler terms, the

'end" relation) between notions ("M,” "D," "S," and "A,"
whatever they may be). When the notions are interpreted as
"meI’Chandise, H "damaged, H ”salable" and. ”asset" (HM, H "D, H

"S” and "A," respectively), the schema becomes a sentence of
accounting which might be read; Some damaged, salable
merchandise is an asset.

One of the most serious impediments to extended intel—
ligent dialogue between accountants is the fact that most
accountants have developed personal interpretations of the
terms common in argumentation, and these personal interpre-
tations interfere with the progress of the argumentation.

But the deductive correctness of the argumentation is apart

 

 

 

 

"in the interpr

:erzzation has, 2

:i‘ the cocrectne

'4'.- . ,_
'7
m ..>
(D D]
'0 H-
m (I)
H
90 I4

O
Q)
vs

3
: (3 (D
E) 55
cr rt ;
ID‘
(1')
.4

"I; <3 ‘4
1

(D )-c-
:5 <.
O

(J z)
'4') n:
(D

(D
‘1 0C}

:3,
,. .

questions
i: accomplish
it the constrt
isneant gene:
shapes or gro
the language
relates, in i
Syn
essenti
in matt
dissent
is to i

Again,
Pr
Cdress
Slmbol

em ti
mOGEr

 

r

 

71

from the interpretations. And, the use of logic in argu-
nwntation has, as its primary purpose, the establishment
of the correctness of argumentation.

This is the advantage of formalization, i.e., of
the separation of the calculus as a formal system
from the interpretation. If some persons want to
come to an agreement about the formal correctness
of a given derivation, they may leave aside all
differences of opinion on material questions or
questions of interpretation.

fi)accomplish this, scientists have turned to symbolization
Bathe construction of axiomatic systems. By symbolization
is meant generally the use of single letters or other
shapes or groups of such in place of the ordinary words of
the language and the introduction of variables. Woodger
relates, in his exploratory work with axiomatics and biology:

Symbolization is in no way theoretically
essential, but its merits have long been recognized
in mathematics and chemistry and it is all but in—
dispensable in a theory in which calculation
is to be performed.

Again,

Practically it is impossible to make much pro—
gress in mathematics and logic without appropriate
symbols, just as it is impossible to carry on mod—
ern trade without checks or book credit, or to build
modern bridges without special tools. . . . Symbolic
reasoning is essentially reasoning on a large scale
with instruments appropriate to such wholesale
undertakings.

There is sometimes expressed a feeling that symboli—

 

lCarnap, Foundations of Logic, p. 208.

 

2Woodger, op. cit., p. 67.
3Cohen, Preface, pp. 22-23.

 

 

zation will des‘t
presentation of
:1" logic in syn
:Lcewalne of ti
:zwetec, it wa
:e'iod ‘oy cons
asset of wise
nation on a
is not unlike
fsinwlae or n

T3996? halanw

fission. PI
amfin'lpulat
Alllty as

EYE not um-
tollZaty-‘On

3ng

 

 

72

zation will destroy an unidentified "something” in the

presentation of an accounting argument, that the utilization

 

of logic in symbolic presentations will, in fact, diminish
the value of the argumentation. Throughout this paper,
however, it was stressed that the use of the axiomatic

method by construction of a logical system was an adjunct,

a sort of visual aid, to the presentation of a particular

is not unlike the accountant's utilization of mathematical

‘ position on a controversial accounting question. As such it
I formulae or models in a discussion of some theory regarding

proper balance sheet valuations. The additional tool
enhances rather than interferes with the value of the dis—
. cussion. Professor Cohen noted, "If logic were indeed only

a manipulation of symbols it would be as devoid of scientific

1

utility as chess or tick—tack—toe.” Certainly accountants

are not unique in their initial reluctance to adopt sym—
bolization in argumentation. Perhaps as Professor Cohen
says,

The opposition to symbolic reasoning, like
the old opposition to the introduction of ma—
chinery, arises from the natural disinclination
to change, to incur trouble or expense for a
future gain. The prejudice against careful ana—
lytic procedure is part of the human. . . . Im—
patience with technique which arises from the
fact that men are interested in results and
would like to attain them without the painful
toil which is the essence of our moral finitude.2

 

lCohen, Preface, p. 22.

21bid., p. 10.

 

   

faocowntants a1
not" the power
tation should
ngnentation.
1e logic oi Fe
2? Propositions
The syster
{which the s
“swolete prc
reduce an argy
functions or
E'Iailahle foy
331d to be '11
Ellen 3991p
52:6 the 108
33539“ are
it COnStI‘u
lite the de
33Wectne:
filly manne

“313% a

 

Winn w)
alllicaby
10lie of
live wen
Bw

line We

 

 

 

 

73

H‘accountants are to obtain the fullest benefits from the
use of the powerful tools of modern logic, serious consi—
deration should be given to accepting symbolization in
argumentation.

‘Hm Logic of Functions
Of Propositions

 

The system approach was said to apply to argumentation
hiwhich the smallest unanalyzed unit may be smaller than
a complete proposition. It is sometimes desirable to
reduce an argument only to sentences or propositions and
functions or complexes of them. Deductive techniques are
available for investigating such argumentation which is
said to be in the form of functions of propositions. The
system approach has also been used by logicians to investi—
gate the logic of certain sets of such functions and these
systems are generally called propositional calculi. But
the construction of such systems is not necessary to uti—
lize the deductive techniques for determining the validity
(correctness) of argumentation which has been analyzed in
this manner. It is sometimes possible, for example, to
analyze a single sentence to be assumed and a single con—
clusion which is claimed to be implied by it. The logic
applicable in such a case is called molecular logic or the
10810 of truth functions, and is the simplest of the deduc_
tive methods.

By the methods of truth-value analysis one can deter—

mine whether the assumption of one or a series of sentences

 

 

 

 

:nth—innctionali
ianeant that to.
collision can t
i: first glance
fictional logi
since .ost argu
tense or fut
agitation of
iicient to she
:essiwe invest
Consider
:fo'olens of b
ilinion on or

A1 is

”2:53

ref

truth-functionally implies another sentence. By implication

7a

is meant that there are no circumstances under which the
conclusion can be false if the assumed proposition is true.
Msfirst glance it would seem that some aspect of truth-
flnmtional logic is what some accountants would wish to use,
since most argumentation in accounting has been presented
hiprose or full—sentence form. But, a brief review of an
application of the logic of truth functions should be suf—
ficient to show that something more is needed for any in-
tensive investigation of accounting problems and arguments.
Consider a discussion among three accountants on the
problems of balance sheet classification. Each states his

ophfion on one simple problem as follows:

 

A: As far as I'm concerned, a bank account is
- not a current asset unless it is in the
i name of the company and is not pledged as
' security for some kind of debt.

B: Let‘s be more positive about this. A bank
account is a current asset if it's in the
company's name or it is not pledged as
security for a debt.

C: You fellows are much too complicated about
these matters. A bank account is a cur—
rent asset only if it's not pledged.
Eb these three accountants agree with each other? Does each
position actually imply the others? Truth—functional logic
can help to answer these questions, and the results will be
surprising to most accountants. The positions could be

restated in the form of truth functions of the constituent

. propositions as follows:

 

 

 

”d‘stands
the
“c“stands
5801
stand
“y'stand
"3'stand
“a“ stand

“V II

\acesnlt of
steam
.t egnivaler
lilhawe the

Aiih mn hel

f‘
w

andthat

Similaritie:
“When tha
C.' the pgsj
It is
SRAOUS co
latent a;
tank, is
All a deb
aceWilla!
“Ms.
eqUiVale
WesSed

ix

 

75

A: —p v q & —r)
B: p D q v —r)
C: p D —r

where the letters and signs are interpreted as follows:
"p” stands for ”A bank account is a current asset"

q stands for ”A bank account is in the name of
the company”

"r" stands for "A bank account is pledged as

security for some debt”

”v” stands for ”or”

"3" stands for ”only if"

"—” stands for ”not"

”&” stands for ”and"

As a result of truth—value analysis of these statements, it
can be determined that the three accountants' positions are
not eguivalent. (By equivalence is meant that the positions
will have the same truth value under all circumstances.) The
position held by A implies (as defined above) those of B and
C, and that of C implies the position of B. Logically, the
similarities end right there. The position of B implies
neither that of A nor that of C, and that of A is not implied
by the position of C.

It is even more important to note that none of the po—
sitions considered express the position ”A bank account is a
current asset if and only if that account is in a solvent
bank, is in the company's name, and is not pledged as security
for a debt of some kind.” Yet this statement is the one most
accountants would say exactly expresses the actual require-
ments. Moreover, this statement, call it Position D, is not
equivalent to any of the other three. Position D may be ex-

pressed logically as:

(x) [3301(6an [SK as cnx & —<3y>(Dy & PSXYM‘]

 

yisexpcession
inionary:

an stan<
‘tl" stan
Us” Stan
"0N” stan
"D" star
”PS" star
"9" eta:
”E II Sta
3%" sta
ll_H Sta
“3“sta

his, the ab<
into English
For eve
then
the c
is m
for
lhtle traj
lltlon is (
hconntant
L05 SQODe
(that are 1
lations,
hom this

“est St;

 

1his to
dealt in

in
functio

€°rm of

 

‘]“fifir—'—__—_ ’ 444444C_________________""'ll!!

76

This expression may be translated by utilizing the following

dictionary:

 

‘ ”B” stands for ”is a bank account"

' ”CA” stands for ”is a current asset”

"S" stands for ”is in a solvent bank”
”CN” stands for "is in the company's name”
”D” stands for "is a debt”

”PS" stands for ”is pledged as security for"
”9"stands for ”only if'

”5"stands for "if and only if"

”&” stands for ”and”

”—” stands for ”not”

"3"stands for ”some”

Thus, the above string of symbols might be loosely translated
into English as

For everything you select, if it's a bank account,
then it‘s a current asset if and only if it is in .
the company’s name and the bank is solvent and it »
is not the case that it is pledged as security
for some debt.

 

Little training in logic is required to see that this formu-
lation is considerably more complex than those presented for
Accountants A, B and C, earlier. This expression is beyond
the scope of the logic of propositions alone. It involves
what are known as quantification theory and the logic of re—
lations, both of which are usually found in axiomatic systems.
From this example, it should be clear that some of the sim—
plest statements in accounting will require substantial ana—
lysis to bring out their logical form in order that they be
dealt with logically.

Although the application of truth—value analysis to
functions of propositions is, in one sense, the simplest

form of deductive analysis, it is also the most basic of

 

 

ieuuctiwe analyt
sully assumed
snorted system
rarely found us
is use requirr
:15: arguments
stole, single
The use (
functions of
again tr
”that does e
Ail—w anothe
functions 01'

CE Stated;

he01ers,
313 not a E
i510g'lcaj
another W
he mine
We 00ml
Aha
yidnld ti;
39911031
stateme

M1

or thQE

77

deductive analytical techniques. The molecular logic is
usually assumed as part of the rules of inference in con—
structed systems such as those described earlier. It is
rarely found used alone in analyses of argumentation since
its use requires that full propositions remain unanalyzed and
nwst arguments require a deeper analysis as indicated by the
simple, single statement above.

The use of the logic of truth functions as related to
functions of propositions is limited in another way. Refer—
ring again to the earlier example, the question was not asked,

"What does each position imply?"but rather,”Does one position 1

imply another?” To utilize truth functional logic with
functions of propositions, the positions to be analyzed must
be stated; for the propositional functions implied by any of
the three original positions are, in fact, infinite. Thus,
tests for implication can be made only on stated positions.
Moreover, to determine that one statement implies another

is not a pppp£_of the latter from the former as premiss. It
is logically possible to show that one statement implies
another without ever being able to derive that latter with
the former as the only premiss. This is true also in logic
more complex than simple functions of propositions.

What do these limitations mean for accountants who
would turn to logic in the analysis of argumentation? The
application of the simplest logical techniques to accounting
statements requires considerable precision in the formulation

of those statements. The use of the logic of truth functions

 

asaoplied t0 fl
Linited. Even
seems to requir
:‘f the proposit
accountants ma
1:1}; slightly
position as tr
:ernine wheth
arguments occ
:‘ther than 53‘
in whole pro
one use may
ants who hat
All their

33 the syst
shown in ty
Ely be teS
efficient

for if the

 

if anothe

‘ifidepstal

 

78

as applied to functions of propositions alone is extremely r
limited. Even the simplest statement concerning accounting
seems to require considerable analysis of the individual parts
of the propositions. For a few cases in which two or three
accountants may express their positions on a single subject in
only slightly different forms, a simple analysis using the pro—
position as the smallest unanalyzed unit of analysis may de—
termine whether the positions are equivalent. However, most
arguments occurring among accountants will involve something

other than simple statements of position, and analysis limited

 

to whole propositions is not likely to be of much value.

Some use may be found when a debate arises between two account— .
ants who have constructed informal axiomatic systems to sup—
port their theories on a particular matter. If the axioms

of the systems exhibit apparent similarities such as those
shown in the example presented earlier, mutual implication

may be tested for. But in these situations there are more
efficient methods for testing within the systems themselves;
for if the axioms of one system are equivalent to the axioms
of another, the latter will be theorems of the former and
vice versa. The primary reason, then, for accountants to
understand molecular logic is that it is usually incorporated

into any axiomatic system.1

 

 

1This fact was alluded to earlier when the comment was
made that accountants might wish to accept certain tau—
tologies in their use of deduction. Tautologies are, as
can be seen from footnote 1, page 44, expressed as truth
functions of propositions, in many cases.

 

 

Certain co

of deductive me

'17; examining ti
onions in are
l. With
porositions,
ire construct
adopting the
are sufficie'
the tool fo
Tithods avaf
only one su
Tiluirement
WA do nc

”it more ii

Alum“ .
3:03.10ny

2.
yttt‘lga

investCi

79

Conclusions
Certain conclusions can be drawn as to the usefulness
of deductive methods in attacking accounting problems simply
by examining the characteristics of deduction and its appli—
cations in areas other than accounting.
1. With the exception of simple truth functions of
propositions, all the recognized deductive methods involve

; the construction of systems. The limitations imposed by

adopting the proposition as the smallest unanalyzed unit
are sufficient to preclude this practice as the sole deduc—
tive tool for analysis of argumentation. Of the system
methods available, the informal axiomatic method seems the

only one suited to the needs of the accountant since the

 

requirements peduliar to the other methods relate to matters
which do not affect the validity of the argumentation and
are more important to the study of logic and higher mathe-
matics. Moreover, practically, the informal axiomatic
method is the most appropriate for initial attempts at for—
malization in a particular subject. The adoption of this
method, with its explicit statement of allowable inferences,
is the support needed for an accountant‘s claim that a con—
clusion which he advocates is ”implied by” or ”logically
fellows deductively” from an argument he has presented.

2. The informal axiomatic method may be used to in~
Vestigate problems varying in scope. The experience of

inVestigators in the sciences indicates that a general

L‘

 

 

:zeory involv ing
:ortinuing ideal
rosary, the mt
shod most use
ie'Tiued boundar
iioation of the
amounting, se
anonumentally
zeros of an in
A535 illdgment
amounting he
felt examine
Talidity, T
33355 in son
aibfilance 1‘
tore Main

‘33? regeam

3- Cr

the E031 0
33510 acct
lit QQnst:
hr the w
3hr the 2
practice
1.
Aethos E

80

Hwory involving the whole of the particular discipline is a
continuing ideal but not, by any means, a reality. On the
contrary, the most progress has been made and the axiomatic
method most useful when approaching problems with carefully
defined boundaries. The accountants' desire for an axioma—
tization of the whole of accounting, or a general theory of
accounting, seems likewise an ideal to be strived for but

a monumentally ambitious single goal. Review of the require—
ments of an informal axiomatic system further substantiate
Hus judgment. At this time, Very few of the theories in
accounting have been formalized in a manner which allows
their examination from the point of View of deductive

validity._ The use of the axiomatic method to formalize the—

 

ories in some particular area of accounting knowledge such
as balance sheet classification or asset accounting seems a
nwre attainable goal at this time and could prepare the way
for research on more complex problems.

3. Consistency being a quality of a deductive system,
the goal of consistent accounting principles deduced from
basic accounting postulates can be attained only through
the construction and adoption of a single axiomatic system
for the whole of accounting. The same requirement holds
for the attainment of the goal of consistent accounting
practices deduced from accounting principles.

4‘ In no subject can the adoption of the axiomatic

method guarantee the absolute truth of a given statement in

L‘

 

 

deviation. The trot-t
mic system rests e:
mm the axioms of t
meme by relying

its: the basis for i

:ree-and-ior-all-ti'
:i‘ basic concepts ‘
garzicular small a
generated imtere st
"incepts, however,
also. Similarly,

its investigation
issmptioms, and

hen the inabil:

i: the axiomatic

7~31€help account
Blue.

6. if the
tation of som
Piibiems will t
“it finding l
finding axioms

1«Twill to Dr

whee .
I‘nlhg the

81

isolation. The truth attached to the theorems of an axio—
matic system rests entirely within that system and hence
umnithe axioms of that system. This problem cannot be
overcome by relying on appeals to self—evident axioms to
form the basis for the system, for the idea of self—evidence
is, in fact, a relative one.

5. The adoption of the axiomatic method will not
once—and—for-all-time resolve disputes regarding definitions
of basic concepts or assumptions in accounting, even in a
particular small area. Its use in other subjects has
generated interest in more adequate definitions of basic
concepts, however, and this can be expected in accounting
also. Similarly, its use in other subjects has facilitated
the investigation of the deductive consequences of certain
assumptions, and this, too, can be expected in accounting.
Given the inability to determine absolute truth by resorting
to the axiomatic method, this second advantage of the method
”my help accountants to decide which assumptions would be of
value.

6. If the axiomatic method is utilized in an inves—
tigation of some particular area of accounting, the major
problems will be adequately defining the boundaries of the
area, finding primitive terms appropriate to that area and
finding axioms to govern those terms which are sufficiently
fruitful to produce the desired theorems. The various works

concerning the underlying concepts of accounting offer a

 

 

 

 

 

 

 

 

ginning point in the
3;: substantial effo
ms and axioms in
ties of inference v
users.

7. Accountant
lzgician to analyze
:emgnize that the
Tremmithout resor
SimDiest anal:

szzians is facili

82

starting point in the search for primitive terms and axioms.

But substantial efforts will be required to present those
terms and axioms in a form which can be dealt with by the
rules of inference which would be used in most axiomatic
systems.

7. Accountants who would call on the tools of the
logician to analyze and evaluate their argumentation should
recognize that the fullest benefits cannot be derived from
them without resort to symbolization to some extent. Even
the simplest analysis into the truth functions of propo-

sitions is facilitated by the use of logical symbols.

 

 

 

 

 

inductive argx

‘4 :1; se these con

c)

{rents such that
are to the orig
argument invalid.
tended to offer
the conclusi
a: inductive arg
arse, can als<
rents. The dii‘
12% grants the
is committed tz
Titer; if one
3.2% is Still
733% exact opp
The stat
“early by n
phenom f
the litmus
“Ihave See

CHAPTER IV

INDUCTION

 

Inductive arguments have been characterized in this paper
as those whose conclusions are not necessarily true; i.e., ar~
guments such that there is at least one statement which, if
added to the original premisses of the argument, can make the
argument invalid. The premisses of an inductive argument are
intended to offer good, but not conclusive grounds for assert-
ing the conclusion of that argument. Hence, the strength of
an inductive argument rests with its premisses. This, of
course, can also be said, in a sense, about deductive argu—
ments. The difference can be expressed in this manner: If
one grants the premisses of a valid deductive argument, one
is committed to acceptance of the conclusion with equal
Vigor; if one grants the premisses of an inductive argument
one is still logically free to hold a position representing
the exact opposite of that conclusion.

The status of an inductive generalization is shown
clearly by the following example from chemistry. It is a
well-known fact in chemistry that hydrochloric acid turns
blue litmus solution red. If, however, a man were to say

"I have seen the litmus solution turn red a hundred times

after the acid is added, but the next time I believe that

83

 

 

 

 

 

K
mill turn green,
*= a logical possib

e "[11
Each has said, in

Latinas between th‘
inactive angweht
The very nati
:mblete listing
argumentation, si
tilt inconclusi‘
inappropriate to
irialid, for it
2'? such an argu'
false, that the
it one time. 1
Either totally
TOPnartiai or
admit only of
Neverthe
if inductiVe
Whimsy 01‘
mm“ of n

,le refers 1

\

i
Max B

Fan) 1m.

2
Lite:

84

it will turn green," he is not contradicting himself—~there
is a logical possibility that he is right. As Professor
Black has said, ”The soundness of an inductive conclusion is

a matter gf_fact,not merely a question of the logical re—

lations between the premisses and the conclusion of the

. . I
inductive argument."

The very nature of inductive argumentation precludes a
complete listing and description of the methods of such
argumentation, since there are an infinite number of ways to
build inconclusive arguments. Moreover, it is perhaps
inappropriate to portray inductive arguments as valid or
invalid, for it can always be said, then, if the conclusion
of such an argument turns out, as a matter of fact, to be
false, that the argument was invalid while appearing valid
at one time. Deductive arguments, on the other hand, are
either totally valid or totally invalid, there is no place
for partial or temporary validity. Inductive arguments
admit only of degrees of strength. 8

Nevertheless, some accountants have advised the adoption
of inductive methods of investigation into accounting
problems, or at least mentioned induction as part of a dis-
cussion of methodological possibilities. Professor Little-

ton refers to ”Inductively Derived Principles.”2 Professor

 

 

1Max Black, Critical Thinking (New York: Prentice-
Hall, Inc., 1952), p. 304.

2Littleton, Structure, Chapter ll.

 

 

 

 

 

 

for exam?-

trader,
. . _ constitu
possibility U
about account:
1e comment seems ‘
anally followed by
fight be beneficiz
more of inducti
question the nece
motive argument
:‘ran Bertrand Ru
“allinierence :'
duction appears
:ere method of
S“itsests, at ti
“it accountin
mt by accepte
a"~"E‘umentation
‘15 “0t induct
ties an Sucl
But, to
Wild be to

\

i
SQhra

2
S ‘ Rain}
erbnepis

Mm

85

Schrader, for example, presented a paper which, he said,

constitutes an exploratory study of the
possibility that significant generalizations
about accounting might be derived inductively.1
The comment seems to imply that this practice is not gen~

erally followed by accountants, and suggests further that it

might be beneficial. With a clear understanding of the

nature of induction, the accountant—researcher might now

question the necessity and advisability of investigating in—

ductive argument any further. No less a logic authority

than Bertrand Russell is quoted regularly for his remark that
all inference is deductive” so that "what is called in— -
duction appears to me to be either disguised deduction or a
mere method of making plausible guesses.” The writer
suggests, at this point, that if what is desired in the cur—
rent accounting quest for better theory is rigorous reason—
ing by accepted rules of logic which ensures unquestionable
argumentation not open to criticism, then what is desired
is not induction in any form. Inconclusiveness character—
izes all such arguments.

But, to close the topic of induction at this point

would be to do injustice to those accounting writers who

have asked only for more thought, more careful consider-

1Schrader, op. cit., p. 645-

2Ralph M. Eaton, General Lo ic (New York: Charles
Scribner's Sons, 1931) p. 59, quoted in turn from p. 11 of
Principles of Mathematics.

 

 

 

 

aim of accounting
if we were
E of validity; we
. the "criticism
want our concl
irreproachable
pen-grounded
to consider ti
tinct from the
established .1
15,3hen, accounta
ire truth of their
for inductive arg
izuould appear t
:mclusively .
is was stat
induction is hat
rare, at times,
it‘.lhlch a pers
ddhaue discu:
drift mainta
in
”be true. (

.i h
I “e We becau

be .

 

‘UWS'ltion ‘

20ft
e
grams, 2
m belief

 

86

ation of accounting argumentation as argumentation. For,

If we were to confine ourselves to the study
of validity, we should be shirking the task of
the "criticism of thought." For in real life we
want our conclusions to be true as well as valid~
irreproachable reasoning can be no substitute for
well—grounded premisses. We are led, therefore,
to consider the ways in which the truth (as dis—
tinct from the validity) of conclusions may be
established.1

 

If, then, accountants are also interested in establishing
the truth of their conclusions, there may still be a place
for inductive argument in the study of accounting, though
it would appear that such truth will be established only in—
conclusively.

As was stated above, a listing of the methods of
induction is hardly feasible. Philosophers and logicians
have, at times, however, suggested certain general grounds
on which a person might hold a given proposition to be true,
and have discussed the question of whether such grounds
justify maintaining that given position. There are three
general reasons one might give for holding a proposition
to be true. One might say (1) I hold this proposition to
be true because it is, or (2) I hold this proposition to
be true because X said s02, or perhaps, (3) I hold this

DrOposition to be true because I know it to be so from

 

 

lBlack, op. cit., p. 249.

2Often, the appeal to authority is treated simply as
a fallacy of informal argument rather than a valid ground
for belief as will be noted later. Earlier in this paper,

 

 

 

 

 

 

 

, .
experience. The l 1r

:msidered as premi:

fair conclusion th
err statements are
position. if the <
in fact that such
also allows thef
third ground, bro:
experience, will
prized as induct
premisses relate
he conclusion-p
arguments on tin
35W tpremiss
Position taken
till be devote
”Made of bel
timed by re

to, Matti
\

W10“ was r
:ue QOIImlOn f
“S paper a
vQI‘y text 1r
duthOpity IQ
Milo“ was
SO“miles i
“tuneup

87

experience. The first two of these are not generally

considered as premisses of inductive arguments having as

their conclusion the proposition in question. But some

mmh statements are often used in support of a stated

position. If the concept of induction is expanded slightly,

the fact that such support is used in inconclusive argumen—

tation allows their discussion in this paper. Under the

third ground, broadly interpreted as the appeal to

experience, will be found those arguments generally cate-

gorized as inductive. In arguments of this type, the

premisses relate to the specific subject matter with which i
the conclusion—proposition in question deals, while in g
arguments on the first two grounds, the evidence pre-

sented (premisses) is rather, in a sense, gbggt_the pro—

position taken as a whole. The remainder of this chapter

will be devoted to a discussion of these three general

grounds of belief, the ways in which they have been

utilized by researchers in other fields of study, and

their possible relevance to the study of accounting.

 

mention was made of fallacies of informal argument which
relate primarily to suppressed premisses. A listing of

the common fallacies would serve no useful purpose in

this paper and can be obtained from any standard introduc—
tory text in logic. The selection of the appeal to
authority for inclusion in this paper in this particular
section was based upon the realization that such an appeal
E§Q_be reasonable grounds for belief in some cases, although
sometimes it may be misused and hence lead to fallacious
argument.

 

W3
iposition hel
us no actual stat
frat which does no
sitiom. At best,
nose only premis
evident." Such a
Declaration of II
We hold
that all me
endowed by
1"tents, thz
There is n
ieid to be true
3‘3 mankind. A;
"With the selec
appealing to 5
such truths '1.
times in the
evidently trl
around the e.

Self‘eV'ldenc

\

 

x lFI‘Om 4
EV") Princip
is self-exit
ttdded midd
already bee
middle in ‘
(“mules
erally :13:

 

88

Argument Based upon Self—evidencel

A position held and claimed self-evident or obvious
has no actual stated premisses, for self—evident means
that which does not need the evidence of any other propo—
sition. At best, it could be called an inductive argument
whose only premiss is "The following proposition is self-
evident." Such an argument is found, for example, in the
Declaration of Independence which states, in part:

We hold these truths to be self—evident,

that all men are created equal, that they are

endowed by their Creator with certain unalienable

rights, that among these rights are life,

There is no doubt that certain propositions have been
held to be true and even self-evidently so in the history
of mankind. As was discussed in Chapter III in connection
with the selection of axioms for a deductive system,
appealing to self-evidence, or even looking for examples of
such truths is fraught with difficulties. There have been
times in the history of man when it was believed self—
evidently true that the earth was flat and the sun rotated
around the earth. Closer to modern times, we find that

self-evidence is asserted for the proposition that two

 

1From the following discussion have been excluded the
two principles which are normally suggested by philosophers
as self—evident: the principles of contradiction and ex-
cluded middle. It was noted earlier that some interest has
already been expressed in not asserting the law of excluded
middle in some special cases. The exclusion of these two
principles from the discussion should not materially affect
its completeness, for these principles are not those gene
erally asserted by accountants as self—evident.

 

 

 

 

J.
freight lines cannot

finite. Yet no de

:maliiied truth of
:u‘rni what is now
the refusal of some
as proclaimed as E
it is equally
trident are not sy
iiies are affecter
isms self-eviden
neck was writtet
Hompany or eve
:nt'nat bank ace
:5 Self-evident
A distinct
self-evident at
life, if EVE?
33: Cannot be
in two plus
‘yn‘Wersany a
hat it is se
find Unmers;
Postulate tp
identify the
only har (11

report on e

thinly not

89

straight lines cannot cross more than once or that space
is finite. Yet modern scientists would not accept the
unqualified truth of the latter two statements. In fact,
mudiof what is now accepted belief in science stems from
the refusal of some scientists in the past to accept what
was proclaimed as self—evident.

It is equally important to note that true and self—
evident are not synonymous. It may be true that the ocean
tides are affected by the moon, but this fact is by no
means self—evident. Similarly, it may be true that a
check was written on the only bank account maintained by
X Company or even that all X Company's checks are written
on that bank account but certainly neither of these truths
is self—evident.

A distinction should, likewise, be made between
self—evident and universally accepted. Universal accept—
ance, if ever any such thing could be so and proved to be
so, cannot be equated with self—evidence. The proposition
that two plus two is four is certainly very close to being
universally accepted; yet one might hesitate to suggest
that it is self—evident. Similarly, although one might
find universal acceptance of the truth of Dr. Moonitz's
postulate that a report on economic activity must clearly
identify the specific entity involved in that activity, it
could hardly be called self~evident. The finding of a
report on economic activity which does not do so is cer—

tainly not inconceivable. The farther one‘s statement

 

 

 

 

rats from siml)1e Pb
pastion of self-ev‘f
Tao lessons ar
:ier disciplines 3
interested in furtl
first, the prospec
evident to all cor
trancement of km
to accept such pr
for universal acr
ads by the Univ
lilies are self-e
directly relate
Tilly recogniz
31“itical readey
d1Sdussing the
The authors he
statements W
it rest.
If Nope
Widens; are
further “we
Stine type 01

Case) the a

90

nmves from simple physical observables, the more likely a
mmstion of self—evidence or even of universal acceptance.
Two lessons are to be learned from the experience in
other disciplines and should not be ignored by accountants
hmerested in furthering the knowledge of their subject.
First, the prospect that propositions are equally self—
evident to all concerned is Very slight; and second, the
advancement of knowledge which may come from the refusal
to accept such propositions may be a high price to pay
for universal acceptance and uniformity. The statement
made by the University of Illinois Study Group that "postu—
lates are self—evident propositions which underlie and are
dhectly related to the accounting discipline, . . . gen—
erally recognized as valid,”1 is very significant for the
critical reader of the monograph. There is no point in
discussing the results of that group’s research since
the authors have refused to accept discussion of the basic
statements upon which the rest of the monograph is said
to rest.

If propositions which have been proclaimed as self-
evident §p§_subsequently questioned and subjected to
further investigation, their justification will involve
some type of argumentation yet to be discussed. In any

case, the argument involving simple claims of self—evidence

 

AStatement of Basic Accounting Postulates and
Principles, p. 7

 

 

I

tiers very little p

expansion of account

Argument
lgreat many 0
re assert are the I
the testimony of o
:5 such argumentat
Tillie we accept
:‘T doctors that a
tend not want tr
{tie to establisl
i: this manner .
aztomobiles and
:‘re dictionary
is are going tc
in matter
3TEdit informs
Bradstreet rep

directly frOm
\

 

d Certait
(an its tru'
ddlanch Or
I

into thfLS Ce
diarantEed t
qeffmition.

 

91

 

offers very little promise for the consolidation and

expansion of accounting knowledge.1 1

Argument Based Upon the Testimony of Others

A great many of the beliefs we hold or propositions
we assert are the result of simple inductive argument with
Hm testimony of others as premisses or support. The value
of such argumentation is undeniable. In the simple matters
of life we accept, without personal experience, the testimony
of doctors that an OVerdose of certain drugs is fatal—we
would not want to learn this first hand, nor would we be
able to establish the truth of more than one such statement
Mathis manner. We accept the testimony of engineers that
automobiles and airplanes do not run on water. We turn to
the dictionary for word meanings without any thought that
we are going to be deceived.

In matters more related to business, we accept the
credit information about X Company supplied in a Dun &
Bradstreet report without having ascertained the data

directly from X Company’s previous creditors. We accept

 

 

" 1Certain propositions are said to be self—evident when
analysis" of the proposition alone enables one to ascer-
tain its truth. A proposition such as ”Balance sheets
balance” or "Income is shown on the income statement” falls
into this category. The truth of such propositions is
guaranteed by what they say—-they are true, so to speak, by
definition. There can be no question as to the truth of
such propositions, but as noted in Chapter III, what is, by
definition, true, offers no information. And, moreover,
these types of propositions are not those generally asserted
by accountants as self—evident.

 

 

 

shot that income *-
rganiration wimo‘it
sting ccmnanv- We
tract pay our de'
:e'fcre neglected to
settle measure qr
YLETS and yardsti
in: these measurf
siandard by which
these facts and E
are on testimon;
“authority." Da
sible, without 5
The most h
not without rel
:EIQVEPT Chemi‘
Exdefinent whp
"~5st Of his 8(
Editi‘iment .
construct his
poppy} fall .1‘
ydnayym. and
it little) f
Wit Me 3C

There, Ye

Q
“Ge 13 hr

92

asiact that income taxes will be levied upon our new

organization without ever before having operated a profit~ i
mmdhg company. We accept as fact that we will be sued if
we don't pay our debt to creditor A, though we have never
before neglected to pay our debts and subsequently been
mwd. We measure quantities of inventory by means of scales,
mflers and yardsticks without having ascertained personally
that these measuring devices measure in accordance with the
standard by which we intend to abide. The acceptance of
these facts and subsequent actions thereon represent reli—
ance on testimony of others, often called an argument from
'huthority.” Daily life would be intolerable, nay, impos-
sible, without some reliance on the testimony of others.

The most highly developed of the physical sciences are
not without reliance on the testimony of others. Without
it,every chemist and physicist would have to perform every
experiment which was made to establish every general state—
ment of his science which was involved in his current
experiment. Without it, each scientist would have to
construct his own measurement tools. The basis of psychiatry
would fall if the established general patterns of human
behavior and reactions were ignored. In short, there would
be little, if any, advancement in scientific knowledge with_
out some acceptance of the reported findings and opinions of
others. Yet these scientists recognize that reported experi-

l ence is not conclusive proof. Even in the most advanced of

 

%

 

 

 

:Emes wherever ¥
anctheresults 05
gear detail so t‘na
:ius
he normal au‘
:fidhmeonthe
ithga.counsel,
2:2. Without rel
nimveto comp}
alw we produc
he wisdom
smeinstances i
lhlfiication ”
hewn appeals
tuned to an
Should be Cons

i

pa

informal a
an inapprc
““100 of at
Le” a need
Wide the
has 1“lpI‘OVe
thdpew Ban
in cOnnECt

dealing Wi

SUCCESS 01

93

sciences, wherever possible the testimony of the scientist
as to the results of his experiments will be presented in
great detail so that the experiment can be performed by
others.

The normal audit engagement is filled with examples
i of reliance on the testimony of others—letters of inquiry
to legal counsel, stock transfer agent, banks, customers,
etc. Without reliance on such testimony, the audit could
not move to completion within a period of time which would
allow the production of timely financial statements.

The wisdom of accepting the testimony of others in
some instances is beyond question; the necessity of the
qualification ”in some instances” is equally undeniable.
One who appeals to the testimony of another must always be
prepared to answer a question as to why that testimony
should be considered as adequate grounds for accepting the
truth of the stated proposition. When listing fallacies
of informal argument, philosophers recognize the possibility
of an inappropriate appeal to authority, introducing the
notion of differentiating between good and bad testimony,
i.e., a need for evaluating a reference to authority.
Consider the accountant who states that financial reporting
has improved over the past twenty years on the ground that
Andrew Barr said so. Such a remark could very well be made
in connection with some premiss in any number of arguments
. dealing with, for example, the progress in accounting, the

success of the AICPA or the SEC, or the growth of stock

%

 

 

:orsrshih There is
host‘s 9051mm or
;:;onent recognizes
21'. is well; if “Qt
information such a?
:12 Securities and
its: organization
'55. ”EC should be
trends, etc. The
recognized authoo
has. he type
bologioians is
iiil’lOl‘ity in ono
iii-no outside o
55 an appealing
foil-rights p
The testi
Villain DOSit'
l? accounting
“tested“ in
Itool‘s of a .
1“ looms
Elsatmem m
treatmentl
amount-mg

aQQQUUtant

94

ownership. There is no logical necessity for accepting the
arguer's position on the grounds he has suggested. If his
opponent recognizes this appeal to authority as acceptable,
all is well; if not, he may be required to offer further
hfibrmation such as that Mr. Barr is chief accountant with
the Securities and Exchange Commission and has been with
that organization since 1938, that the chief accountant of
the SEC should be in a position to opine on reporting
trends, etc. The example here represents an appeal to a
recognized authority on the subject with which the argument
deals. The type of fallacious appeal to authority envisioned
by logicians is the use of the testimony of a recognized
authority in one field to support a statement about some-
thing outside the field of that person's competence, such
as an appealing to statements of Mr. Barr to support a
civil—rights position.

The testimony of others as a grounds for accepting a
certain position is particularly relevant to the history
of accounting. Whatever it is that is considered ”generally
accepted" (theory, principles, practices, etc.) is the
result of a build-up of authoritative statements accepting
it. Individuals who are considering a particular accounting
treatment normally search the literature for support of that
treatment. It might be suggested that the multiplicity of
accounting practices is at least partially the result of

accountants' seeing that there is g9_logical necessity for

 

 

 

 

 

atoeoting the test it
:i' recognized exper

"ell-hoown authorit

12:12 that there a}:

:‘I inooortant under

an others suggEE
innioated. 3%:
there most exper
Laiei evidence i
use, simple ap'
expert, and the
233' agree with
till never 1%
inclusions .
if a sim'
anyway he s
for accepting
sent either
outs in art
exDerience o
olosions in
ttied when
mation has

W Weber

95

accepting the testimony or experience of others, even that
of recognized experts. Professor Paton, perhaps the most
well—known authority in matters of accounting theory, states
both that there are_”postulates" in accounting (in the sense
of important underlying assumptions) and what those postu—
lates are. Yet with this testimony readily available, some
nmmbers of the profession continue searching for postulates
and others suggest that there are no postulates in the sense
indicated. But, the history of science shows that even
where most experts have agreed upon a particular matter,
later evidence has been known to prove them wrong. In any
case, simple appeal to Professor Paton's authority as an
expert, and that of many other expert accountants who

may agree with his position and his list of assumptions,

will never logically necessitate the acceptance of these

 

conclusions.

If a simple appeal to the testimony of experts, however
many may be selected, is not completely satisfactory grounds
for accepting a given statement, its supporters might pre—
sent either the deductive argumentation used by those ex—
perts in arriving at their conclusions or the evidence from
experience used by those experts in arriving at their con—
clusions inductively. The problems which will be encoun—
tered when deductive argumentation is presented for exam~
ination have been noted in Chapter III, mainly the problem

Of proper formulation of the argument and the selection of

 

 

 

 

 

 

 

stable axioms or ‘
ice argumentation

cased in the next

13;
Arguments has
escorted by a Si
sch o‘oseryations
1:; technical expé
is :hat argument
stance to the ge
1r. experience s
sciences are th
richer of obsex
Earlier exhibi‘
iases of combi
toin ohenomenc
nioioh this oh
“accent, ca]
“Eclusive do
could he des
Mid alrea<
aThrone by
R m limi
The co
”Penance
thematic

Qdmmex ‘

96

mfltable axioms or starting points. The subject of induc-
tive argumentation on the basis of experience will be dis—

cussed in the next section.

Agggment Based upon Experience

Arguments based upon experience can range from positions
supported by a single personal observation to a series of
such observations or a group of complex descriptions of high—
1y technical experiments. The popular notion that induction
is that argumentation which moves from the particular in—
stance to the general is most related to the argument based
on experience since the general statements and laws of the
sciences are thought to be conclusions based on a finite
number of observations. The example from chemistry used
earlier exhibits this situation: it is concluded that all
cases of combining two specific liquids will produce a cer—
tain phenomenon on the basis of a series of experiments in
which this phenomenon has occurred. The fact that such an
argument, called an "inductive generalization” is not
conclusive despite the general nature of its conclusion (and
could be destroyed by a single negative instance) has been
noted already. It has also been shown, by reference to an
argument by "analogy,” that induction based on experience
is not limited to the notion of inductive generalization.

The common characteristic of arguments based upon
experience is the use of data resulting from individual
observations, whether accidental or intentional, simple or

complex. Each argument includes, as premisses, evidence

 

 

D b‘ ;. C
izsoooort 01 ”n“

cricction. Add, 11

aihr,eyidence I
:5mluve been so
ensnence as to
rdmem there a
figmwungbetwe
nmlusions oro t
hmaer,some g6
ammmnts dealic
is inductive ge
ization.1 Eacl
thwtogether
fidits past a
adlition, some
bfiioal meth
hill the thrg
till be made
no dET‘eement

m0
” argument:
\

. There
tttested y
Stet. only
EPQDOSed g
c'mliam Kn
his: 19D
Orinductj

97

flisupport of the conclusion. Evidence is the key to
induction. And, like the testimony of others, considered
earlier, evidence may be either good or bad. There are no
formulas for discriminating between the two kinds. While
there have been suggestions through the history of logic
and science as to how evidence might be gathered and
analyzed, there are no rigorous rules or laws for dis—
tinguishing between good and had evidence or for drawing
conclusions on the basis of that evidence. There are,
however, some general patterns of reasoning followed in
arguments dealing with observations from experience known

as inductive generalization, analogy and statistical general-
ization.l Each of these argument patterns will be discussed
below together with brief comments on its major problem(s)
and its past and possible future use in accounting. In
addition, some brief comments will be made about the ”hypo-
thetical method” which has some characteristics in common
with the three argument forms listed. No specific reference
will be made to a special "scientific method” for there is

no agreement amongst scientists or logicians that one method

Of argumentation occupies the central place in scientific

 

l . .

There is by no means complete agreement among logi—
Cians as to the types of inductive argument, so that any list
éuggested here will not coincide with all writings on the sub-
Ject. Only in rare, and not unchallenged cases, has an author
proposed that his listing is complete. See, for example,
William Kneale, Probability and Induction (Oxford University
Press, 1949). It is more common to suggest that the types
Of inductive arguments cannot be enumerated with accuracy.

 

 

 

included in the to

illit‘c'el.” .

Zrdoctive General

/

in argument
consists of prem‘
fact was noted i
:r.1ya sample of
clusion which i
from the premis
ations, or wha‘
we it a more
items of that

The stro

complete 01‘ a
x

l
. F01” Ex
mung Of It
dhygh make
fled by ol
M Ne
only deSCI‘i
inscientif.
o .
Wt [Lox
the method
16 be disc

he

98

investigations.1 The various argument forms proposed in
the past as representing the "scientific method” are
included in the four types selected for comment here,

however.

Inductive Generalization

 

An argument involving an inductive generalization
consists of premisses relating the evidence (that a certain
fact was noted in some instances, which of necessity are
only a sample of the total possible instances) and a con~
clusion which is of a universal form. Put in another way,
from the premisses that the observed objects, events, situ—
ations, or whatever, in a group have a certain characteristic
(be it a property or a relation), it is inferred that all
items of that type have that characteristic.

The strongest argument is that which rests on the most

complete or adequate evidence, with the latter phrase being

 

1For example, Russell has described the method as con—

sisting of "inventing hypotheses which fit the data, which
are as simple as is compatible with this requirement, and
which make it possible to draw inferences subsequently con~
firmed by observations.” (Human Knowledge, Its Scope and
Limits [New York: Simon M. Schuster, 19h8], p. 311). When
this description is coupled with his remarks about induction
quoted earlier, the ”method” of the scientists sounds rather
unscientific. Karl Popper (The Logic of Scientific Dis—
covery [Londonz Hutchinson of London, l949fl) maintains that
the method of science is a form of the hypothetical method
to be discussed later. Black (Critical Thinking), Robinson
The Principles of Reasoning, An Introduction to Logic and
Scientific Method, Second Edition [New York: D. Appleton-
Century Company Incorporated, 1930]), and Searles (Logic

and Scientific Methods [New York: The Ronald Press Com-
pany, 19 J) all refer to several patterns of investigation
as scientific methods.

 

 

 

 

from of the PTO
Sedan argument is
raid-n or" certain '
zaiion is the argu
azacenents as "Al:
ides.“ Most of
azatenents of bic
:5 argumentation
hand some of th
held prior to ti
depressions
1n the ole
sciences, collc
dole.1 Thus, 1
themes were 6
“MODS, and
5“digested. d
Characterist-
and we find
“Cheerios
mar“coals.
The mo

\
1
For
gum“) ir
“Tithe, A
mung;

99

the crux of the problem and subject to some interpretation.
&mh an argument is that reported earlier about the combi-
nation of certain liquids in chemistry. Inductive generali—
zation is the argument form which supports such scientific
statements as ”All magnets attract iron” or ”Heat expands

bodies."

Most of the elementary and primarily descriptive
statements of biology, chemistry and physics are the results
of argumentation of this kind. In this group are also
found some of the economists' generalizations, such as that
held prior to the end of World War II that wars are followed
by depressions and perhaps the profit motive assumption.

In the classificatory, defining stages of the physical
sciences, collections of observations have been most valu—
able.1 Thus, in the study of animal life, the classificatory
schemes were established on the basis of innumerable obser~
vations, and categories such as fish, bird and man were
suggested. As observations became more sophisticated, the
characteristics of each category were carefully established
and we find redefinition leading, for example, to the
reclassification of bats and whales to the new category of
mammals.

The most serious problem of which to be aware when

statements are to be established by the "leap" from partic—

 

 

1For some interesting commentary on the role of in-
duction in definition formulation see Georg Henrik Von
Wright, A Treatise on Induction and Probability (London:
Routledge and Kegan Paul, Ltd., 1951), Chapter 6-

 

 

in instances to 2
gleteness of the e
inconplete evidenc
as, present and
meet be examine
callable, the at
erase of that do
32-? this reason
accumulate evide
bias. hsimple
is the conclusi
:‘thitude of 01
African tribe.
Smilbic locatj'
Eiidenee must
which COUtl‘ac‘
'Iiew, 00mm“:
for inductiv
amm‘fla‘Cior
been consid.
ldtludeS 6X
Edgatm '1,
Dim Eehe

induethe
Mutiny (

0? bids f

The

 

100

char instances to a universal generalization is the incom—
pleteness of the evidence. To be sure, the problem of c
incomplete evidence cannot be avoided since all instances——
past, present and future~—of a universal generalization
cannot be examined. (If, in fact, all the evidence were
available, the argument would not be induction at all but

a case of that described earlier as perfect induction.)

For this reason it is necessary to make every effort to
accumulate evidence which, though incomplete, is without
bias. A simple and commonly used example of biased evidence
is the conclusion that all people are black based upon a
multitude of observations by the members of a primitive
African tribe. Here the evidence is biased by the geoe
graphic location of the observers. More basically, the
evidence must not be biased by the exclusion of instances
which contradict the general statement being made. The

View, commonly attributed to Aristotle, that the evidence

for inductive generalizations consists solely in the simple
accumulation and enumeration of positive instances has long
been considered erroneous, and the evaluation of evidence
includes explanation of what might be called initially
negative instances. Unlike the yet to be discussed Statis—
tical generalization,a single negative instance renders the
inductive generalization false-hence the need for close

scrutiny of apparently negative instances and the elimination

Of bias from the accumulation of favorable instances.

There appear to be few general statements in accounting

 

 

 

 

:reory which might
non observations (
idly accountants
idea of double-em
heir observation
:eeedhad been an
ation of duality
otters to be ac
'33 so, the press
Pension liabili‘
to: observed, 0
hobonents of 1
negative insta
hi all busing
That ”the cri‘
accounting sy
h continue 2

the Other ha
\

n 1PPofe:
”2m the ba‘
Foe necESSa
out of the
ht

J
they}, EVem
have this .
arm the 0
the evider
hased.

c Ray!
“honours
9‘ T.

r_l

101

 

theory which might be termed inductive generalizations based
upon observations of business facts. It is possible that
early accountants followed this line of reasoning when the
idea of double—entry bookkeeping was originally conceived.
Emir observations that the business facts they had encoun—
tered had been analyzable into two parts, the accounting

notion of duality, may have led to the conclusion that all

 

matters to be accounted for could be analyzed in this manner.
If so, the present controversies over the recording of
pension liabilities and long—term leases, business facts,

not observed, observable or even imaginable by the original
proponents of the double—entry system, may be cases of
negative instances which falsify the generalization regard—

ing all business facts.l

Mr. Dein, in fact, once observed
that "the criteria for the admission of data into our
accounting system has been whether we are ingenious enough
to contrive an equality of debits and credits.”2 If, on

the other hand, the owners‘ equity of the balance sheet

 

lProfessor Schrader (”An Inductive Approach”) suggests
that the balancing feature of an accounting system may be
the necessary product not of the duality of business facts
but of the fact that measurements are made. He suggests,
on p. 647, that ”if balancing is a condition required of
every event admitted to the record, the aggregate must still
have this characteristic.” While his conclusion differs
from the one expressed by this writer, he, too, notes that
the evidence upon which the conclusion is based may be
biased.

 

2Raymond C. Dein, ”A Glance Backward at Research in
Accounting," The Accounting Review, XXXVI (January, 1961),
p' 70

 

 

 

 

isceally a residue
system, then, ensu‘.
determined as affe
{looted in the SB“
Outside of t‘
share are few cas
:f the use of arc
:ooniversal gen
says,1 economic
units or entiti
economic activi
and not of some
ZCIlV'l'Ey being
generalizatioc
statements of
as "us finar
chose statem.
cations that
identify the
meoncepti
litre EEnerg
hemmed" ‘
like1v the
1s rmm ts

Should be
\

M01

 

 

102

is really a residual notion, the mathematical nature of the
system, then, ensures that any information which can be
determined as affecting assets or liabilities must be re—
flected in the system.

Outside of the assumption of the double-entry system,
there are few cases in the literature of accounting theory
of the use of arguments leading from individual observations
to universal generalizations. That, as Professor Moonitz
says,l economic activity is carried on through specific
units or entities, is the result of the definition of
economic activity which requires that something be acting
and not of some finite series of observations of economic
activity being carried on. The usefulness of inductive
generalization as an argument for establishing general
statements of accounting theory which are normative, such
as ”All financial statements should identify the entity
whose statements they are” appears severely limited. Obser-
vmfions that this, that or the other statement should
identify the entity involved are biased by the observer's
preconceptions of what ought and ought not to be done. A
more general statement such as ”All assets should be
recorded" suffers the same difficulty. Moreover, it is un-
likely that such a statement would even be argued inductive—
ly from the premisses that this, that and the other asset

should be recorded, but rather from some more general state—

 

 

 

 

lMoonitz, Postulates, pp. l3ff.

 

 

 

 

 

 

zensregardihg the
:e'ationship of 885‘
The University
appears to be a uni
rinsipal characte'
innonetary terms.
iasis of evidence
:reracteristics o
statements as ass
in evidence, to:
evidence are the
Eossibllity of c
instances which
int the instan
One way '11
hhhh about b
accounting the
“h- That to
economist doe
”steam; the e(
ties? that
hhfle 311 d6
13 interest.

Q1“his it

\

l
m
2

m

103

ments regarding the objectives of the recording and the
relationship of assets to those objectives.
The University of Illinois Study Group suggests what

appears to be a universal generalization relating the

prhmipal characteristics of assets.
1

”They are measurable
flsmonetary terms.” This conclusion is reached on the
basis of evidence which seems to consist of noting the
characteristics of those items which are listed in financial
statements as assets. Here, too, there is bias inherent in
the evidence, for the only assets which are included in the

evidence are those which have been measured. There is no

possibility of observing a negative instance in a set of

instances which has, as a requirement for set membership,
that the instance be a positive one.

One way in which observation, analysis and generali-
zation about business or economic facts may be of use to
accounting theorists is in the area of definition improve—
ment. That the accountant deals with the same data as the
economist does not automatically force the accountant to
accept the economists‘ way of looking at things. Copi
notes2 that though the librarian, bookbinder and biblio~
phile all deal with the same basic material——books——each
is interested in that material in a different way and will

classify it according to his interest. Accounting and

1A Statement, p. 16.

QIntroduction, pp. 462ff.

 

ennomics, while dee
derail, the same
:‘nssification syst
it totally appropi
in accountant. P
ids sort when he
reserved for perso
:c‘profit‘ are t
concept when appl
sifloations and/o
resistance-as c‘
hhbrt can eas
sent definition
Classifications

h inductively

as well as thC

A“ hrgum
mathis the
them) eye
moon to be
is exhihlte
that the 0t
soothe? ha;

\

1
P0st

\

the

104

 

economics, while dealing with the same material, are not,

after all, the same thing; and it is possible that the i
classification systems and definitions of the economist are

rmt totally appropriate for the purposes and interests of

the accountant. Professor Moonitz suggested something of

this sort when he proposed that the term ”income” be

reserved for personal income and that ”The terms 'earnings'

or 'profit' are then available to describe the related

1

concept when applied to accounting entities.” New clas—

sifications and/or definitions may meet with some

resistance-—as did Professor Moonitz's suggestion——but

 

support can easily be found for the belief that some pre—
sent definitions are not entirely adequate. While new
classifications based on inductive generalization would
be inductively argued, they might be supported at least

as well as those presently in use.

Analogy

An argument involving analogy consists of premisses
relating the evidence of noted similarities between two
objects, events, situations or whatever, which are not
known to be identical, a premiss that another characteristic
is exhibited by one of these objects, etc., and a conclusion
that the other object has that characteristic also. Stated

another way, from the premisses that two things are alike

 

lPostulates, p. 14.

%

 

 

igsowe rest)eCtS it
rather. Eaton (1'13“
arcinooctive gene?
iaezindoctive gene
. , . proceed
olars of a g1
marticula
iron statemer
particulars 1
those partic'
telegical argume
arc scientific ll
research where e
in who have so
initial. The (:01
1'1“. experiment
Simon to both
Biologists haw
35 life on ott
iih'nysics, a,
tent of a wav
tisatiation
The MO
fished by me
7% need f0
itiepant dl

WES of at
\
1

Rate

2
Fror

105

flisome respects it is inferred that they are alike in
mwther. Eaton distinguishes between argument by analogy
and inductive generalization through enumeration by stating
Umt inductive generalization

proceeds from statements about some partic—
ulars of a given class to a generalization about
all_particulars of that class; analogy proceeds
from statements about some properties of given
particulars to statements about other properties of
those particulars.

 

 

 

Analogical argumentation is common in both everyday life
and scientific investigation. It forms the basis of medical
research where experiments are performed with nonhuman ani—
mals who have some characteristics in common with the human
animal. The conclusions of arguments stemming from success—
ful experiment propose that successful treatment will be
common to both the research animals and the human animals.
Biologists have argued by analogy about the possibilities
of life on other planets of the solar system. In the history
of physics, analogical argument played a part in the develop~
ment of a wave theory of light through an analogy with the
organization and structure of the solar system.2

The problems encountered when conclusions are estab-
lished by means of analogical argument center around
the need for selecting relevant similarities and recognizing
relevant dissimilarities. In medical research, different

types of animals are used for different types of experiments

 

 

 

lEaton, op. cit., p. 555.
2From examples by Herbert L. Searles, op. cit.

%

 

 

 

iepemiing on the me
argument that the i
11‘ mother did, the
{trough the same b
in companies are
ration. Unfortun
Lite that of adec
:‘re decision as i
is material com
sick is not get
‘1'? length. Yet
sell dimension
:easuring devi<
possibility ex
foe measuring
:eut.1 Thus,
iitions which
EKDEPiment’ 1.
hitting he Cc
nWitty be 6

The "as

mt tfansac

L“. agcwntj

Wording <
\\
i
1 This
lingth 0f
“‘3“ York

106

depending on the medical problem to be attacked. To an
argument that the price of one stock will fall as the price
of another did, the fact that the two stocks were obtained
through the same broker is hardly as relevant as that the v
two companies are in the same industry or geographical lo—
cation.. Unfortunately, the notion of relevant similarity,
like that of adequate evidence, is subject to interpretation;

the decision as to what is relevant is a subjective one.

The material composition or cleanliness of a measuring
stick is not generally considered relevant to a measurement
of length. Yet when dealing with measurements of extremely
small dimensions, the absence of dirt particles on the
measuring device becomes an important condition; and the
possibility exists that in some cases the composition of
the measuring stick may be a relevant factor in the experi-
ment.1 Thus, though the scientist may suggest those con-
ditions which he considers to be relevant in his particular
experiment, his report may still include comment on con—
ditions he considers of doubtful relevance lest they subse—
Quently be determined to be important.

The "as if" approach to the handling of some account—
ing transactions is an example of the analogical reasoning
in accounting. This approach is seen, for example, in the

recording of stock dividends and fixed asset acquisitions

 

 

1This suggestion comes from a study of the concept of
length of P. W. Bridgman in The Logic of Modern Physics
(New York: The Macmillan Company, 1928).

%

 

 

ct involving cash

rated, and the rec
:eristics of such
men accountants

business fact and

audit so, in whz
of such argument
hthur Andersen

of long-term le
between such ob
he argument 31
The s

the concl

Other for
acquires
rentlp w

”Lid CODCiudes
Mid in the

Same discuss
TEQogniZ-mg
tithtatmn a,
0990hents O

Often draw

hints for

hi'thur Am.

of that a

 

 

 

107

not involving cash. Similarities to cash transactions are
noted, and the recording is effected based upon the charac—
teristics of such transactions. Analogy is likewise used
when accountants are confronted with a new situation or
business fact and question whether it is to be recorded
and if so, in what manner. An excellent example of the use
of such argumentation is found in the discussion of the
Arthur Andersen & Co. position on the accounting treatment
of long—term lease obligations.l An analogy is drawn
between such obligations and conventional types of debt.
The argument states, in part, as follows:

The similarities between leases and debt support
the conclusion . . . that leases are frequently an-
other form of financing under which the lessee
acquires an important property right . . . concur—
rently with the creation of a fixed obligation.2

and concludes that such leases should, therefore, be re—
corded in the manner of other financing obligations. The
same discussion offers an example of the importance of
recognizing relevant dissimilarities in analogical argu—
mentation and the subjectivity of the concept of relevance.
Opponents of the recording of lease rights and obligations
often draw an analogy between lease agreements and commit-
ments for future operating expenses. The authors of the
Arthur Andersen discussion attempt to falsify the conclusion

of that analogy by pointing out what they consider to be

 

1Accounting and Reporting Problems of the Accounting
Profession, Second Edition (October, 1962), pp. 27ff.

“-

2Ibid. p. 31.

 

 

 

relevant dissimila'
being compared. rI
analogical argumer
in those who disa
of the similariti
Whether anai

for support of m
tum as the ear
identify the 210
example, no inv
similarities he
WOW; astro
5“treat that 5
have been 1001

the meme 0

accounting si

Mr

A“ argt
consists of
tongue is
smiths 0f
concmsion
or Perhapp
or evehts

ence Carr

\

10a:

108

 

relevant dissimilarities between the two types of things
being compared. That the firm's position is also based upon
analogical argument renders it equally subject to question
by those who disagree with its estimate of the relevance

of the similarities contained therein.

Whether analogical argumentation offers any possibility
for support of more general statements in accounting theory
(such as the earlier example that activity reports should
identify the acting entity) has not been explored. For
example, no investigations have been made to detect possible
similarities between the data of accounting and that of
history, astronomy or biology. The writer does not wish to
suggest that such similarities exist, but only that none
have been looked for; and the search is made difficult by
the absence of an authoritative collection of those general

accounting statements in need of some support.

Statistical Generalization

An argument involving a statistical generalization
consists of premisses relating the evidence that a charac—
teristic is found to some extent in a sample or several

samples of the total possible objects, events, etc., and a

 

conclusion that this characteristic is to be found to some
or perhaps the same extent in an examination of such objects
or events. Closely akin to this argument form is the infer—

ence Carnap calls predictive or external,1 which involves

 

lCarnap, Logical Foundations.

 

%

 

 

ice conclusion thz

Conclusions
iration may, but

associated with

aproportion of
iavliie, state
. . Ill 01’ "MOS
from argumenta
:cre precise i
Leading to co
in: not alwap
Likewise, mar
laws Father
Some dj
haiifier "r
aBeets from
Examined we
hat a larv
idttep is
El kind of

Tilt]
\

1313

 

 

 

the conclusion that a characteristic noted in one or several
samples will also be found in another sample not overlapping
the first.

Conclusions of arguments involving statistical general—
ization may, but need not, indicate the precision commonly
associated with the word "statistical." Black forms the
conclusion of such an argument as ”Most (or, such and such

1

a proportion of) cases of A are also cases of B.” In every—

day life, statements of the form ”This usually happens when

H H

or ”Most of the time when represent conclusions
from argumentation involving statistical generalization. In
more precise terms, medical science has produced arguments
leading to conclusions such as that pennicillin is sometimes
but not always useful in treating numerous types of infections.
Likewise, many of the laws of modern physics are statistical
laws rather than universal statements.

Some distinction can be made between two uses of the
qualifier "most.” To conclude that most bank accounts are
assets from the premisses that the several bank accounts
examined were assets is not the same as from the premisses
’that a large percentage of those examined were assets. The
latter is a statistical generalization. The former involves
a kind of weak inductive generalization.

The major problem in establishing a conclusion involving

 

lBlack, op. cit., p. 327-

 

109 ‘

 

 

 

: statistical gene

'1
(D
'0.

Institutes a
generalized about
is instances exz
irdnctive argume
discussion of it
Tit evidence mu:
question relati
enanined. in i
"for relevant s
similarities t
:aciveness of
sample is of
itneralizatic
the conclusi.
in analogy j
the QOllClus:
is, in a se
his Duint
Eighty pep.
tetheen ill
randomly l
is not Sll
exteriem
Samills 0
which in

me he:

 

*

110

a statistical generalization is the need for evidence which
constitutes a representative sample of the objects to be
generalized about. The problem of representativeness of
the instances examined is shared, of course, by all the
inductive arguments mentioned. In connection with the
discussion of inductive generalization it was noted that

the evidence must not be biased, which is equivalent to a
question relating to the representativeness of the instances
examined. In the discussion of analogical argument the need
for relevant similarities was noted in order that accidental
similarities be excluded as evidence to ensure the represen—
tativeness of the sample. But the representativeness of the
sample is of particular importance to the statistical
generalization. A single case which does not correspond to
the conclusion of an argument by inductive generalization

or analogy is sufficient to falsify that conclusion, but

the conclusion of an argument by statistical generalization
is, in a sense, harder to falsify. An example will make
this point clear. Evidence may indicate that approximately
eighty per cent of the machines of a certain type last
between nine and ten years. Additional evidence that a

randomly selected machine of that type lasted eight years

 

is not sufficient to falsify the conclusion established by
experience. If the original evidence was based on a random
sample of machines of that type, another random sample
Which indicated that the machine life was between eight and

nine years would not be sufficient to falsify the previously

 

established concl

norobability the

other hand, with
evidence which c
Liam that t
The notion
cably associate
however inform:
ever possible
theory in app]
sample) in su
generalizatic
aence that u
5° analogs.
nation, no]

in the fina

\

lCarng
(if an hypo
to the Tel
011 things
teng'mnal
Telat'we .
argument
releyame
not'mn‘
and a Dr:
applicab-
Other ill
W
. “Emails
lty t0 \
Diicate‘
Wed ha
pPOpOSe

lll

established conclusion. The second sample would produce only
a probability that the first conclusion was false. 0n the
other hand, with an inductive generalization or an analogy,
evidence which contradicts its conclusion produces a
certainty that the conclusion is false.

The notions of sampling and probability are inextri—
cably associated with arguments by statistical generalization,
however informal. Scientists who follow this method, when—
ever possible utilize the concepts of modern probability
theory in appraising the acceptability of the evidence (the
sample) in support of a conclusion stated as a statistical
generalization. However, neither scientists nor logicians
agree that the same concepts of probability are applicable
to analogy, inductive generalization and statistical general—
ization, nor even which concepts apply to which arguments.1

In the final analysis, the applicability of modern statis—

 

lCarnap, for example, (Logical Foundations) proposes

that one concept of probability relates to the probability
of an hypothesis on a given evidence statement and another
to the relative long—run frequency of ”one property of events
or things with respect to another." He suggests that no ex—
tensional concept of probability, dealing as it must with
relative frequencies, will be appropriate for analogical
argument since the strength of the analogy rests with the
relevance of the similarities which is an intensional
notion. Distinction is made elsewhere between the frequency
and a priori theories of probability with the former more
applicable to statistical generalization and the latter to
other inductive conclusions. P. Coffey, in The Science of
Logic, Vol. II: Method, Science and Certitude (London:‘*—
Longmans, Green and Co., 1918), suggests that the probabil—
ity to be attached to analogical argument is further com—
plicated by the recognition that the relevant similarities
used have only a probable connection with the similarity
proposed in the conclusion. (pp. 15uff.)

 

 

Jul methods mus .

" ' Y‘
c the particula-

Though not re
reactivity pert
:eneralization is
a: ted. An audi‘
rents is the con
experience (evic
2;: similar argu
rent, the concl

statement of ti

reasonable for

acomment is i

nnic‘n rests 0

invoical cour

auditor‘s are
he notions

it interpret

if the audi

Statis
EStalinist
T"linen oi‘i

tiltlvtsemel
\
1PI‘E
testing
“than.

r

112

 

tical methods must be determined on the basis of the facts
of the particular problem being investigated.

Though not related to accounting theory, in at least ‘
one activity performed by accountants the statistical
generalization is commonplace, if perhaps not explicitly
stated. An auditor‘s expressed opinion on financial state—
ments is the conclusion of an inductive argument based upon
experience (evidence). That evidence in turn is arrived at
by similar argument. For example, in a normal audit engage-
ment, the concluding work paper on inventory includes a
statement of the auditor‘s position that the inventory is
reasonable for inclusion in the financial statements. Such
a comment is the conclusion of a long and detailed argument
which rests on premisses relating the results of tests of
physical counts, footings,extensions and pricing. The
auditor's arguments are inductive and hence inconclusive.
The notions of "materiality" and ”fair presentation” can
be interpreted as recognition of the inconclusive nature
of the auditor‘s arguments.1

Statistical generalization as a form of argument to
establish significant general statements in accounting
theory offers only as much promise as those conclusions

themselves offer. Generalizations such as that "Most assets

 

lPresent trends in the field of auditing to include
testing procedures which embody the elements of modern
statistical theory reflect further recognition of this fact.

 

1 1| .
nerecorued of e
1” n:
werecordeo (wn.

:tserration) are

be night questin
if they were use
accounting posts
the group of ge
2f the goods at
trough exchan
producers.“ i
hires no rep
5? aspears to
sent involvil
conclusion f
500% staten
all argument
cualiiicatj
Statistic?
itSSop M00
regarding
(dices, h

in hlS m
Fetannin

all adore

Estable

buted i

113

are recorded" or even that ”Eighty per cent of all assets
are recorded" (which might be supported by research and
observation) are not generally found as premisses for
further argumentation in questions of accounting theory.
One might question the fruitfulness of such generalizations
if they were used. Professor Moonitz proposes as a basic
accounting postulate (which this writer would classify in
the group of general statements of accounting) that ”Most
of the goods and services that are produced are distributed
through exchange, and are not directly consumed by the
producers." Presumably, he feels this generalization re-
quires no report of the evidence which supports it though
it appears to be in the form of the conclusion of an argu—
ment involving statistical generalization. Granting this
conclusion for the moment, the important thing to note
about statements such as this one is that the conclusion of
any argument using them as premiss is subject to the same
qualification; that is, it must take on the nature of a
statistical generalization also. If, for example, Pro—
fessor Moonitz wishes to further argue to some conclusion
regarding the accounting for goods and services at exchange
prices, he is subject to the limitation of the qualifier
in his premiss unless further premisses are offered
regarding the accounting for things not exchanged. Perhaps
an additional analogical argument might be proposed to
establish, albeit inconclusively, that things not distri—

buted in exchange have exchange prices. Or it might be

 

 

Enpothesized that

through exchange 2
assumptions, some
in argument is nc
lnlhapter 111.
the truth of the
generalization ,

for it.

A method
ill ConnectiOn
Eui'agtsting an
and ascerta'u
insures are
Tnosa Staten
shanty are
istated as
‘lith a for
varying in
other) can
of “pme
Can be oh
not pom
about me

Expense,
\

1v

114

hypothesized that all goods and services are distributed
through exchange and from this, coupled with some other
assumptions, some conclusions may be deduced. But such
an argument is not inductive but deductive as discussed
in Chapter III. The truth of the conclusions rests with
the truth of the premisses, one of which is the original
generalization, and its truth still rests on the evidence

for it.

Hypothetical Method

A method of scientific investigation often discussed
in connection with inductive argumentation is that of
suggesting an hypothesis, deducing some of its consequences
and ascertaining whether the hypothesis and those conse-
quences are in accordance with subsequent observations.
Those statements which are implied by Newton’s law of
gravity are consequences of such a method. The law itself
(stated as follows: Every body attracts every other body
with a force which is proportional to its own mass but
varying inversely as the square of its distance from that
other) cannot have been established by observing instances
of "forces," for "forces” are not the sort of things that
can be observed. Movements of objects can be observed, but
not forces. What Newton proposed were certain hypotheses
about motion, and from these are deduced ”the motion to be

expected of bodies in various kinds of situations.”1

 

lWilliam Kneale, op. cit., p. 100.

 

 

 

 

then the postulate
observation, the
tation in genera“.
that an hypothes
argument. To tr
We, no finit
hypothesis as t
Often, tht
test by direct
are made to e:
quences deriv
Consequences
are tested a
there is sai
1“this sity
bEtnnes 1m
Etarantees
must be tr
comuS'lOt
“en‘know
“affirm
problem;
fines not
“thee, 4
or a Da

0? in;

115

When the postulates themselves are subject to test by
observation, the evidential problem of inductive argumen- ;
tation in general presents itself. Suppose, for example,
that an hypothesis were suggested by means of analogical
argument. To the extent that the.hypothesis is general in
scope, no finite amount of evidence can establish that
hypothesis as true.

Often, the postulates themselves are not capable of
test by direct observation (like those of Newton). Attempts
are made to establish such postulates by testing the conse—
quences derived from them or trying to arrive at these same
consequences by inductive argument. If the consequences
are tested and found to be in accordance with observations,
there is said to be support for the hypothesis also. But
in this situation, a peculiar characteristic of deduction
becomes important. While a valid deductive argument
guarantees that if the premisses are true the conclusion
must be true also, there is no such guarantee that true
conclusions cannot be deduced from false premisses. A
well-known fallacy of deductive argument, usually called

' involves essentially this same

"affirming the consequent,’
problem; affirming the consequent of a conditional argument
does not imply the affirmation of the antecedent (hypothesis).
Hence, to establish, however strongly, that the consequences

of a particular hypothesis or set of hypotheses are true,

or in accordance with some finite amount of observational

 

 

 

 

evidence, does no
the hypothesis .
which are in ace
guarantee that t
vhioh would prov
supporting the
the hypothesis
the iudgment o
The relevance
judged. The
be deduced th
relationship
mints previc
must he (let
vhhehs have
for the em
itself,
ThusJ
“when“
matters c
(Emma
homer
com;Pad
ital me
in the

begun

 

ll6

evidence, does not necessarily provide equal evidence for
the hypothesis. Nor does the ability to deduce conclusions
which are in accordance with some observational evidence
guarantee that the hypotheses used are the only hypotheses
which would produce these conclusions. Whether the evidence
supporting the consequences is of importance in establishing
the hypothesis depends on facts outside the realm of logic;
the Judgment of the scientist is an important factor here.
The relevance of the hypothesis to the conclusions must be
Judged. The availability of other hypotheses from which can
be deduced the same consequences must be considered. The
relationship of the hypothesis to other scientific state-
ments previously investigated and tentatively established
must be determined. All these considerations and many
others have led the scientist to attribute the evidence

for the consequences of an hypothesis to the hypothesis
itself.

Thus, the hypothetical method, which, like the inductive
arguments mentioned, involVes eventually a reference to
matters of observation and evidence, faces the same general
difficulties as those arguments. The truth of the hypothesis,
however tested, does not necessarily follow without fear of
contradiction or qualification. In addition, the hypothet-
ical method is troubled by the same difficulties involved
in the use of the deductive methods as noted in Chapter III.

Despite the multiplication of limitations, the method of

 

 

 

 

hypothesis, deduc
valuable in scie‘
overeative imag
implications cam

its;
A1

rthis brie
ductive argume
the usefulnes
accounting pr
1- Modv
evidence as
as true. rm
Group may b
Grove to u
bummed
determined
be the re
ment, or e
truths i]
accoblnta
Statemeh
Saul s
”turner
attemp

If the

117

hypothesis, deduction and test is considered extremely
valuable in scientific investigation, with observations

or creative imagination suggesting hypotheses whose logical
implications can be explored and possibly tested.

Aspects of Induction Significant for
Argumentation in Accounting Theory

 

 

This brief examination of the forms of individual in—
ductive arguments suggests several comments relating to
the usefulness of inductive argumentation in handling
accounting problems or constructing accounting theories.

1. Modern scientists have refused to recognize self-
evidence as adequate grounds for accepting a proposition
as true. Though the postulates of the Illinois Study
Group may be true, or accepted as true, the claim of that
Group to the self-evidence of the postulates is neither
Justified nor appropriate. Rather, self-evidence has been
determined a relative notion. Apparent self—evidence may
be the result of pre—conditioning the mind through environ-
ment or experience. The lack of an abundance of self—evident
truths in the more exact sciences should indicate to the
accountant-theorist that the likelihood of finding general
statements of accounting which could be said to be univer—
sally self-evident is very slight. And, this method of
argument offers little hope for the accounting theorist
attempting to expand or consolidate accounting knowledge.

If there are postulates or principles which are inductively

 

 

 

determined, they
2, Scientis
im the course of
the reputation E
they are relyint
testimony of ot
the progress 01
that any appea
amd the discus
of the author
in argument,
mists or pr:
extent that
the present
has not bee
1“ bractice
Bulletin 1m
deferred f
Would be h
accountin
he or at
that fut
resillts.
3.
stateme

1131111311

118

determined, they are most certainly not self—evident.

2. Scientists often rely on the testimony of others
in the course of their work, but take into consideration
the reputation and competence of the authority upon whom
they are relying. Failure to rely to some extent on the
testimony of other investigators could completely stifle
the progress of science. Still, the scientist is aware
that any appeal to authority is open to further question
and the discussion may finally be reduced to an examination
of the authority's basis for his conclusion. In an account—
ing argument, an appeal to the authority of accounting the—
orists or practitioners would be valuable only to the
extent that the testimony is accepted as sufficient. To
the present time, simple reference to accounting experts
has not been sufficient to preclude differences of opinion
in practice. Reference is made to Accounting Research
Bulletin No. 43, Chapter 10b, in support of recording
deferred federal income taxes. Undoubtedly, this Bulletin
would be considered the testimony of highly respected
accounting experts. Yet the controversy over the record—
ing of deferred taxes still exists. This writer believes
that future references to authority will produce the same
results.

3. The most common misunderstanding of scientific
statements, laws and theories is that they represent cer—

tainties. Professor Flanders, for example, notes "Science

 

 

is also a certain
is. ilvin R. Jenr
tical sciences a
the "laws" of sc
in Accounting Rb
supposedly unch
“luvs" of accov
explanation of
science would
iiven time wa
it Short is
learned men <
mouth his
aI‘gvment the
illusion (p
the number
finality b:
“my ate
000cmSive

subseiluen

\

ing) :22:

1961), p
2

”my ’ m E

31)
at
LEnglev
is 112

 

119

is also a certain body of knowledge, the knowns of science.”1
Mr. Alvin R. Jennings speaks of the principles of the phy—
sical sciences as ”immutable laws."2 Failure to understand
the "laws" of science is also indicated by Professor Moonitz
in Accounting Research Study No. 1 when he seeks to contrast
supposedly unchanging facts and laws of astronomy with some
'Haws" of accounting, the latter being, for him, ”merely an
explanation of behavior at a particular time.”3 Of course,
science would not progress very far if what was known at a

given time was considered the last word on the subject.

The story; is told of Galileo having invited some of the

 

learned men of his time to observe the moons of Jupiter é
through his telescope. Some refused his offer with the
argument that if they did see anything, it must be an
illusion (perhaps with the assistance of the devil), since
the number of planets had already been determined with
finality by astronomy.4 The contention regarding the cer—

tainty attached to physical laws has been consistently and

 

conclusively refuted by the number of such laws which have

subsequently been discarded. In contrast to the status of

 

lDwight P. Flanders, "Accountancy, Systematized Learn—

ing, and Economics," The Accounting Review, XXXVI (July,
1961), p..568.

 

2"Accounting Research," The Accounting Review, XXXIII
(July, 1958), pp. 547—554.
3pp. 6 and 7.

4W. Ward Fearnside and William B. Holther, Fallacy
(Englewood Cliffs, New Jersey: Prentice-Hall, Inc-. 1959),
P. 112.

 

 

the laws of maths
“those of all 0t
and in constant
covered facts . "j
Nor have s
such that it 0;
attached to th
established b1
Chemistry, F
thephysical
objective, 1
be a(haired a
CO“clusions
PecoI‘ding 0
dgcumentatf
i- m
Ofggientj
h Suppor‘
has had t
inv'mlab.
Such eva
the Prov
“it it

“thou
\

1
tiv it .
\v)
M

120

the laws of mathematics, Professor Einstein once wrote
'%hose of all other sciences are to some extent debatable
and in constant danger of being overthrown by newly dis—
covered facts."1

Nor have scientists adopted a theory of probability
such that it can be said that a specific probability is
attached to the many scientific laws which have been
established by inductive argument, such as many laws of
chemistry. For the accountants to seek the certainty of
the physical sciences, then, is perhaps an ill—conceived
objective. The attribute of scientific investigations to
be admired and imitated is not the infallibility of the i
conclusions so much as the meticulous accumulation and 1
recording of evidence, the scientist’s penchant for
documentation.

4. Arguments based upon experience form the bulk
of scientific history. In any situation where the evidence
in support of the conclusion is not complete, the scientist
has had to evaluate the evidence. The logician offers no
inviolable rules for separating good evidence from bad.
Such evaluation lies outside the realm of logic and within
the province of the researcher in the particular discipline.
And, it is a simple fact that no evaluation is possible

without a presentation of the evidence to be evaluated.

 

1”Geometry and Experience," from Sidelights of Re1a~
tivity (New York: 1923), reprinted in Feigle and Brodbeck,
Op. cit., p. 189.

 

 

 

Except in ne
clusions in the
unaccompanied by
those conclusio
tistics, case s
aspects of the

efforts made t
of these repor

10 ludge for

be documentec‘

‘evidence is ;

be, by the r

Striated~

Care i

Mounting
ment to th
Study has
Pecogn'lZe
believss
the natu:
many of
Mum.
Studv.

)0 have

121

Except in newspaper accounts, the reporting of con—
clusions in the physical and social sciences is seldom v
unaccompanied by the reporting of evidence upon which
those conclusions were based. These reports include sta-
tistics, case studies, etc., and comments on control
aspects of the observations as well as descriptions of
efforts made to preclude prejudicial observation. Readers
of these reports, colleagues of the researchers, are free
to judge for themselves (by personal experiment, resort
to documented sources, or whatever method) whether the
evidence is adequate to warrant a position which cannot
be, by the nature of its argument, conclusively demone .1 )
strated. ‘

Care in the presentation of evidence is not common in
accounting discussion, and its absence is a serious impedi—
ment to the acceptance of any given position. Dr. Moonitz’s
study has been criticized for its failure to specifically
recognize accounting objectives and purposes.1 This writer
believes the criticism is the result of a failure to present
the nature of the evidence on which Professor Moonitz bases
many of his conclusions. While he claims clearly neither
deductive nor inductive reasoning as the pattern for the
study, the facts selected for examination and argument appear

to have been governed by financial reporting considerations,

 

1”COmments on 'The Basic Postulates of Accountingf,”
Journal of Accountancy, CXV (January, 1963), pp. 44—55.

 

 

 

 

and, more specifi
ierations. To ti
selection are no
mderstood for t
viii be reached
it not substant
accountant -arg\
is adequate su
that evidence .
5- What
hint are ObSE
data are bus
1“ Wovens
bus‘meSS far
mics lead
Stern to f0]
iiejmce.
“illness j
hymns}
Perhaps a
“il‘incyp-I
times i
Obit of E

mm be

122

and, more specifically, public financial reporting consi—
derations. To the extent that these criteria for data
selection are not made clear, if necessary justified, and
understood for their effect on the type of conclusions which
will be reached, the argument is subject to methodological
if not substantive criticism. If colleagues of the
accountant—arguer are to be free to judge whether evidence
is adequate support for a position, they must be aware of
that evidence.

5. What constitutes evidence in an inductive argu—
ment are observations of data. In accounting, observable
data are business facts. What can be argued inductively
in accounting theory depends on what can be observed in
business factSrwwhat they have in common, what character—
istics lead to distinctions between them, what rules they
seem to follow. Moreover, the data must be selected without
prejudice. Professor Schrader's experiment with inductive
argument involved observations limited to exchange trans—
actions;1 and the bias in the evidence must be recognized.
Perhaps a more subtle bias could exist in any accounting
"principles” which Professor Littleton comments can ori—
ginate in practical experienceeflcan be derived inductively

2

out of accounting actions." The results of such induction

will be principles of accounting actions, which may or may

 

 

 

lSchrader, op. cit.

2Littleton,'Structure, p. 186.

 

 

 

 

mt be the equiv
a.“ observatior
kw”

re: 0: data 8.1T”

T'UUOSES, among

obaracterist
aiized about

asnditioni‘n

ing system
“that are d
accountan‘
This writ
involving
as rspor
Ethereal?
restric
ternal
univer

busine

 

123

not be the equivalent of conclusions which could be reached
upon observations of the data before accounting action. The
use of data already subjected to accounting actions, pre—

supposes, among other things, the defining characteristics

'abiding in the present accounting classificatory scheme.

Hence, when it is suggested that ”principles,” ”rules” and
”conventions were generalized out of . . . experiences in

i

meeting accounting needs,’ it is entirely possible that the
data have been preconditioned to reflect only certain of their
characteristics or relationships. And, what can be gener—
alized about is what is built into the data by this pre-
conditioning.

If what are desired are generalizations about an account—
ing system, such preconditioning is perhaps acceptable; if
what are desired are conclusions about the data with which
accountants are confronted, the situation is not acceptable.
This writer suggests, for example, that argumentation
involving some aspect of internal accounting reports (such
as reports for capital budgeting purposes) may include
generalizations about business facts including, but not
restricted to, those facts relevant to some aspect of ex-
ternal reporting. Stated in a more general way, the total
universe which might constitute evidence, the set of all
business facts, though infinite, may be divided into an in—
finite number of nonexclusive subsets, the membership require_

ments for each depending on the purpose or objective of the

 

 

 

 

 

iivision. And, 83
heevamined, the
class of rules f<
it. the sets to b:
mt exist.

6. The us
produce a co-or
serve as a gem
to agree with
of iccounting
:bveiy derive
exsentence f:

the accounts
tively argue
between pm
in ertralog
inductivs

avtomaticg

OI“ the Bab

equal rear

dilate. 1

Could be

ment (1
mains a

7

\
1

12A

division. And, granting an agreement on what sets are to
be examined, the accountant cannot appeal to any prescribed
class of rules for arriving at conclusions about the data
in the sets to be examined, for such a set of rules does
not exist.

6. The use of inductive argumentation alone will not
produce a co-ordinated body of general statements which can
serve as a general theory of accounting. Hence, it is hard
to agree with Professor Littleton, who suggests, in Structure
of Accounting Theory, that "Accounting principles (induc—
tively derived) will necessarily be interrelated, because
experience falls into patterns of related elements."1 If i
the accountant attaches differential status to his induc-
tively argued conclusions, such as, perhaps, a distinction
between principles and postulates, the distinction must be
an extralogical one. For, conclusions reached in separate
inductive arguments are neither interrelated nor ranked
automatically by the use of the same pattern of reasoning
or the same basic data. In a sense, the conclusions hold
equal rank, if the evidence is evaluated as equally ade_
quate. No hierarchy of conclusions results. A distinction
could be made, perhaps, based upon the scope of the state-
ment (i.e., how general it is), but the distinction re—
mains an extralogical one.

7. Whether at a high level of abstraction or in a

1p. 231.

 

 

barrov, speciali:
hill not resolve
zeal concepts as
repeatedly that
is not establis
accept such co
veuld simply p
This fact, am
some conclusi
clusions Serb
late h-5 of
that money -'
has beep at
true.1 Yet
inmoney, .
Such mease
tist is s
scientist

fag
set
him
all
bu

ti

 

125

the use of induction

narrow, specialized field of practice,
will not resolve, beyond question, disputes regarding cen—
tral concepts and assumptions. The point has been emphasized

repeatedly that what can result from inductive argumentation

is not established but probable truth. Yet, to refuse to

accept such conclusions until complete evidence is available

would simply preclude investigations on any new subjects
among others, has encouraged scientists to accept

This fact,

some conclusions for the sake of progress. The probable con-
cluSions serve as a basis for further investigation. Postu—
late A-5 of Accounting Research Study No. l, which states

that money is the common denominator for measuring purposes,

has been attacked on the grounds that it is not entirely

true.1 Yet, to explore the implications of measurement made
it is necessary to assume that

in money, or beans, or peanuts,
such measurement is made. Perhaps the attitude of the scien-

tist is summed up in the following quotation about a soc1a1

scientist who
reaches a state, finally, in which he has to
All these

face comprehensive sets of statements.
sets may be composed of statements which seem to

him plausible and acceptable. There is no place for
an empiricist question: Which is the Htrue set?
but only whether the social scientist has28uffic1ent

time and energy to try more than one set

1”Comments on ’The Basic Postulates of Accounting'
Journal of Accountancy, CXV (January, 1963), p. 49.
2Otto Nurath, Foundations of the Social Sciences (In-
ternational Encyclopedia of Unified Science, Foundations of
the Unity of Science, Vblume II, No. 5; Chicago: The Uni-

versity of Chicago Press, 1944), p. 13-

 

 

 

 

it can be seen, n
:igations is not
rather partly th:
Least, certain 1
ply accepted, at
iris respect, t
zbrmof the and

form any indiv

126

 

It can be seen, now, that the nature of these further inves— ,

tigations is not solely that of inductive argument, but
For, at the outset at

 

rather partly that of deduction.
statements or generalizations are sim—

least, certain laws,
In

ply accepted, and the scientist goes on from there.

this respect, the method of investigation resembles the

 

form of the axiomatic method discussed in Chapter III more

than any individual method of inductive argumentation.

 

 

Consider
alert in the
lhdtations,
matter of ac
hmcluded:

he

he use

concep

do the
Literally

w deducti

llviégsg

he, hr.
lhhe four
ills is
ments ar
are nev

content
\

1

r

CHAPTER V

   

TERMINOLOGICAL DIFFICULTIES AND A
"LOGICAL” SUGGESTION

 

A Terminology Problem

Consider the accountant, properly trained and profi—

cient in the methods of logic, aware of their power and

limitations, and desirous of approaching the subject
Mr. Paul Grady has

matter of accounting with these tools.

concluded:
Accounting principles (which, it appears,
he uses here as synonymous With postulates and

are not drawn from natural laws norl

concepts]
do they rest on inductive or deductive logic.

Literally speaking, of course, nothing rests on inductive
Yet as noted in Chapter

or deductive logic but more logic.

II, reasoning is what the logician subjects to scrutiny.
And, Mr. Grady would certainly admit that if, as he says,
'%he foundation for their [principles, eth] existence usu—
ally is based on business experience, contractual arrange—
ments and legal requirements, the principles, or whatever,

are nevertheless supported by reasoning whose substantive
If reasoning

”2

 

Content consists of the matters he mentioned.

lGrady, o cit., p. 55.

2Ibid.
127

 

 

 

islmvoived, the
pestion of when
med must be ans
That there
ioarticular ac
seriously argue
course of acts
other than gm
mould find no
and terminolc
Mishhat th
Gimme Class
and imports
Yet, a
Distillate,
AmpOrt tC
Vary in it
use of th
used in (
This} $0
of the e
which E

bv em,
\
1

UAGEr
hoctol

128

is involved, the accountant—logician is interested. The
question of where and how these tools can and should be
used must be answered.

That there is nothing more to accounting than taking
a particular action in a particular situation has not been
seriously argued. Even the decision to follow a particular
course of action in a situation must make use of something
other than the exact circumstances of that situation or we
would find no similarities in accounting records, reports
and terminology. Absence of any reasoning would mean, sim—
ply, that there is only one level of accounting knowledge
or one class of accounting statements, all of equal rank
and importance.

Yet, accounting literature abounds with such terms as
postulate, principle, concept, standard, and many more which
purport to relate to general statements of accounting which
vary in importance. Two difficulties have arisen with the
use of these terms. First, the different terms have been
used in connection with what appear to be the same things.
Thus, for example, there is an exploration of many aspects
of the entity "concept” by Dr. Li,l the entity ”theory” of
which Dr. Lorig writes,2 and two entity ”postulates” proposed

by Professor Moonitz. In these writings it is not always

 

1David H. Li, "The Nature and Treatment of Dividends
Under the Entity Concept,” The Accounting Review, XXXV
(October, 1960), pp. 674—679, and others.

2 . .
Lorig, op. Cit.

 

 

 

 

 

 

hssible to tell
intended to deny
simply that the
Level of genera
mals have bee
he of these te
ems-the sets
same members ,
the positioh
‘dddnitz, in 1
that there a
WW- Some
what terms
Such dime:

A.
this
such
ants

Add, Pref

is“ the

129

 

possible to tell whether the difference in terminology is
intended to deny that the other terms are applicable or 5
simply that the subject is being treated at a different
level of generalization or importance. Secondly, when pro—
posals have been made as to the members of the set named by
one of these terms——the set of postulates, principles,
etc.-—the sets are neither isomorphic nor do they have the
same members. For example, Arthur Andersen & Co.l holds
the position that there is only one postulate; Professor
Moonitz, in Accounting Research Study No. l, proposes
that there are fourteen; and Professor Chambers2 lists
forty. Some authors are kind enough to tell the readers i
what terms are to be considered synonymous;3 most are not.
Such differences have caused Professor Storey to comment:
Accountants have been extremely careless in

this matter, and the situation has degenerated to

such a state that it is doubtful w ether the account—

ants really understand each other.
And, Professor Gaa, in recognition of the situation, admon-
ished the profession strongly.

Since there is substantial disagreement on
terminology, our starting point ought to be to

 

 

 

 

1This position is explained substantially in The Pos—
tulate of Accounting—What It Is, How It Is Determined,
How It Should Be Used (Arthur Andersen & Co., 1960).

 

2Chambers, Towards a General Theory.

3For example, E. A. Spiller, Jr. ”Theory and Practice
in the Development of Accounting," The Accounting Review,
XXXIX (October, 1964), pp. 850—859.

“Storey, op. cit. p. 62.

 

 

 

reach agreen
to use so 5
llberatiODS
can we disc
about "903“
“practices
the same 1
ascertain
partieulax
These inc
nno‘olenns in is
the status or
hat is, if E
entity “cone!
contracts or
stood that \
general int
about oonsl
terminolog
discussion
knowledge

ll
if new

ll

a Danti
939%? is
“11%ng
PTOTESE
“Std in
Qinles

\

bles‘

130

reach agreement on the important terms we plan

to use so frequently and so profoundly in our de—

liberations, literature, and reports. . . . How

can we discuss efficiently and read intelligently

about ”postulates," "concepts," ”principles,”

"practices,” etc., when we are not all speaking

the same language? We must waste time first to

ascertain how a given term is being used by a

particular speaker or writer.

These inconsistencies in usage need not present serious
problems in isolated cases when the point at issue is not
the status or acceptability of the concept or whatever.
That is, if passing reference is made, for example, to the
entity ”concept” in a discussion of long—term construction
contracts or installment sales, it will normally be under—
stood that what is meant is simply ”something” of more
general interest or applicability than specific statements
about construction and installment contracts. But the
terminological inconsistencies do present problems when the
discussion is related to the organization of accounting
knowledge and not the knowledge itself. That is, speaking
Of "postulates” is to be distinguished from speaking about
"a particular postulate." Since what is at issue in this
paper is the application of logic to something, the ter-
minological inconsistencies do affect the discussion, and
Professor Gaa‘s warning is to be heeded. Is logic to be

used in studying and developing concepts, postulates, prin-

Ciples, practices, theories, conventions?

 

1Charles J. Gaa, ”Uniformity in Accounting ’Princi-

plesfl" Journal of Accountancy, CXI (April, 1961), p. 49.

 

 

 

 

 

 

Judging fro
“postulate," "p1
nention,“ all a
eccountant‘s vc
sill disappear
nade about, an
in the past h:
sent as to th
The objective

1)

 

131

Judging from the continual usage of the terms ”concept," .

l

"postulate," ”principle,” ”practice,' "theory" and "con—

vention," all appear to have become a permanent part of the
accountant’s vocabulary. It seems most unlikely that they
will disappear now. This writer believes that the comments
made about, and the distinctions made between, these terms
in the past have been inadequate to permit a direct state—
ment as to the applicability of specific methods of logic.
The objectives of this chapter are, as follows:
1) To suggest what it is hoped will be a
useful set of distinctions between
these terms which will
a) generally coincide with present
usage, but which ‘
b) allows the identification of spe— ‘
cific logical relationships.
2) To suggest those methods of logic, if any,
which appear appropriate for investi—

gations related to the accounting matters
falling in each of the defined categories.

 

3) To indicate the advantages of the pro—
posed interrelated structure, or set of
distinctions, over those suggested by
other writers.

 

Terms Distinguished—An ‘
Interrelated Structure
The Basic Approach
Few accountants, if any, would question the necessity
of reasoning in some form in the presentation of accounting
discussions. Undoubtedly, the most valuable attribute of

reasoning, from the point of view of progress, is that

 

attribute which
to accept a com
are accepted.
truth or falsit
oi the neasonin
concerning the
independently
disagreements
through the c
it the reason
ital analysi:
to the acoep
atgunent is
i this type
nost assure
if account;
Peason, ch.
150% inv
the when
Micah};
ten 111 (

 

method bl
Qhatter
ent 3126

“A Witt

 

132

attribute which characterizes deductive logic~the necessity
to accept a conclusion if the premisses on which it is based
are accepted. Whatever disagreement may exist concerning the
truth or falsity of premisses may be discussed independently
of the reasoning. But also, whatever disagreement may exist
concerning the reasoning used may, likewise, be considered
independently of the acceptability of the premisses. In
disagreements of the latter type, a final answer can be found
through the careful analysis of the specific inferences made,
if the reasoning is expressed in a manner suitable for log—
ical analysis. The problem of disagreement is thus reduced
to the acceptability of premisses, and it is known, once the
argument is in the proper form, just what the premisses are.
If this type of problem is the most difficult to solve——and
most assuredly it is in any discussion——it is to the benefit
of accountants to isolate it from other problems. For this
reason, the significant relationship in the current proposal
is one involving deductive logic. The basic organization in
the proposal is predicated on the use of deductive logic, spe—
cifically the informal axiomatic system as discussed in Chap—
ter III (which is essentially the same as the hypothetical
method briefly discussed in Chapter IV). It was shown in
Chapter III that such systems can handle problems of differ—
ent sizes and complexities, with varying degrees of rigor,

and with or without the use of complex symbolization tech~

 

 

 

clones. These d
opportunity to U
iyor and minor

sit the system

Tnelroposal
On page 1
interrelated s
:‘neory, come
this structur
conventions :'
the position
informal and
allowable i~
desired), t
concepts, 3

l

The
hence) a
on a 596

desumpt

It

133

niques. These characteristics offer the accountant the
opportunity to utilize such systems in discussions of both
major and minor accounting matters, and, to some extent, to
suit the system to the degree of sophistication of his audi—

ence.

The Proposal

On page 134 is a simplified graphic presentation of an
interrelated structure of accounting terms consisting of
theory, concepts, postulates and principles. An extension of
this structure to include identification of practices and
conventions is based upon this simplified structure. Given
the position to be held by the accountant in the form of an
informal axiomatic system (including the identification of
allowable inferences in accordance with the degree of rigor
desired), the structure allows for the distinction between
concepts, postulates and principles, as follows:

1. concepts—the basic terms or primitive
notions of the system;

2. postulates—the assumptions or axioms
of the system; and

3. principles—any theorems of the system.

The whole system will be said to constitute a theory.
Hence, a theory might be described as an argument or position
on a specific subject, consisting of basic terminology,
assumptions, and demonstrated (proved) conclusions.

It is then proposed that an accounting practice be

 

 

WHQLQOHUPWW

 

 

134

ILLUSTRATION 1

STRUCTURE OF A THEORY

ooSooUdm

 

 

 

>
wodeHmdmm
afioosw

 

 

wswzowpwmm v

 

nmnnodas anz
theory by adoptio
arms in that the
ftra definition
"any theories.
The follow
proposed defin‘.‘
comparison is
attempt to rel
inc terms are
ilstinctions

inapplicanii

lhe p
1“a with c
Aden thi
ing the g
0f: for
Tile sub:
as from
Clamp
coilpat
for SC

limit

 

135
understood as an action in accordance with the principle of a
theory by adoption of a specific definition of the basic
terms in that theory, and that the term convention be used
for a definition of some basic term which is common to
many theories.

The following section is devoted to a comparison of the
proposed definitions with current usage of these terms. The
comparison is not one between definitions, but rather an
attempt to relate the proposed definition to the use to which
the terms are currently put. The advantages of the proposed
distinctions are pointed out as well as the applicability or

inapplicability of additional methods of logic. .

 

Analysis of the Proposal

Theories

 

The proposed definition of the term ”theory" is in keep-
ing with current usage in many respects. That it is possible,
under this definition, to have more than one theory concern-
ing the same subject is consistent with present discussion
Of, for example, the entity theory and the proprietary theory.
The subjects of different theories may range in importance,
as from a theory of depreciation to a theory of balance sheet
classification. Since different axiomatic systems may not be
Compatible because of different axioms, it would be possible
for some theories to be incompatible, as are some present

positions relating to deferred taxes and long—term leases.

 

 

 

Providing a
"’”di“" 0? some ch.

126.5 appeared t

no

arguments unt'
singcted, or
1: direct cor
rigorous exa

for progress

problems an
iPo'clems d
Etdo‘dht'lhp
i1assin“ic
Edd probl
thstion
1Q'd1 (105
deferre
GIG-tbtic

tdies

3% to

m, Fe

 

136

fi

Providing a definition of theory which allowed for a

theory of something less than the whole of accounting know—

ledge appeared to this writer to be not only in keeping with

common usage, but also an absolute necessity. The construction

of a general theory of accounting, advocated and even

attempted by some, is an admirable project, but, as discussed

in Chapter III, a tremendously ambitious one. And to delay

the application of logical analysis

arguments until a suitable general theory has been con—
structed, or to limit analysis to such a theory, seems
in direct contradiction to the present interest in more
rigorous examination of accounting problems and the desire

for progress. One small problem carefully analyzed and

to limited accounting

 

brought to solution is more progress than no large or small

problems analyzed or solved. Moreover, not all accounting

problems deal with the whole of the
accounting. For example, there are
classification which do not involve

and problems of cost accounting not

Questions. The question of continued adherence to histor-
ical cost does not rest on the recording and reporting of
deferred income taxes. And, even more to the point, the
question of how to record and report deferred federal income
taxes should arise only after the question has been answered
as to whether there are such things and whether they are to

be reported, The present concern over a general theory may

 

subject matter of
problems of balance sheet
revenue considerations,

involving disclosure

 

 

 

be an indication
my be a reactic
cast, but there
facing accounta
general theory
in Chapte

for some accon
“th 0i acco
:0 the eXC1LlE
lnmentory at
M cost ac
provides f0.
WSW re
Q1Die of 1‘
theory Sin
ence to h

m I”tonic
lddiy‘ldu:

Sti

weSent
StPthe

be) by

Dies 0

totem

ments

gene]

 

137

be an indication that something is amiss in accounting, and
may be a reaction to the case—by—case approach used in the
past, but there is certainly no proof that all the problems
facing accountants are related to the absence of such a
general theory.

In Chapter I it was noted that the justifications given
for some accounting actions indicate the presence of a hier—
archy of accounting statements by appealing to one statement
to the exclusion of some others. For example, reporting
inventory at the lower of cost or market departs from histor—

ical cost accounting but is considered appropriate since it

 

provides for foreseeable losses. The careful construction of
a theory regarding inventory reporting which includes a prin—
ciple of lower of cost or market pricing may show that the
theory simply did not include a postulate relating to adher—
ence to historical cost. The apparent hierarchy may turn out
to result from the omission of unnecessary postulates in
individual theories.

Still, the proposal in this paper is in keeping with the
Present usage in the sense that if_such a theory were con—
structed, its concepts, postulates, and principles would
be, by definition, the_basic concepts, postulates and princi-
ples of accounting which are now being sought. The appli_
cation of the axiomatic method in several accounting argu~
ments may, in fact, contribute to the development of a

general theory by revealing that a particular series of

 

 

 

assumptions appez
acid provide so
the postulates r
”caterer, that a
nothing more ti
insofar a

an accounting
membership in
M carefully
terminology ;
lowed in pro
latter requf
aTimer is e

Dibblem of
atlahtages

to a form

its Val-1d

Of COUSiC

diam be.
those WE

tier 1‘6:
idioms

“18 m

the dy

138

 

assumptions appears necessary in each case.

Such experience

would provide some real evidence that these assumptions are
the postulates of a general theory. It should be remembered,
however, that a general theory of accounting will still be
nothing more than a theory.

Insofar as there is apy_reasoning to support any position
on accounting matters, the minimum qualifications for
membership in the proposed class of theories can be met by
(l) carefully analyzing the reasoning to isolate its basic
terminology and assumptions, and (2) outlining the rules fol—
lowed in proving the position from those assumptions. The
latter requirement is the most important if the accountant—
arguer is ever to isolate the problem of premisses from the
problem of reasoning. And therein lies one of the major
advantages of the proposal. If the argument g§p_be reduced
to a form which allows review of its reasoning and tests of
its validity, its premisses and conclusions are deserving
of consideration. If the argument cannot be so reduced, it
might better be called only the discussion of an opinion,
whose weight rests solely on the persuasiveness (for what—
ever reason) or perhaps the reputation of the arguer; the
discussion will be convincing to some and not to others.
This description, to a great extent, approximates the sit—
uation in professional writing of the past and present.

There are no rules or regulations, no methods of logic,
inductive or deductive, which the accountant can apply to

produce for him a theory to express his ideas on a particular

IIIIIIIIIIIIIIIIIlll::::;________________

 

detect. he muss

tics of inierenc

tests, the settin

annertaken be
in gstem—buil<
niortunately,
ethos Given
midea, and a
"an be asked 2
ironed true?
tfihmum
in another s

The theory j

The p}
“in is, 1
“‘5 genera
“concept"
loconmtar
incpme,

the term
obligat
is lute
meory
Verl C

Fem

139

subject. He must have a working knowledge of the deductive
rules of inference generally used. One of the most difficult
tasks, the setting up of the boundaries of the subject, must
be undertaken before the system is constructed. Before that,
the system—builder must have an idea and an objective. And,
mfibrtunately, for ideas as for systems, there are no algo-
rithms. Given the construction of an appropriate system for
an idea, and accepting the validity of its proofs, the question
may be asked as to whether the theory can be said to be
proved true? No! To the extent that it rests on the truth
of its postulates, which may rest either on other postulates
in another system or some method of inconclusive induction,

the theory isn't true; it is merely useful.

Concepts
The proposed definition of concepts, like that of the—

ory, is, in many ways, consistent with present usage. Though
not generally thought of in.relation to an axiomatic system,
"concept” is often used synonymously with "basic term."
Accountants speak of an entity concept, or a concept of
income, of capital, or of cost. Though the proposed usage of
the term as part of a definite system carries with it the
Obligation of the system—builder to explain what the concept
is intended to mean, it also allows the meaning to vary from
theory to theory. In this respect, the proposal coincides
very closely with current usage, for discussions of income

reporting and measurement often deal with different concepts

 

 

 

 

:‘f income tee-l
:cmept, etc).
{zo‘ildbe reiler
ations but mean
in still a
inline with u
i.esearch Study
mich consistt
continuity.
sols Study (in
NW or co
in with has
ESSUmptions
Eases to be
graph}? in
dimSsior
Fslates e.

[1183) SUC

ter 111)
terms Of
The dig
Ellen j

in aQCn

140

of income (e.g., the accretion concept, the purchasing power
concept, etc.). Two arguers can both assert that all income
should be reflected in a statement of the results of oper—
ations but mean entirely different things by their assertions.
In still another respect, the proposed definition seems
in line with usage in accounting literature. In Accounting

Research Study No. 1, each postulate is given a brief title

 

which consists of some key word or phrase such as entity or
continuity. Similarly, each of the postulates of the Illi—
nois Study Group1 relates to some basic idea such as, again,
entity or continuity. For these and other such reports deal-
ing with basic assumptions of accounting, the wording of the
assumptions may differ, but the basic terms appear in many
cases to be the same. Even in the Paton and Littleton mono—
graph,2 in which concepts are equated with assumptions, the
discussion of each assumption is preceded by a title which
relates to the very same basic notions noted in other stu-

dies, such as entity and continuity. It was noted in Chap-

ter III, that the axioms of a system relate to the basic
terms of the system and assert relationships between them.
The differences in the wording of the actual statements

given in the various studies about the underlying assumptions

in accounting may, in fact, be different axioms involving

 

1A Statement of Basic Accounting Postulates and Prin—

Ciples.

2W. A. Paton and A. C. Littleton, An Introduction to
Corporate Accounting Standards (American Accounting Asso—
ciation, 1955).

 

 

 

IIIIIIIIIIIIIIIIIIII::::;______________

 

 

 

insane basic te
:‘rese terms. In
i: appear to be
nth the defini
Under the
newer, limit
:‘nose noted at
identification
1“economic act
orewen “fat:
The req
defined for
‘3": this pro
that in son
the System
understw
leans Sys
the terms
and Used
1“ in ac
cm was
Sta“din
detail
1933,
8One
late<

nus.

141

the same basic terms or perhaps different definitions of
these terms. In any case, that there are such terms which
do appear to be basic in such discussions seems to coincide
with the definition of concept proposed in this paper.

Under the proposed definition, concepts would not be,
however, limited solely to basic terms of accounting such as
those noted above. For example, common words or phrases with
identification beyond the subject of accounting, such as
'economic activity” or "time period,” and perhaps ”report”

or even "fair,"

may be basic terms in some arguments.

The requirement that whatever basic terms there are be
defined for the system could prove to be a major advantage
of this proposal in the long run. It is true, of course,
that in some cases this requirement has been ignored when
the system—builder believed that the terms were sufficiently
understood by themselves. For example, in the case of the
Peano system relating to arithmetic, the author believed that
the terms "number” and ”successor" were generally understood
and needed no further comment. Such might also be the case
in an accounting argument in which one of the basic terms
was "asset.” If all that was needed was a general under—
standing of the term for that particular discussion, no
detailed definition of the concept need be given. Neverthe—
less, it was also pointed out in Chapter III that the use of
some very common terms as primitive in systems has stimu—
lated interest in developing more adequate and useful defi—

nitions (as, for example, interest in determining what

 

 

 

 

cdlyis innolve
:masd interest
n.mmelcome or
it. David Anders
Even i
torial boc
ment as tr
important
inhoi the de
cation oi? some
enpense, qfiyj
iii the WOUO
undies ado
alpear “m gy
Can 0t
aIminis
9‘: their C‘
leech,n 0
am frwit
midi an
there we
Ifilling
lh £1in
be con
then)
went

\

Fina
1963

142

really is involved in the definition of "number”). In—
creased interest in definition improvement would not be
an unwelcome or an unneeded development in accounting.
Mr. David Anderson has remarked:

Even in the narrow confines of our own terri—
torial boundaries, we accountants are not in agree-
ment as to the exact meaning of some of the more
important and elementary terms which we employ.1

Much of the day to day work of recording involves the appli—
cation of someone‘s definition of such words as asset or

expense. This important fact is given explicit recognition

in the proposed definition of a practice which specifically

involves adopting a definition of the basic terms which

appear in the statement of a principle-

Can other methods of logic help in the selection of

appropriate basic terms for a theory or in the establishment

0f their definition? As discussed in Chapter III, the 58—

lection of terms is governed by considerations of simpliCity

and fruitfulness, and is generally the result of both careful

study and a certain amount of trial and error. NOT are

 

there methods of deductive logic to assist in selecting de—

fining characteristics. If one wishes to argue, for example,
in faVOr Of a particular definition, another system might

be Constructed. But the basic terms of that system will,

the“: require an understanding in order for the new argu-

ment tO be followed and finally accepted. Within the

VlDaVid S. Anderson "Communications Problems of

Financial Reporting,” dournal of Accountanc , CXV (April,

 

innenor‘i oi the
knead of 21 def?
ment, and the
":eatnenry, sa;
in Chapter
nt -rom ext
has been he lpi
aresuit of c:
:‘nenrist to m
is form pa}
may Sugge
"alteristic <
such a Case

manned,

Km the E

fitment
assets ‘
mal) WY
Lthan 0‘
“5% 01’
the a

laplj

383

1A3

framework of the definitions proposed in this paper, the term
in need of a definition would be a concept, say an income
concept, and the new argument for a specific definition would
be a theory, say a theory of income.

In Chapter IV it was pointed out that the method of ar—

gument from experience, specifically inductive generalization,

has been helpful in the classificatory defining process. As

a result of careful observations it may be possible for a

theorist to make some generalizations as to what characteris—

tics form part of the definition of, for example, an asset.

He may suggest that the definition of asset include the char—

aCteristic of being capable of being converted into cash. In

SuCh a case, he must be prepared to reveal his argument which

led to that conclusion, indicating the principle he followed

in selecting business facts to observe and the various facts

Observed. His argument will be open to question, then, as to

(1) the possible bias produced by his selection principle,

and (2) the adequacy of his observations. Using an indUCtive

argument he cannot hold it to be necessarily true—even lf

assets were the same sort Of thing as, say, a species Of ani—

mal, which they are not, there is the possibility that fur~
the? observations might contradict his conclusion. In the

ch as assets,

use Of man-made notions and classifications su

the applicability of this method of loaiC seems Particu’

larly limited, but at least not impossible.

Argument by analogy may likewise be of some limited
' .t
aSSistance in building a definition. For example, one mlgh

 

 

‘niiize premisse
andnonoroi‘it 01
nnoroiit organ
figment of e
that character
lations. Ont
all methods 0
refutable wit
example, We
and nonproff
inference t
“f a Profit
It was p01

arguments

ties and
The mix
Interppe
EmEminent

nature

It

logic

Lme a»

Wide

work

dost

\

144

utilize premisses listing the characteristics which profit

and nonprofit organizations have in common, and conclude that
nonprofit organizations have, as a purpose, the most efficient
management of economic resources under their control, since
that characteristic is exhibited by profit—motivated organi—
zations. On the other hand, analogical argument may, like
all methods of inductive argumentation, lead to conclusions
refutable with the addition of a single new premiss. For
example, premisses listing the characteristics which profit
and nonprofit organizations have in common might support the
inference that nonprofit enterprises have the characteristic
of a profit motive exhibited by profit—motivated enterprises.
It was pointed out in Chapter IV that of major importance in
arguments by analogy is the selection of relevant similari—
ties and the consideration of all relevant dissimilarities.
The evaluation of the argument rests with the individual's
interpretation of what is relevant. Thus, though analogical
argumentation may be useful, it is inherently limited by its
nature as a method of induction.

To the important question of whether through the use of
logic there will be true or right definitions or concepts,
the answer must be the same as that with theories. N01 The
concept may be useful in a theory. The definition may be
workable. But for truth, the question must remain open.

Postulates

!

The use of the term ”postulate,’ as proposed here,

 

 

 

 

 

 

 

 

 

 

{——

nincides in one

inc current acco

essentially to 2

ice proposed dc
system is part
a: axiomatic 1
definition.
The iai‘
lates to the
he major d.
‘failnre is
hemiodsly
Efforts tc
include t
hand, in
SO‘C’dlle
and 00mm
htsg 01‘
is not
made,

A900»

\

Dog-

145

coincides in one major respect with the common, if not all
the current accounting, usage of the word in that it refers
essentially to an assumption. In the sense that an
assumption is, by common understanding, something not proved,
the proposed definition of postulate as part of an axiomatic
system is particularly appropriate; for, the postulates of
an axiomatic system are, ip_tha§_system, simply true by

definition.

 

The failure to confine the members of the set of postu—
lates to the assumptions of a general theory of accounting is
the major deviation from current accounting usage. But this
failure is deliberate. It is the automatic result of the
previously expressed opinion against directing all research
efforts to the development of a general theory which would
include §h§_basic postulates of accounting. On the other
hand, in View of the wide range exhibited by some of the
so-called basic postulates suggested already (from specific
and complex statements to such simple observations as fair-
ness or integrityl), the proposed definition of postulates
is not so unreasonable. Depending upon the argument being
made, postulates could range from the very specific to the
very general, and may differ from theory to theory.

Where would the postulates of a theory come from?

Accounting literature in support of any accounting position

 

1Gordon W. Stead, ”Towards a Synthesis of Accounting
DOCtrine, " The Accounting Review, XXIII (July, 1948),
pp - 355-359-

 

IIIIIIIIIIIIIIII'll-Illlllll---_____i

 

sustains stateme

is in some may <
he mriters to
logically a pas
statements wit
logical impor‘
:‘nough left c
drawn. Such 2
difficult ta
fessor Moon

in this fac
being able
and t0 nar
assillllpt‘lo
lines

There an:
hstulat

be inve

Oi SOmc

used 1

the a:

he a]

eitch:

Dia(

146

contains statements which, it must be supposed, are intended
to in some way contribute to the argument. It remains for
the writers to analyze their work to (1) find out what is
logically a part of their arguments, i.e., to separate the
statements with logical import from those with only psycho—
logical import, and (2) seek out those assumptions that,
though left unsaid, are needed to support the conclusions

drawn. Such analysis may, and probably will, be an extremely

 

difficult task. The partial logical analysis of two of Pro—
fessor Moonitz's postulates shown in Chapter III testifies
to this fact. But, the ultimate rewards, the ability of
being able to evaluate the reasoning on an objective basis
and to narrow the areas of disagreement to specifiable
assumptions, seem to warrant the undertaking.

Where do new postulates come from for new systems?
There are no rules for their invention; though, since the
postulates are only assumptions in a system, they can, indeed,
be invented. Postulates of one system may be the principles
of some other system. A slight alteration to assumptions
used in other systems may offer possibilities. For example,
the assumption that all assets have value in exchange might
be altered to assume that some assets do not have a value in
exchange. The implications of the latter assumption in
place of the former in some theory may be very interesting.

What of the acceptability of postulates, one of the
maJOr questions of today? For apparently simple postulates

such as, for example, Professor Mattessich‘s assumption

 

IIIIIIIIIIIIIIII'll-Illlllll---_____i

 

“here exists a
Little criticis
reader is confi
rest partially
change the mat
theory and no
accepted so i
deviation fr
of things oz
to question
To Prc
the time p
help here‘
ciidenoe
lattessic
claim is
inductii
0f use
hunt
degree

Whose

147

’There exists a set of elementary additive time intervals,"1

little criticism may be forthcoming (at least, until the
reader is confronted with some of the conclusions which may
rest partially upon this assumption). This fact does not
change the nature of the postulate as an assumption in his
theory and nothing more. And, a strange assumption to be
accepted so freely by those accountants who try to avoid any
deviation from reality and fact~for, the existence of a set
of things called additive time intervals certainly is open
to question.

To Professor Moonitz‘s more complex assumption regarding
the time period, more criticism can be expected. Can logic
help here? As pointed out in Chapter IV, the claim of self-
evidence is seldom, if ever, justifiable. Even for Professor
Mattessich's classic simple assumption quoted above, the
claim is certainly not appropriate. Argument by analogy or
inductive generalization, as discussed in Chapter IV, may be
of use here. If a postulate of some system has been accepted
(obviously not universally so, but rather accepted to some
degree), the theorist may present an analogical argument
whose conclusion is a postulate of his new theory. For exam—
ple, suppose a theory were available regarding long—term
debt, which theory included a postulate that such debt is to

be recorded. An analogical argument detailing the similari—

 

 

1Mattessich, Accounting and Analytical Methods, p. 32,

 

 

 

:tsbetween deb‘
aiiieading t0 '5
aremrded may
:nmiease obli
iebusiness f:
punsmay lead
obligations. T
inductive in
problems exp‘
imonciusive
hgic offer
Eductive 1
postulate
Which sum
the new 9
meountii

must be

“M!

iii
agree
Ltheme
are 6
My
cret

For

148

ties between debt obligations and long—term lease obligations,
and leading to the conclusion that lease obligations are to

be recorded may support a postulate in a theory about long-
term lease obligations. Or, some extended observations of

the business facts surrounding the operations of pension

plans may lead to some new generalization regarding pension
obligations. But both these types of argument are, of course,
inductive in nature and subject to all the questions and
problems explained in Chapter IV. The arguments will be
inconclusive. Nor would turning to some method of deductive
logic offer solution to the problem. For, any method of
deductive logic requires premisses. And the criticism of the
postulate at hand will simply be transferred to the premisses
which support it, reducing the situation to one of justifying
the new postulates. Can logic, then, prove the truth of the
accounting assumptions in a system—the answer here, too,

must be in the negative.

Principles

While the proposed definition of a principle does not
agree with present definitions and usage in all respects,
there are important similarities. Principles, as proposed,
are definitely derived from postulates; the advantage of this
proposal is that the phrase "derived from” is given a con~
crete meaning. This has not been the case in other proposals.

For example, in the suggestions found in the writings of

 

 

 

toward Spacelc1 i
dich underlies
postulate and ti
out of a princ
float the result
accordance wit'
review of Chat
he, may be o
for adequacy
is neither f;
flier PEQEELE
fairness in
hunting a
oethoq 0th
a“diluent f
Character
8th a (16
The
ileum
with Cu
€th 01
Die, t

\

W11
Dies
We I
dug:
SQVQ

149

Ieonard Spacekl it is held that there is one basic postulate
which underlies principles, and the relationship between the
postulate and the principles is that the ”reasoning” in sup—
port of a principle is to be evaluated or rated to the extent
that the results achieved by the principle‘s adoption are in
accordance with the basic postulate. It may be seen from a

review of Chapters III and IV, that the reasoning, if deduc—

tive, may be tested for validity, or if inductive, reviewed

for adequacy and bias of evidence. But deductive reasoning

is neither fair nor unfair, only Valid or invalid. And the

maJOT roblem in the evaluation of an inductive argument is

fairness in the sense of possible prejudiced evidence.

Adopting a position such as Mr. Spacek's, perhaps there is no

methOd other than some type of court to evaluate whether the

aTEUment for some ”principle" pOSSesses the sought—after

characteristic of fairness—there is no logical way to make

such a determination.

The present proposal as related to principles offers a
fléXibility that seems particularly attractive and in line
With Current usage. Principles need not be consistent with
For exam—

eaCh other if they are parts of different Systems.

p18: the construction of a theory of inventory pricing by 3“
lHNeed for an Accounting Court,” The Accountin T§EZ§?W’

.XXXIII Jul 1 8 , 68- 7 ' ”Solution to the P

ples Dilemmg:” gge)Ac§§unEing3Rgview, xxx1x (April, 1964),
pp. 275“2843 "Are Accounting Principles Generally Agceptzd, ‘
£9239§l_9£_5929222§£§yg CXI (April, 1961), pp- 41-4 : an
Several others.

 

H

 

accountant whose
in: reach dif‘
accountant whOS
tapers. ObjeC‘
irdluencing be
iered to be t‘
situation is
part of nanu
plant but as
at the home
Princi
of general
pith curre
principle
sure of 1
relating
principl
tencp '1
M
iv the
iPOVir

hr, p

\

own

150

accountant whose purpose related to intra-company transfers
might reach different principles than one constructed by an
accountant whose purpose related to sales to outside cus—

tomers. Objectives of the theory can make a difference by
influencing both the basic concepts (such as what is consi—
dered to be the meaning of ”cost”) and the postulates. The

Situation is not unlike the classification of some costs as

part of manufacturing expense if incurred at a manufacturing

plant but as general and administrative expense if incurred

at the home office.

Principles, as proposed, need be of no SpeCifiC degree

0f generality or applicability. This, too, is in keeping

With current usage. In ACCOunting Research Study NO. 7: some

principles are very specific, such as A—8, relating to disclo—

Sure of rental charges on material lease contracts, or 0—1,

relating to valuation of current assets. On the other hand,

principle A—lO speaks of disclosing the effect of inconsis—

tency in the application of any other principle- Willi—35

. l
and Reporting Problems of the Accounting ProfeSSion refers
to the principles of recording all assets and liabilities,
providing for all costs, and matching costs and revenueS.
Mr, A. Carl Tietjen suggests a principle of fair presentation.
1
Arthur Andersen & Co., 1962:

 

2A. Carl Tietjen, ”Accounting Principles, Eraigéges
and Methods,” W: CXV ”p“ ’ ’
pp. 66—68.

 

locounting Resez
fron“Profit is
activity“ to “:
inuested capit
lithIl of whai
that all stat
as some of ti
system night
about long-i
those prim-9
assets and
in an
clilies wi?
sense of
their tn
hostulat
Sintern.
It
an act.
Turnip
amen
iosal
Thrt
Siti
to

th:

A

 

 

151

Accounting Research Study No. 3 suggests principles ranging
from "Profit is attributable to the whole of business
activity" to "Stockholders' equity should be classified into
invested capital and retained earnings." The proposed defi—
nition of what may be called a principle would not require
that all statements with this label be of such sweeping scope

as some of those suggested above. In fact, principles in one

 

system might be postulates in another. For example, a theory

about long—term leases may require a postulate such as one of

 

those principles suggested above regarding the recording of all
assets and liabilities and the matching of costs and revenues.
In any case, by virtue of the definition proposed, prin-
ciples will be derived by logical means, in a specifiable
sense of the term ”logic,” from postulates. To that extent,
their truth would follow directly on the acceptance of the
postulates and the prescribed rules of inference in the
system. More than that cannot be claimed.
In only one major area does the present proposal require

an actual change in the language of the accountant. The con—

 

formity of financial statements with ”generally accepted

accounting principles”

has no real meaning under this pro—
posal unless and until a general theory of accounting is

'forthcoming and pronounced accepted. Since such is not the

 

situation at present, one might suggest that the phrase has
no more meaning now than it would have under the proposal in

this paper. Still, in case such a situation should arise,

 

:‘oo principles z
principles of a
accepted accouv
into the histo
:ersies over ‘
its eventual

There have be

 

liautices :
Teiated to
ing but I‘;
lilth the
“Piracy
simple 5

Cedip

i310u
ill

 

 

152

the principles of that general theory would become ppp_basic
principles of accounting. Though the phrase "generally
accepted accounting principles" is almost unalterably etched
into the history of accounting, the extent of present contro—
versies over its meaning may be interpreted as a sign that
its eventual elimination may not be entirely undesirable.

There have been, in fact, some suggestions to that effect.1

Practices

The proposed definition of a practice, like the other
definitions, conforms in many respects to present usage.
Practices are definitely different from principles but are
related to them. Practices are not statements about account-
ing but rather accounting actions—~the accountant's actions
with the facts of business. There is no difficulty in
separating principles frOm practices—it is the separation of
simple statements from accounting acts. It was noted in the
discussion of concepts that much of the work of the prac—
ticing accountant involves the application of definitions.
The proposed definition of a practice takes full cognizance
of this fact.

Consider one of the suggested principles in the pre—

ceding section——that all assets and liabilities be recorded.

 

1See, for example, Tietjen, o . cit., and Carman G.
Blou h, "Principles and Procedures, Journal of Accountancy,
CXI April, 1961). pp. 51-53.

 

 

 

 

lepracticing a
1'? an asset to h
to provide for :
he considered a
invested capit;
adopt a defini
involving mp
he acted upon
concepts,
No moth
Erect to ac<
huts the C
are illogio
itiinij'mn
tut, one s
“mascot“
MET, p
minim

has him

 

 

153

Wm practicing accountant has to adopt somebody‘s definition
of an asset to be able to act upon this principle. Similarly,
to provide for all costs it is necessary to decide what is to
be considered a cost. To classify stockholders' equity into
invested capital and retained earnings, it is necessary to
adopt a definition of invested capital. Even principles
involving such terms as disclosure and current assets cannot
be acted upon without selecting some definitions of those
concepts.

No method of logic seems appropriate for use with re—
spect to accounting practices, which statement perhaps sup—
ports the comment sometimes made that accounting practices
are illogical. An accountant may give reasons for selecting
definitions for the basic terms in some principle or other;
but, one should not confuse the phrase ”reasons for" with
"reasoning behind" or ”reasoning” as used throughout this
paper. An accountant may give, as the reason for selecting
definition, that he believes it is easier to apply, that he
has adopted it on other occasions, that others have adopted
it, etc. The selection process is not logical in any strict
sense of the word logic. To the extent that practices, as
defined in this paper, are not selected by logical methods,
and require definitions which also cannot be developed
entirely by logical means, there is some support for the

Opinion that methods of logic may be of little immediate,

 

 

 

 

practical use t
is with do

are to suggest
icing account
log out his f

opinion on ti

that there n

and wrong

like COUCE

The
definpt-l
rent ops;
'uith th

QhSisi
thousl
We“ ‘
Singp

aQQOo

 

 

154

practical use to accountants.
As with definitions, it would not be strictly appropri—
ate to suggest that some practice is right or wrong. The prac-
ticing accountant adopts some definition of income in carry-
ing out his functions. The present honest differences of
opinion on the definition of the concept ”income” suggest
that there may be no right and wrong definition for a concept,
but only definitions with varying degrees of usefulness for
different purposes. Given these differences of opinion, it
is difficult to support the position that there can be right
and wrong practices; better to say that accounting practices,

like concepts, have varying degrees of usefulness.

Conventions

The proposed definition of a convention is, of all the
definitions proposed here, the least in conformity with cur-
rent usage by accountants. The major problem here is not
with the definition of the term, however, but with the in—
consistency in current usage. Conventions are used synon—
ymously with rules, concepts and standards, and sometimes
even with principles, postulates and practices; yet, no
single usage appears to predominate. In usage outside of
accounting, however, a convention, or what is conventional,
is associated generally with action. For example, the common

phrase "It is conventional in the industry to do . . ." or

1

"I did the conventional thing.' In the proposed definition

 

 

 

 

:i‘ convention, a
is noted io
elepent of an a
accountant to z
2'? an adequate
ciple. To be
':e able to re
actions invo',
classifying
the def:
25 a matter
9:11th tho
in? Wine
term 00min
Can
35 the So
EQVETneC‘
some (16
have ac
“i log

htthg (

Sign

Obie

bet

4

 

 

155

of convention, a relation to action or practice is maintained.
As noted in the discussion of practices, the essential
element of an accounting practice, that which allows the
accountant to act with a principle in hand, is the adoption
of an adequate definition of the basic concepts in that prin~
ciple. To be able to record an asset, the accountant has to

be able to recognize one. But there are many accounting

 

actions involving assets (their recording, valuing and
classifying in financial statements, etc.) which might
fit the definition of a practice. For the sake of order, and
as a matter of common sense, it seems the accountant must
utilize the same definition of the concept of assets. Hence,
the proposal that a convention is the definition of a basic
term common to many theories (and their principles).

Can logic assist in establishing conventions? No. Just
as the selection of a definition in a particular case is not
governed by methods of logic, so the decision to select the

same definition in several cases. It may be common sense to

 

have accountants utilize the same definitions; but, no method

 

of logic will produce an ironclad convention just as no

method of logic will produce an unchangeable definition.

Summary

The preceding sections have accomplished the first two

 

objectives of this chapter, to establish useful distinctions

between certain terms currently used by accountants which

 

‘diahow ident'
ins,and to so
investigations o
Trusts, postc
asnnnitutin

ii p the

and i
is a

ix;
~_/

Cont
rem
by'
nit
rat
on

156

will allow identification of specific logical interrelation—
ships, and to suggest the methods of logic appropriate for
investigations of matters classified in the given categories.
Concepts, postulates and principles have been characterized
as constituting a theory in the form of an axiomatic system.

1) A theory, then, is a specific type of thing,
and the axiomatic method of deductive logic
lS applicable.

2) Concepts are basic terms of a theory and
require definitions. Terms are not selected
by means of logic. In constructing defi—
nitions of the concepts, inductive generali—
zation and analogy may be of some small help)
but no other method of inductive or deductive
logic appears appropriate. And, definitions
cannot be considered true or false.

3) Postulates are the assumptions of a theory.
Analogical argument and inductive generali—
zation can be helpful in suggesting postu—
lates, but cannot establish them as true.

Likewise, another system constructed to argue
to a conclusion which is a postulate in some
System cannot establish the truth of the pos—

tulate.

4) Principles are theorems of a theory, and, as
such, are the result of the application Of
the designated deductive rules of inference
in a given axiomatic system. They Will be
true ;p_g_system.

5) Practices are actions in conformity With th?
principles of some theory by adOpting a defl”
nition of the terms in those prinCipleS' NO
methods of logic relate to the selection Of a
practice. The accountant may put his argument
for the adoption of a particular definition in
the form of still another axiomatic syStem:
but such a move will put him in a position of
having to define his new concepts and JuStlfy
his new assumptions. Judgment is Still the
primary factor in practicetflndement Of what
is to be accepted (what theories, what defl—
nitions), and what is suitable in the Circum—

 

 

stances
ti Convert
common
tical
are, ‘
ment.

There ha
hgterninol<
pnsent prop
proposal lit
vhich speci
he areas

natenent
none ap

The
inconne

Tith Mr

its mad

limits:

honey
save

nE

\

 

 

stances.

6) Conventions, as definitions of basic terms
common to many theories, are related to prac-
tical considerations, and like practices,
are, in the final analysis, a matter of judg—
ment.

Comparison with Other Proposals

 

There have been other proposals for organizing account—
ing terminology, to all of which this writer considers the
present proposal superior. The major advantages of the present
proposal lie (1) in the ability to isolate those areas in
which specific methods of logic can be applied, by defining
the areas in terms of logical relations, and (2) in the
statement as to those areas to which some method of logic
can be applied.

The proposal of Mr. Spacek was discussed briefly above
in connection with principles. This writer agrees heartily
with Mr. Spacek's warnings that "unless the arguments
are made available for study, we cannot hope to advance our
profession,"1 and,

. without a clear statement of the reasoning

on which the principles rest, no community of

professional thought can exist.

However, the proposal made by Mr. Spacek would seem to have

several serious drawbacks if given consideration in practice.

It appears that all statements on accounting matters save

 

1Spacek, ”Need for an Accounting Court,” p. 375.

2Spacek, "Solution to the Principles Dilemma,” p. 279,

 

 

"oat accounting

one general st
seen, nor is it
add practices.
one conpletelp
ants today.
ical method,
principles a
accountant,
“Tint to
Like ‘
it". Stead1
undeiinab
For hr. 5
accouwti
ture Q‘p
siennar
account
Site f
his ".
Sitar
Cage

ton

 

 

158

that accounting should be fair, would be relegated to the
same general status. What is a principle is not easily
seen, nor is it possible to distinguish between principles
and practices. Such a distinction seems necessary unless
one completely ignores the usage of these terms by account—
ants today. Further, there is no indication of what log-
ical method, if any, could be used for ascertaining that
principles are in accordance with the postulate. The
accountant, or perhaps, the accounting court, is still left
subject to psychological persuasion.

Like the Spacek proposal, the structure suggested by
Mr. Steadl is based upon a single, undefined (and perhaps
undefinable)notion, which is both ethical and subjective.
For Mr. Stead, there is ”one all—pervading principle of

accounting,”

and that principle is Integrity. The struc—

ture or hierarchy proposed includes a principle, canons,

standards and rules, the whole group together being called

accounting doctrine. His principle is "an invariable cohe—

sive force," his ”canon” a ”goal derived from the principle,"

his "standard” a "measure against which conduct must be

squared,” and his ”rule” a “detailed mandate in a particular 9
case.” He supports and justified his proposal in the

following words:

The value of this approach is conceived to be
the ability to derive the lower orders from those

 

lStead, op. cit.

 

 

immediately
formulatiqo
dictates l
hiortunately,
cred against,
her. The qua
derivation an
can one evalo
it er may ta
sorts of thj‘
in are mil
to their co
derahly £3
The ‘
concepts
tonventp.
assumptj-
theen t
mild

tier,

most (

tloos

tied

disc

as

\

A

 

FTIIIIV' 4___—_______________________T::::IIIII'II

159

immediately above and thus constantly check any
formulation in any ethical framework against the
dictates from which it is supposed to stem.

l

Unfortunately, such phrases as "derived from,’ and "mea-

I 1

sured against,‘ and "squared with,’ are not explained fur—
ther. The question remains as to HOW one goes about this

derivation and, in the absence of methods of doing so, how
can one evaluate the path that a given "derivation" by ans
other may take? Nor is there any way to determine what

sorts of things fit into his categories, unless, of course,

 

you are willing to simply accept his personal opinion as

to their content. Moreover, his categories stray consi—

 

derably far from present usage of most of those terms.

The Paton and Littleton monograph2 offers a structure of
concepts and standards, and also such things as principles,
conventions, practices and procedures. Concepts are
assumptions; standards rest on concepts. The distinctions be—
tween the remaining terms and their interrelationships are
difficult to determine and appear to vary on occasion. How-
ever, the usage of most of these terms does not conform to
most current usage. Nor does the study offer logical rela—
tionships between the groups, or methods of moving from one 1
group to another. And, examination of the accounting matters
discussed under the various headings does not give an answer

as to the methods used by the authors. Further, there is no

 

lIbid.

2W. A. Paton and A. C. Littleton, Introduction.

IIIIIIIIIIIIIIIIIIIIIIIIlllll---_____l

 

 

:ethod of ident
:entuelonsS in
those used as
Professor
pliers a stru
concepts, to
cards, and a
principles :
tested for
used are n
in general
with. Th
ditional
limited
clear.
interre
Ponies:
to son
it do:
minoi
tion

uhir

 

160

method of identifying into which group some particular state—
ment belongs unless that statement happens to be one of
those used as part of the study.

Professor Littleton's Structure of Accounting Theory
offers a structure of definitions, reasons, principles and
concepts, to which are added conventions, practices, stan-
dards, and a number of other levels. It is suggested that
principles are inductively derived and somehow deductively
tested for interrelationships. The types of induction to be
used are not entirely explained, and the problems of induction
in general (as noted in Chapter IV) are not fully reckoned
with. The use of deduction is suggested as limited to tra—
ditional syllogistic organization, and where this very
limited deductive method is to be considered useful is not
clear. There appear to be no other discernable logical
interrelationships between his categories. In particular,
Professor Littleton suggests a category of concepts (which,
to some extent, coincides with the present proposal in that
it deals primarily with broad understandings of basic ter—
minology), but there is no indication as to a single rela—
tionship between, for example, concepts and principles,
which can be analyzed and examined by means of some method of
logic. If for no other reason, the proposal is somewhat
weakened by the very multitude of the terms used and

considered significant.

 

 

 

 

 

iroiessor V
imas'which coo
principles, pro
ciples are prep

“hen logical
uological re:

suggestions 2

levels of th

iicult to or
logical gig
inter does
ductive st
view is,
made thro
“°i lust
A one
ment re
group c

of log

crap:

lute

 

161

Professor Vatter has offered a ”hierarchy of accounting
ideas" which consists of objectives, conventions, doctrines,
principles, procedures or methods, and postulates.1 Prin—
ciples are presented as distinctive in the hierarchy for
"their logical significance.” Conventions are said to have
no logical reasons, but may be, in fact, arbitrary. But no

suggestions are made for logically relating the various

 

levels of the hierarchy in any strict sense, and it is dif—
ficult to understand where his principles, then, get the
logical significance he asserts that they possess. Professor
Vatter does remark, ”Accounting is not a purely logical, de—

ductive structure similar to Euclidean geometry.”2 This

 

View, is, to a great extent, in accordance with the comments
made throughout this paper that the whole of accounting is
not just one big logical system such that everything is bound
by rigid rules of deduction. But Professor Vatter‘s state—
ment refers to ”accounting” and not a specific level or
group of levels of his hierarchy. And, the role of methods
of logic at any given level in his hierarchy is left gen~
erally unexplored.

The position expressed by groups of the American Insti—
tute of Certified Public Accountants is, perhaps, the most

well—known at the present time. This proposal envisions a

 

1William J. Vatter, ”Postulates and Principles,”
%%E£nal of Accounting Research, I (Autumn, 1963), pp, 179-
7.

2Ibid., p. 186.

 

—

 

structure of poo
guides for appl
captions upon
ride some four
iolihe the Sp:
that accounti
particular a
terms used 2
”winter to 0
terms, imp
aneat ex
@0381 doe:
as to how
WStulatv
missing
patio“
Strum
with r
lumen:
tulat
that
the:

has

 

r—————————————————::::Jlllll|||llll

162

structure of postulates, principles and rules or other
guides for applying the principles. Postulates are as—
sumptions upon which the principles rest, and are to pro-
vide some foundation for developing principles and rules.
Unlike the Spacek proposal, it provides for the simple fact
that accounting is not made up of simple statements relating

particular actions in particular situations. Moreover, the

 

terms used as names for the categories in the structure

 

appear to correspond with present applications of those
terms, though current usage has been, perhaps, influenced to

a great extent by the position of the Institute. The pro-

 

posal does present two difficulties. It lacks a suggestion
as to how to go about deriving principles and rules from
postulates (which admittedly is an extremely important
missing link), and it fails to indicate a satisfactory expla—
nation of what place ”practices" or ”procedures” hold in the
structure and in accounting in general (are these coincident
with rules?). Too, the position may be questioned for its
immediate requirement (or supposition) that the set of pos—
tulates is small in members.1 Still, "few" (like fair, ade—
quate, sufficient, etc.) is, of course, a relative term, so
presumably specific studies which purport to set out the

basic postulates need only to reflect the individual‘s inter-

 

lThat question is, in fact, raised by a correspondent
to the Research Division of the Institute, and the comments
are reprinted in ”Comments on ‘The Basic Postulates of

Accounting,‘" Journal of Accountancy, CXV (January, 1963)
pp. 44—55.

 

—_ l

 

notation of wh
One of the
been that they
examples of ti
posted in the
possibility <
ill appear or
that one or
done in a s
”Dressed ‘
ihsals wer
its categ
lostuiatg
Attract'l
that) for.
SWAT c
Comm,

eiampp

mode

lies

the

163

pretation of what constitutes "a few."

One of the major difficulties with most proposals has
been that they were accompanied by extensive discussions and
examples of things that fell into the various categories sug—
gested in the structure. For example, the case for the
pOSSibility of a set of most basic assumptions in accounting

may a . .
ppear much stronger if it is unhampered by a statement

that on
e of those assumptions is that accounting should be

done ' -
in a stable unit of measure. Thus, the position

ex re
p seed by the AICPA received little criticism until pro—

posals -
were made regarding the statements which fell into

the c '
ategories suggested. Similarly, the suggested one-

postu
late system of Mr. Spacek may appear particularly

attra '
Ctlve because of the specific postulate offered rather

than f ' - -
or its ability to add logic or reasonableness to the

stud -
y Of accounting problems. It should be noted that no

controv ' -
erSial accounting terms or statements were used as

exam ' -

ples in the discussion of the terminological proposal
made ' ' d

in this chapter. This writer oes not believe that the

erally corresponds to

present state of accounting ideas gen

the - -
proposal in this paper, and the proposal is ”0t a“

att - -
empt to justify that present state. The pTOPosed defi—
ccounting

ni ‘
tions may be considered appropriate to the a

theory of the future.

Nevertheless, there haVe been some particular researCh

lication of logic or

e
ffOPtS which have either claimed an app

 

 

been considered
ooh these effo
relation to W
In a great ext
structure con
Toe proposal
suggests the
logical metk
lates, cone
and W the

10 determ
logic, an,

Toe next
identifi
methods

most ad

16M

been considered such by other members of the profession. As
such these efforts are deserving of special consideration in
relation to the present proposal and related suggestions.

Tb a great extent, each study of the group accepts a basic

Structure consisting of postulates which ”underly” principles.

The probosal in this chapter accepts the same structure, and

suggests the applicability and appropriateness of particular

logical methods to specific areas of investigation (postu—

lates, Concepts, etc.). Having described in Chapters Ill

and IV the logical methods available, it should be possible

to determine if these studies do utilize some method of

logic, and if so, to identify the method or methods used.

The next chapter will be devoted to such a determination and

identification, together with an indication of whether the

methOds used in the studies coincide with those proposed as

most advantageous for the task undertaken.

 

 

In very
investigatiox
that logical

can be an ll
TY

are m
world
sider

by cc
lootessor

Research

watched

the em

ottlice

F'tttttt ‘__""""'IIIII-IIIIIIIIIIIIIII‘IIII

 

CHAPTER VI
CURRENT ”LOGICAL" STUDIES EXAMINED

In very recent years, there have been some accounting
investigations which Professor Storey calls "ample evidence
that logical studies of accounting are not only possible but

can be an important contribution."1 He offers the opinion:

 

The real breakthroughs in accounting theory
are more likely to come from . . . individuals
working more or less alone on what are now con—
sidered the fringes of accounting and uninhibited
by consideration of accounting practice.2

Professor Storey refers specifically to the early Accounting

 

Research Studies, the work of R. J. Chambers, generally sum-

marized in his Towards a General Theory of Accounting, and

 

the efforts of R. Mattessich, which culminated in the recent
publication of his book, Accounting and Analytical Methods.

An investigation approximately equal in scope with these
is The Statement of Accounting Postulates and Principles by a
Study Group at the University of Illinois, which, its authors
claim, is an attempt "to express and explain in concise terms
the core of a general theory of accounting," and in which

they are concerned with "the realism of our conclusions as

 

lStorey, op. cit., p. 62.
2Ibid., (Footnote) p. 62.
165

 

 

all as their it
that this study
I. D
eurScorey, L03
an effort to h
H
group.
to equal
son in an as
locouutiog “.
that the st
theory with
Each

and 3 haul

from uhl
Organizz
\‘Tlter

F,llill
166 lllll

well as their internal consistency and logic.”1

It appears
that this study should fit the qualifications set by Profes-
sor Storey, for the Group notes that its concern "was not i
an effort to make our conclusions palatable to any specific
group."

An equally ambitious study was made by Dr. Marvin Carl—

son in an as yet unpublished doctoral dissertation entitled

Accounting Theory as a Logical System. Dr. Carlson states

 

that the study "is an experiment in constructing accounting

theory with the aid of formal logic.”2

Each of these five studies (Research Studies Numbers 1

 

and 3 have been considered as a unit) accepts the basic idea

that there are, in accounting, some assumptions or postulates

 

from which other accounting statements are derived, which
organization is essentially the same as that proposed by this
writer in Chapter V. The objective of this chapter is to
analyze each of the studies in the following general pattern:

1) determine the components of the struc-
ture suggested;

2) ascertain what method or methods of
logic the authors claimed were utilized;

 

3) identify the method or methods of logic
which appear to have been used; and

4) compare the method used to that sug—
gested as most appropriate for the type
of problem being investigated.

 

lPreface, page unnumbered.

2Marvin Lee Carlson; "Accounting Theory as a Logical
System,” Unpublished Ph.D. dissertation,(Commerce, Univer—
sity of Wisconsin), p.

 

IIIIIIIIIIIII'llllllllllll---_____

 

'1. connection wit
citation utilizl
Is noted in Chat
fusions} are t
i.e., that the
sly conclusio
:h.t method 0:
:Zusions. Th
hat extensi'
hue unstate

53 dies wou

is it Shggc
“Elm t

{“90 ores

t9 examh

Current

least 9;

in WAG

Wind.

“Shut

are r

Paad

not

167

‘ hiconnection with the identification of the method of argu-

mentation utilized, some preliminary comments are necessary. '
As noted in Chapter II, only arguments (premisses and con—

clusions) are the proper subject matter for logical methods;

i.e., that the subject matter of logic is argumentation. If

only conclusions are presented, there is no way to determine

what method of logic, if any,was used to arrive at those con—

clusions. The writer does not wish to deny the possibility

 

that extensive reorganization (and perhaps the addition of
some unstated premisses) of the discussions in some of the

studies would produce an identifiable logical argument. Nor

 

is it suggested that the authors used no logical methods

 

prior to the preparation of the studies for publication, and

then presented only conclusions. But the writer is committed

 

to examination of the selected studies as they are written.
Current interest in the use of logic by accountants is at
least partially prompted by a desire to be able to ascertain,
in some objective manner, that reasoning behind a position is
sound. And whatever methods of logic may have been used in
argumentation taking place outside the final published study
are not available for examination by this writer or any other
reader-

When arguments are identified as inductive, there will
n0t, generally, be a discussion of the strengths and weak—
nesses of the arguments. As noted in the discussion of in-
duction (Chapter IV), evaluation of an inductive argument

rests with evaluation of its premisses——its evidence. Such a

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII-_____

 

a
.fiytnooloes ev
3% prejudlceu ‘

ufischafacterl

.4

its objective
then an 2
writer‘s firs
ence to a se‘
'mtioos deve
seen in the
ally unsuc<
commonly u
he found ‘
that some
their no
simple 1
tainly

might

are e)

the 1

been

that

com

in

F:

A

l 68 -

task involves evaluation of its completeness, its freedom

from prejudice, and its relevancy to the proposed conclusion.

It is characteristic of induction that such argumentation is *
subject to individual evaluation with respect to such matters,

and identification of argumentation and not its evaluation is

the objective of this chapter.

When an author has claimed deductive methodology, the
writer's first efforts were directed at locating some refer—
ence to a set of allowable inferences which guided the deri-
vations developed by the author. These efforts, as will be
seen in the discussion of the individual studies, were usu—

ally unsuccessful. The writer recognizes that there are some

 

commonly used rules of inference, descriptions of which could
be found somewhere outside these studies. It is possible

that some authors did not desire to increase the length of

 

their works by inclusion of a list of references. Yet, a

Simple reference to the location of such a list would cer—

 

tainly be sufficient for most readers; and, such reference
might add considerably to the weight of whatever positions
are expressed. For those readers who would not investigate
the list of rules used, the very fact that such a list had
been adopted by the author would give the reader confidence
that the discussion was formally sound. The reader could
concentrate his efforts on the substance of the argument.
This advantage is exactly what characterizes deductive logic.
For those readers who wgulg_question the validity of the

argumentation, a list of rules or a reference to it is an

IIIIIIIIIIIIIIIIIIIIIIIIIIl---_____i

 

imolute necessit
are of all the '
(erect osycholc
:my logic-mind
fer attentiom
'mlidity of th
thhtioe inf.
mismatics,
lliate. But
threading t

Sim
heat of
atttmpt
Fehurt
3%th
0lller
Sitoe

Vlde

' 169 ‘

absolute necessity. As was noted in Chapter III, the accept—

ance of all the premisses in an argument may be useless

(except psychologically) if the argument is not valid, and i
many logic—minded readers of the study might wish to direct A
their attention initially to ensuring themselves of the

validity of the argumentation. Perhaps, if the rules of

deductive inference were as commonly known as the rules of

mathematics, the failure to refer to them would seem appro—

 

priate. But it is unlikely that the accountants interested
in reading the accounting studies examined here are familiar

with commonly used deductive inferences.

 

Accounting Research Studies Numbers 1 and 31

Structure Terminology

 

Since these studies were undertaken at the direct re—
quest of the AICPA Council, they understandably represent an
attempt to follow the structural format envisioned in the
rePOrt by the special committee on research programs. That
structure consists of postulates, principles, and rules or
other guides for the application of principles in specific
situations. ,Postulates are said to be assumptions, and pro~
Vide a foundation for principles.

In addition to the three basic levels mentioned above,

the studies contain references to some things called

 

1The two studies are here discussed together as they
were intended to present the parts of an integrated whole.
This position is made clear in the Preface to Accounting
Research Study No. 3.

 

 

IIIIIIIIIIIIIIIIIIIIIIIIIllll--_____,

 

ll
concepts .

Ema

to tudy No. l E

to materiality
assets (p. l9).
ll. l8), and p
To. 3 indicate
arms and cor
cl definitio'
fit into mhj
hits are t‘
35W} betwe

115 that (:0
“AM moo?
rlher

'v‘llllth 3‘1‘

lot all
discuss

the has

9th Q

Which

”the

Stat.

Qati

Dar

170 II

'boncepts." Examples of things in this category according

to Study No. l are fairness (p. 3),1 going concern (p. 39),
and materiality (p. 47). Study No. 3 mentions concepts of
assets (p. 19), economic benefits (p. 21), accrual accounting
(p. 48), and profit (p. 53). Though the authors of Study

No. 3 indicate that "certain definitions are given to key
terms and concepts" (p. 53) and then proceed to give a series
of definitions, they do not state which of the things defined
fit into which category. In any case, it appears that con—
cepts are things to be defined. The only relationship indi—
cated between concepts and postulates, principles and rules

is that concepts are some of the words used in the statements

 

which would fall into the latter three categories.
There is one serious matter of terminological conflict
which significantly affects the examination in this paper.

Net all the ”principles" set forth in Study No. 3 will be

 

discussed in this paper because the writer contends that on

 

 

the basis of the authors‘ own usage of the terms identifying
each category in the structure, those ”principles" given
which relate to pricing or valuation do not fall under the
authors’ heading of principles. Rather such comments or
statements represent "rules or other guides for the appli—
cation of principles to specific situations.” The sup—

port for this contention follows.

 

1Moonitz, The Basic Postulates. For the remainder of
this chapter, the specific page references will be incor—
porated into the text which will make clear what source is
being referred to.

IIIIIIIIIIIIIIIIIIIIIIIIlllll---_______,

 

hinciple A

:ncedure, there
l

. . u
this . . -

specific differ
that such a
toe statement
and procedure
ciplcfi in th
cedars“ and
“Tiling int
“procedure
iouentory.
aoo curre
tussion c

to money

 

 

171 Ill

Principle A, Study No. 3, reads, in part, ”any rule or
procedure, therefore, which assigns profit to a portion of the
l
1!

whole There is no indication in either study of a l

specific difference between rules and procedures, but even
without such a differentiation the statement is significant.
The statement indicates that the authors consider that LEASE
and procedures are separate fppp and pplpy_the level of pp;_—
ciples in the total structure. The usage of the terms ”pro-
cedure" and "rule" in the studies indicates the type of things
falling into those categories. In Study No. 3, the term
"procedure” is used with respect to alternatives for pricing
inventory——alternatives such as net realizable value (p. 27)

and current replacement cost (pp. 29 and 34). In the dis- .

 

cussion of the measurement of assets which are money or claims
to money, the statement is made ”as a general rule, the

valuation of these assets should be based . . .” (p. 24,

 

Underscore added.) Later, in the discussion of fixed assets,
the authors call it a procedure to reflect the result of
"buying cheap” and "using dear” assets through
. . . the use of the current (replacement) costs of
the services rendered . . . and the separate classi—
fication of the related gain or loss. (p. 34)
Reference to Study No. 1 reveals the same usage. Discussion

of consistency (pp. 43—44) reveals that procedures are to be

consistent, and the example given of a change in procedure is

 

1The text of the postulates and principles examined is
given in Appendix A for reference.

 

acreage in the b
:2 conservatism l
:rccedures of as
.,._..——
“set of pricing
grimciple D in
Viper pricing
an intent to t
sponyms,
The wri‘
3.th“ WEI,

“irimcipie

Cations
the Oll‘
"once (
minat
Tesla.

he“

172 III

a change in the basis of measuring revenue. The discussion
of conservatism (pp. 47—A8) speaks primarily of conservative
procedures of asset valuation. That study also refers to a ’
"set of pricing rules.” (p. 55) Finally, in the statement of
principle D in Study NO. 3, the authors use the phrase ”the
proper pricing (valuation) of assets . . .," which evidences
an intent to use the words ”pricing” and ”valuation” as
synonyms.
The writer concludes that since the authors consider
valuation and pricing synonymous, and since the examples of
procedures and EEEEE are ways of pricing, pricing and valu—
ation must be a matter of procedures and rules. But those
"principles” in Study No. 3 regarding measurement relate to '
pricing or valuation, and present specific suggestions for
the application of a general statement (e.g., that all assets
and liabilities should be recorded) to specific situations.
As such, these comments, or ”principles,” meet the qualifi-
cations for membership in the structural category called, at
the outset of the studies, “rules or other guides." Accept-
ance of this position has led the writer to limit the exa—

mination of principles in Study No. 3 to those which do not

 

relate to specific suggestions as to how something "should
be measured” or priced. On this basis, principles D and F
in their entirety, and those portions of principles E and G
which propose specific pricing suggestions have been omitted

from consideration.

 

 

histology Claim

irofessor pt
"teeny reliance
inelopment of
Later he indict
so in the to
influences“ d
late concern?
'lniortunatel
peatedly st:
strident, at
and the eu
etident pt
those trt
must con
sentient

nit) U16 k

l

‘PT-‘lnr
illi- A

i— 173 fi

Whthodology Claimed
Professor Moonitz himself stated, in Study No. 1, that
Wmavy reliance must be placed on deductive reasoning in the i
development of accounting postulates and principles." (p. 6)
later he indicated1 that both induction and deduction were
used in the two studies, and that the postulates were "the
inferences" drawn by means of induction from ”the mass of
data concerning the environment in which accounting functions.”2
Unfortunately, in the same article, Professor Moonitz re—
peatedly states that the postulates of Study No. l are self—
evident, and then proceeds to discuss the search for them
and the evidence reviewed. As already explained, self—
evident propositions are those that require no support, those
whose truth is apparent to anyone confronted with them. One
must conclude that Dr. Moonitz meant, not self~evident as—
sertions, but perhaps assertions to which no violent objections

would be raised. In any case, given the claim of induction

 

lMaurice Moonitz, ”Why Do We Need 'Postulates' and

'Prificiples'?” Journal of Accountancy CXVI (December, 1963),
pp- 2‘46-

2Ibid., p. AM. This statement appears in contrast to
the comments of some readers of the studies. Compare, for
example, with the comments of Professor Metcalf who speaks
Of "the heavy reliance upon the deductive process evidenced
by " Study No. l.("The ‘Basic Postulates' in Prospective,”
The Accounting Review, XXXIX [:January, 1964 , pp. l6~2l.)
Also compare with the editors of the Journal of Accountancy
who applaud Professors Moonitz and Sprouse for having worked
in "accordance with rigorous rules of logic” to ”achieve a
logically consistent set of principles.” (”The Approach to
Accounting Principles,” CXIII [ll/lay, 1962:] , pp. 36—37.)

 

 

i: the inoestiga
'2': claimed in tl
llo speciii
arriving at de:
:Laim any use
all other gut
concern them:
The only are

in the deve

A res
that the
res so mm
liesente
Althea

lilth 03
“Allie
tulag

lop p

den

its

 

l7A I

in the investigation of postulates, deduction, then, is to
be claimed in the development of principles.

No specific claims are made regarding methodology for
arriving at definitions for concepts. Nor do the authors
claim any use of logical methods in the development of rules
and other guides, for it was not intended that the studies
concern themselves with the things falling into that category.
The only areas, then, where logical methods are claimed are

in the development of the postulates and the principles.

Methodology Observed—Study No. l

A reservation was made at the beginning of this chapter

 

that the only basis on which one can identify the type of
reasoning behind a conclusion is the information actually
presented for examination. This reservation is especially
applicable to the discussion of the two research studies.

Much of the material in these studies is related to what

 

"follows from” or is "involved in" or "implied by” the pos-
tulates and principles prOposed rather than to the argument
£9£_the postulates or principles themselves.

Consistent with Dr. Mbonitz‘s claim, the use of any
deductive method is generally not in evidence in Study No. 1.
Absent from the study is any formal argumentation which could
be identified as deductive. However, special attention must
be given to the second sentences of postulates A-3, A~4 and
A-5, which Professor Moonitz claims are ”implied” by the
first sentences in each of these postulates. The author calls

these implications "simple and direct, requiring nothing

L

 

 

"are than “913
motive lOAI—C'
The write
:2 the impllc
Since the ill
references to
o._e interns
term to the
Leading to
direct. l

ence shou

spe
the simn

somethi

Simil
on dl

dire

 

175 III
nmre than explicit statement by way of ‘proofl,”(p. 26)
The claim in these cases, then, is to some method of de—
ductive logic.

The writer cannot identify what logical inferences led
to the implications stated as being so simple and direct.
Since the first parts of each of these postulates make no
references whatever to "a clear identification” of anything,
pppp intervening propositions must have contributed this
term to the conclusions; and the method of logic capable of
leading to such conclusions can have been neither simple nor
direct. An example of a simple and direct deductive infer—

ence should illustrate this fact. From the postulate (A—3)

 

 

 

 

Economic activity is carried on through
specific units or entities.

the simplest and most direct inference is a proposition
something like the following:

If there is a report on some economic activity,
there is an entity which carried on that activity.

Similarly, from postulate A-A, ”Economic activity is carried

H

on during specifiable periods of time, the simplest, most
direct implication is
If there is a report on some economic ac—

tivity, there is some period of time during

which that activity was carried on.
The implication claimed in postulate A-5 must be even less
simple and direct, Though the postulate specifies money as
the common denominator used to measure goods and services,

the discussion of the postulate holds rather that it is

 

"exchangeability" of goods and services which is measured

IIIIIIIIIIIIIIIIIIIIIIllll---______i

 

items of money
:i‘mey is impli'

225m eabilit

   

2: goods and se:

22106 of deduc
identific:
rave been usec'
the special 0
the fact thai

1’ figment

‘ifilng 0
"MG 211'.
lfismlat
Stated‘
dence (
in the
3‘99 ti
3W c
We
1L5

mm

9Q

’52

A

176 ‘

Miterms of money. Thus, if anything about reports in terms

of money is implied by the postulate, it is about reports on

exchangeability of goods and services and not about reports *

on goods and services per se. The writer cannot identify any I

nmthod of deduction which would produce the claimed implication.
Identification of specific methods of induction that may

have been used to arrive at individual postulates (other than

the special cases just discussed) is further complicated by

the fact that each postulate is not preceded by a discussion

or argument especially intended to support it. What appears

to be intended support of some individual postulate may be

found in more than one place in the study, while for other

 

postulates no support can be found at all.

Detailed search of the monograph itself reveals no
listing of observations or report of such observations which
would allow identification of the method of logic leading to
postulates A—l, A—3, A—u, A—5, B—3 and 0—3, as they are
stated. Perhaps the author felt that no presentation of evi-
dence (or the manner in which he collected it) was necessary
in these cases. We might even suggest that these postulates
are the ones Professor Moonitz considered self—evident. In
any case, no evidence is presented. It is interesting to
note that if one were to accept as evidence for postulate
A~5 the author's statement ”The point of agreement is that
money is used almost universally as the common denominator in
economic and business affairs” (p. 18),the inference to the

postulate would be invalid. For, the statement quoted

recoghizes hegat
:‘reoosihlate, ‘
:hhegaiive ihs
There are

mo the test:'
for which tha
ire fihal as:
notations f
3-14 and 0—5
View state<

Zoe 905ml

(

hela

reta

give

the

shr

The (we.

5‘3 omit

SWOT

We a\

h. h

 

177 |

recognizes negative instances of the generalization made in
the postulate, while the postulate asserts that there are
no negative instances.

There are clear-cut cases of the use of argument based
tumn the testimony of others. That is, there are postulates
for which that part of the discussion which at all relates to ‘
the final asserted conclusion in the postulate involves only
quotations from other authors. In this group are postulates
0-4 and 0—5. To this list could be added postulate 0—2 if it
were stated in the declarative rather than a normative form.

The postulate, as stated, is:

Changes in assets and liabilities, and the
related effects (if any) on revenues, expenses,
retained earnings, and the like, should not be
given formal recognition in the accounts earlier

than the point of time at which they can be mea—
sured.

 

The question of what meaning to attach to "and the like" will
be omitted here, though an interesting question indeed. In
support of a postulate that formal recognition i§_ngt given,
the author refers to an article in the Journal of Accountangy
(p. Ml). To the development of the postulate in its original
form, however, the writer can attribute no method of induction.
Because of the absence of formal argumentation in the
monograph, the idea of evidence and observation may be very
broadly interpreted to include the giving of some examples,
however few. On the basis of this broad understanding of
eVidence and argumentation, postulates B-l, B—2 and B~4 can
be identified as the result of the method of inductive gener-

alization. This list might be extended to include postulate

 

L

 

“iii it were 5'
“mm The post\
In the

entity '5hC
indefinité
ehtity ha

as hemaif
importing a '
tamer suggef
geeseht acoo
fosthiate a:
the ho met
The r
ohesehtatf

soh Moohi

oi distr'

178 I

04.if it were stated in declarative rather than normative
fbrm. The postulate, as stated, is as follows:

In the absence of evidence to the contrary, the
entity should be viewed as remaining in operation
indefinitely. In the presence of evidence that the
entity has a limited life, it should not be viewed
as remaining in operation indefinitely.

&uxmrting a postulate that the entity is viewed in the
manner suggested, the author offers several examples from
present accounting practice. To the development of the
postulate as stated above, however, the writer can attri—
bute no method of logic.

The remaining postulate, A—2, is peculiar, both in its

presentation and in the discussion connected with it. Profes—

 

sor Moonitz claims that it describes "the dominant method"
of distribution (p. 22). This claim gives the impression
that he has made observations of various methods of dis—
tribution. Earlier he states simply:
Furthermore, the goods and services produced,

are, for the most part, distributed through exchange ‘

of some sort, and not consumed by the producers

themselves. (p. 8)
Two examples are offered of g99§§_(but “0t services) WhiCh
had been consumed directly in the past but are now being dis~
tributed through exchange. Under a broad interpretation of
what constitutes evidence, the writer (with reservations)
identifies the method of logic used in developing the postu-
late as statistical generalization. A less generous inter-

Dretation would attribute to the postulate the method of

argument based upon the testimony of experts—in this case,

 

L

 

hiessor Moonitz

mgr-ooh .

'93:'rohoio '

 

Thohg'h the
iateiooihg the
hate my ref
zshooessfhl .
enlist in ti"
ieihttihe in
the writer i
set ih the

Ytiehehce

70 tho

MIC '11

 

179 I

ITofessor Moonitz. lkicather method is apparent from the

monograph.

hhthodology Observed—Study No. 3

Though the authors claim that deduction was used in
developing the principles from.the postulates, efforts to
locate any reference to the rules of inference used were
unsuccessful. The seriousness of this omission was noted
earlier in this chapter. Despite the absence of a list of
deductive inferences which the authors allowed themselves,
the writer has attempted to identify the method of logic

used in the development of the principles, as stated, by

 

 

reference only to the study. The writer found no formal use
of the postulates in the development of principles despite
an explicit statement that ”The ’basic postulates of
accounting' developed in Accounting Research Study No. l
are an integral part of this statement of principles.” (p. 55)
As stated earlier, discussion of Study No. 3 is limited
to those principles which are not measurement rules or
pricing suggestions. The remainder of the principles are A,
B, C, the first sentence in E, the first two sentences in H,
and that portion of G not dealing with pricing. Principle A
stands out from the rest for two reasons. Unlike all the
Others, it is expressed in the declarative form. And, the
manner in which it is supported in the monograph is unusual.
The method of logic which appears to have been used in the

development of this principle is argument based upon the

 

llIIIIIIIIgIIIIIII........--_________

:estimohy of othe
is: of quotatic
i. Yatoh top
grihciple state
The mohog
arr; method of
‘ze‘Loomeht of
to. groups or
he ihtehoed
ohihoioles,
stated pri
ieha‘ole a
we
‘3? Moo
of any j
0‘: )ng
Chaote

“as
has

 

180

testimony of others——an inductive argument. Premisses con—
sist of quotations from G. 0. May (pp. lO—ll, 13—14) and
W. A. Paton (pp. lO-ll), and a simple assertion that the

principle states a fact that is "universally true.” (p. 14)

The monograph contains no information which would allow

any method of deductive logic to be attributed to the de-

VelOpment of the remaining principles, either individually,

in groups or in totals. Some single sentences appeared to

be intended by the authors as support for some specific

prinCipleS, but the relationship of these statements to the

Stated principles must be an indirect one demanding consi—

derable additional argument and premisses to establish any

l2522§l_00nnection. Summarizing, then, With the exception

Of principle A, the monograph holds no evidence of the use
Of any lOgical methods. Regarding the exception, the method

Of logic used was inductive, and one which was noted in

Chapter IV as Often weak.

Com arison of Observed and

___E________________'_f’#

Proposed Methodology
The use of inductive methods was suggested in Chapter V

as appTOpriate‘for the investigation of postulates of a the—
cry; SPGCifically recommended were analogy and inductive and
Statistical generalization. To some extent, the methods
observed in Study No. 1 (if any method could be identified)
correspond to those suggested. But the discussion in the
monOgraph allowed identification of some postulates as
1r'eSLllting from arguments based upon self—evidence or the

 

 

testimohp' of exp
the“: the use of

etch, Which,

1'? principle s

to: o'oserved

‘3‘:th the p?
we logic?
“hat were
The 903)?“-

Islets sx

"9M

The

the

lBl
testimony Of experts. The writer considers it unfortunate
that the use of induction was extended to include these
methods,Which, as noted in Chapter IV, are very often with—
out merit, Though it was suggested that deduction (specifi—

cal ' -
ly aXiomatic method) was appropriate in the development

of ' -
princlplee, the use of deductive methods of any kind was

not - .
Observed in reaching the principles of Study No. 3 from

the
pOStulates of Study No. l. While it is conceivable that

bot - -
h the prinCiples and postulates could be formulated in

som -
e logical manner and proved consistent with each other,

th -
at exerCise would not be evidence of any implications of

th
e postulates or that the statements were on two different

1
eVels such as the postulate—principle scheme would claim.

Postulates and Principles of the
Il is Study Group

lino

Structure Terminology

The Study Group has stated that their work follows from

- . . the current interest in "basic POStulateS”
and ”broad principles” and because these terms
peoessarily occupy a central position in account—
log theory. (Unnumbered page f Preface.

The stUdy envisions a structured whole, called a ”general

theory Of accounting," consisting of postulates, principles
and concepts, Postulates are underlying assumptions (p. 6);
primeiples are ”basic propositions which express significant

(p. 23) Concepts, though not

relationships in accounting."
portant basic

defined in the monograph, appear to be im

terms in accounting.

 

influencing th
\l
anironment .
and procedure

which accomm‘

‘.v

‘.ietnodology
The St
regarding; i
“discover
in nature
T‘Popositi
deductiv‘
ndirecti
Tostula
“a com
the 1
incl .
of de
prop
for

not

 

182

The study presents another category called ”environ—
mental conditions,” which are much like postulates in
in ' —
fluenCing theory but are ”impressed upon accounting by its

enviro ” -
nment. (p. 6) A fifth category, consisting of ”rules

and H .
procedures, is necessary to cover the actual actions

wh' . .
10h accountants take in given situations.

Methodology Claimed

.__—.__———————————-—'

The Study Group makes several specific statements

r .
egarding the methodology used. Postulates are claimed as

III .
dlscovered' by inductive observations“ (p. 7), ”inductive

i
n nature” (p, 23)J and, even more specifically, ”self—evident

propositions.” (p, 7) They are ”not formulated through

dedu ' - . .
ctive reasoning.” (p. 6) Principles are ”universal,”

”directive in nature,” and ”more deductive in nature than
postulates.” (p. 23) Concepts are said to be ”formulated“ by
”a COmbination of observation and the application of deduc—
tive logic.” (p. 12) Observation is involved in their cris—

inal development, and the concepts are then tested by the use
of deduCtive logic to determine if they can be “derived
properly” from postulates and matched against other concepts
The authors

for contradiction and incensistencies (p. 12).

note that

unting principles are not neces—
counting postulates .

individual acco
tes do have consider—

sarily derived from the ac
[for] While accounting postula
able influence on principles, much of this is exerted
indirectly through the influence of postulates on
concepts, the latter often influencing prinCiples

directly. (p. 23)

 

“:9. authors adopt
statement of the
"se. follows h
"estatement l
ire: C,oal.“ (p
The write
radiates we
1914' are self
Eisthlates a
find both
Lied or to

"Jifohmenta

EEc'Jmpt 1C
W» of

w .
WWldt

 

183

Mm authors adopted a distinctive consistent form for the
statement of their accounting principles, which form, they

note, follows Professor Littleton‘s notion of principles as

"a statement linking a goal with the means of achieving

that goal." (p. 214)

The writer understands these comments to claim that

postulates were developed by the use of inductive logic (that

they are self—evident), that principles were deduced from

pOStulates and/or concepts, and that concept formulation in-

Volved both inductive and deductive methods. The methods

used or to be used in arriving at those things which are en—

Vironmental conditions or accounting practices are not ex—

plained, The authors note only that the environmental

assumptions must be accepted by accountants in their develop—

ment 0f accounting practices (p. 4), and that principles

"provide guidance” for the actual carrying out of the account-

ing functions of analyzing, recOrding and reporting (p- 32)-

Methodology Observed—
Postulates and Concepts

_____________________~#—
The Illinois monograph is like the AICPA studies in

. . . N
that much of the discussion is devoted to the ”significance

or Himplications” of the proposed postulates, concepts and
principles rather than to their development. And, as With
those studies, the reservations made at the beginning Of the
Chapter regarding the limitation of the examination to the
Pablished study itself apply to the Illinois monograph.

 

 

‘ l
he autnors
'.zszulates cannoi
{serving all ac

it: authors‘ as

2:2: evidence
fixative ar
Zia-Lite evid
i'femselves
5% assumed

3? survey
Elliever ,

3‘9 Writ

i“Wes s

184

The authors‘ claim of induction in connection with the

postulates cannot be denied outright. The impossibility of

observing all accounting and all accountants, coupled with

the authors' assertion (without reservations) that the postu—

H . . . .
lates a£§_ the assumptions which are made in accounting”

(p. 6) , describes a situation in which conclusions have been

drawn from necessarily incomplete evidence. Presentation of

that eVidence in support of the conclusions would constitute
inductive argumentation. It seems at least possible to accu—
mulate evidence in this case since the authors claim to limit
thEmselves to what i§_being assumed rather than what should

be assumed. The evidence might have consisted of some type

Of SUTVey of accountants to determine what they were assuming.
HOWever, the monograph gives no details of such a survey; and
the writer doubts that the authors really meant to give the
impression that one was taken (except, perhaps, among the

authors themselveS)-

Still, the impression of some type of survey or a
series of observations persists. In suppOrt Of POStUlate 1
(ACCOunting data and reports have validity and userlneSS for

Widely differing purposes.),1 the authors state:
A Considerable variety of interests finds h
accounting data helpful, some of them us1ng suc
data for one purpose, some another, . ...
(and, the data are used by interested partiea)
- . . for many decision—making purpOSGS- P-

'ndicate a survey Of

These comments do not, however, 1
es is given in

lThe text of the postulates and principl

Appendix B for reference.

 

 

mutants as to
12311: from a am
151%, could eel
the make. on
"ing an aeeom
of postul
generalizatlo
premisses be:
Pestula

table ente
accountabii

W1 Ftpert
“a.

meg

3181's agz

lates a

Eldere.

stats

hem

lent

ind».

lai

185

ewcountants as to their assumptions. Rather they appear to
result from a survey of the users of accounting data, and
fence, could only have led to an assumption which the data—
Lmers make. Only if the supplementary claim regarding its
being an accountant's assumption is ignored can the develop—
ment of postulate 1 be attributed to the method of inductive
generalization; and, even then, the monograph contains no
premisses beyond the statement quoted above.

Postulate 2 (Economic activity is engaged in by identi-
fiable enterprises and these enterprises constitute units of
accountability and centers of interest for accounting analysis
and reports.) is said to follow directly from the fact that

individuals and other interests for various
reasons find themselves interested in the status and

progress of given enterprises. (pp. 8-9)

Here again, only by ignoring the assertion that the postu—
lates are accountants' assumptions can this sentence be con-
sidered related to the development of the postulate as
stated. But the relationship cannot be identified as one of
Premiss and conclusion; and, there are no additional state—
ments presented which allow identification of any method of
inductive logic as having been used to arrive at the postu~
late as it is stated.

For postulates 3, 4, 5 and 6, the reader is presented
with only argument on the basis of testimony~~to be specific,
the testimony of the authors. For example, in support of
postulate 5 (An enterprise will continue without significant

change of environment and activities unless there is per—

 

 

siasiue evidence to the

a continuance

conditions and tr
the most reasonai

In, for postulate 6
an enterprise engage
oeriods within its l
Show state the as
nice.“ (p. 11)

In summary, t
no evidence which
to be attributed
humans, self-av

if exports, and
other than the
That sect
generally devc
Of those tone
ment in supp
gaming how
definitions
“Servat‘w
gem defi
Wei had

coneGifts

186 I

suasive evidence to the contrary.) the authors offer:
A continuance of approximately the present

conditions and trends has generally proved to be

the most reasonable assumption. (p. 10)
And, for postulate 6 (The flow of economic activity in which
an enterprise engages can be related to specified time
periods within its life on a meaningful basis.), the authors
simply state the assumption is "implicit in present prac-
tice." (p. 11)

In summary, the discussion in the monograph presents
no evidence which would allow any method of inductive logic
to be attributed to the development of the postulates save,

perhaps, self—evidence or argument based upon the testimony

 

of experts, and in the latter case, the experts are no

 

other than the arguers.

That section of the study dealing with concepts is
generally devoted to giving and explaining the definitions
of those concepts selected as important rather than argu—
ment in support of the proposed definitions or evidence re—
garding how the definitions were developed. Though some
definitions at first appear to be the result of extensive
observations of the items now considered instances of the con—
cepts defined, the definitions given are discussed as though
they had been developed from some argument as to what the
concepts should include. For example, the authors report:

In accounting, the term asset has a special
meaning. . . . It appears that assets are simply

all those items which are listed on the left
side of the balance sheet. (p. 16)

 

— t

 

his definition appea:

tossed accounting dat
the extent of such 0'
mother definition g
increappraising th
determine if any si
definitions are no
oosition that the
observations whic
But, the new def
Winchell chara<
indication that
“dated for m
Other than the
assets, It .1
duct‘mn 0mm.
three Charac
would have y
“ition of a
The cl
authors dc
this of
revenue (
are Sim:
accompa'
useful.

9.0“er

A

This definition appears to result from observations of pro-
cessed accounting data, though no indication is given as to L
the extent of such observations. Later, it is noted that

another definition given in the study would ”furnish a basis
for reappraising those items we now include as assets to
determine if any should be excluded." (p. 17) Since the
definitions are not the same, the authors must be taking the
position that their definition is not the result of the same
observations which led to the ”special meaning" quoted above.
But, the new definition offered follows a discussion of ”three
principal characteristics of assets.” And the authors give no

indication that the things whose characteristics were inves—

 

tigated for the formulation of the second definition were any
other than those things which are, by the present definition,
assets. It is certainly not clear how observation or in—
duction could have led the authors to decide that these
three characteristics were common to all assets since they
would have had to know the characteristics or have a defi—
nition of assets in order to select the things to observe.
The characteristics, definitions, or descriptions (the
authors do not always give definitions as such) of the con—
cepts of transaction, liability, ownership equity, revenue,
revenue charge, enterprise status and enterprise progress,
are simply presented by the authors for acceptance with some
accompanying recommending comment such as, "We find it more
useful." The discussion of realization (called sometimes a

concept, sometimes a test), like that for the individual

 

—

‘ of
modulates, conSists

need is what is accep

lethodology 0h served—

lor that part of
doles, some of the
mineioles study are
in the introductory
any rules or refer
themselves to use
omission. The au
the principles iv
mould allow idem

as having been
Qfission of some

Moreover, the
tives,“ stati'

“01‘ the defim
allow Comm

gmliter 11’
”tilt to he

that somet‘

means of 5

Offers no

if .

then to:

for exa

‘.

l

A

188 -

postulates, consists of simple assertions that what is pro-

posed is what is accepted. i

Methodology Observed—Principles

For that part of the monograph which deals with prin—
ciples, some of the comments pertaining to the AICPA
principles study are equally appropriate. For reasons noted
in the introductory section of this chapter, the absence of
any rules or references to rules which the authors allowed
themselves to use to reach their conclusions is a serious
omission. The authors do not actually present support for
the principles in the form of formal argumentation which

would allow identification of any method of deductive logic

 

as having been used. Rather, they direct themselves to dis—
cussion of some of the "implications" of the principles.
Moreover, the principles are like Dr. Moonitz's ”impera—
tives," stating what should be done. Neither the postulates
nor the definitions contain the relationships needed to
allow conclusions to be deduced in this form. As noted in
Chapter II, it is not possible to deduce statements of what
ought to be from statements of what is. If one concludes

If

that something ”should be found, it must be concluded by
means of inductiVe argument of some sort. The monograph
offers no evidence to this effect either.

If the principles are restated in the more familiar if—

then form, the normative aspect of each is eliminated. Thus,

 

for example, principle 1 might be restated as follows:

 

— _

ii accounting;
plete record of a
in changes 1

ur
In this form, the co

nesuiiicient, but

the achieved.1 As
envision what might

Leading to this pri'
m the monograph .2

above, the deriv at

___’—————-——

1is restated
of action (the de
of record) is $0“.
information abou
much a record is
also. it does v
able informatiO'
of record descr
ther such a re<
ment takes int<
the ii-then re
in the form of
If rt
aotiviti
events a
prise a:
tained.

Such a DTODC
iT-then rel;
information
Writer did
itended to :
intormatio
record of

2h v

rules of
W

“9 (39“th
iitductio

ll

If accounting develops and maintains a com—
plete record of all events and actions which result
in changes in enterprise assets and equities, then
reliable information concerning enterprise activ— ,
ities is provided.

hithis form, the course of action recommended is recognized
as a sufficient, but not a necessary condition for the goal
to be achieved.1 As an experiment, the writer attempted to
envision what might have been the deductive argumentation
leading to this principle from the postulates and definitions

in the monograph.2 To achieve the principle as restated

above, the derivation required several nonsubstantive

 

lAs restated, the principle asserts that if a course
of action (the developing and maintaining of a certain kind
of record) is followed, the goal (provision of some reliable
information about something) will be achieved; or, where
such a record is found, reliable information will be found
also. It does not deny that in some enterprises some reli~
able information may be available in the absence of the kind
of record described,nor that some enterprises may have nei~
ther such a record nor some reliable information. The restate~
ment takes into consideration the logical interpretation of
the if—then relationship. The principle should not be stated
in the form of a necessary condition, which would be written:

If reliable information concerning enterprise

activities is provided, a complete record of all

events and actions which result in changes in enter—

prise assets and equities is developed and main—

tained.
Such a proposition, under the normal interpretation of the
if—then relationship, denies the availability of reliable
information in the absence of a complete'record. But the
writer did not believe that the authors of the study in—
tended to suggest or would hold the position that reliable
information cannot exist in a situation where a complete
record of the matters listed is not kept.

 

2A very limited informal axiomatic system was used. The
rules of inference used were those proposed by Quine in
Methods of Logic, Revised. The rules of the system include
no controversial inferences and are common to most natural
deduction systems.

 

 

—

 

motions, additional <

active (and perhaps ‘1‘
relationship of validit
depended also on the W.
ahhors intended to sa
2e writer emphasizes
1169933951 in the v
ation, assumptions
tomograph. As such
tsmher of possibili
ieen considered by
with to determi'
Principles 1
ii a stated goay
831 and 8b are i
“a“ be restate(
meshing ex
expanded to a.
$th derivati
M additiona

definitions
\
11h a.
writer “0t
a‘Qc‘i‘iutini
10%10 of
“Within
the assu“
the tran
that WOU
vat/10“ J

A

- 190

asmmmtions, additional definitions, and at least one sub~
stantive (and perhaps unacceptable) assumption regarding the 1
relationship of validity or completeness to reliability, and i
depended also on the writer‘s interpretation of what the
authors intended to say in the postulates and principles.1
hm writer emphasizes that the derivation (or any other) 12
pgt_reported in the monograph, and that it requires infor—
mation, assumptions and rules which are not reported in the
monograph. As such, it represents only one of an infinite
number of possibilities, any one or none of which may have
been considered by the authors. From the monograph there is

no way to determine what has been considered by the authors.

 

Principles lb, 2a, 2b, 7 and 8, and, with the addition
of a stated goal to be achieved, principles la, 2, 3, 4, 7a,
8a, and 8b are of the same general form as principle 1, and
can be restated to eliminate their normative aspect. The
possibility exists that the experiment noted above might be
expanded to allow the derivations of these principles; but
such derivations appear to be possible only with a multitude
of additional substantive and nonsubstantive assumptions and

definitions and considerable personal interpretation of the

 

1In addition to the matters mentioned already, the
Writer noted that the principle as stated presents a minor
accounting but a major logical problem which involves the
logic of dealing with the null set. From a strictly logical
standpoint,the principle would have to be restated to include
the assumption that the complete record does have at least
one transaction in it. Obviously, this assumption is not one
that would annoy the accountant, but without it, the deri-
Vation is impossible from the point of view of the logician.

 

—

the used in the monog

E. and principles 7 am

free principles may t
aiding, then, the c‘
'.‘m use oi deductive
rinciples or by the
:‘zeiorm oi the pri‘
are type of observ

Comparison of Obse
imposed hethodol<

____/
i‘he inductiv
development of p
Liable argument:
:ation, the tes
'vlith respect t<
principles, ti
monograph off
The strv
in some othe
Cmtiter \I.
lating ade<
csEStions c

note that
W 0i a<

may not

A

 

Qwel‘er

 

fl

words used in the monograph. By combining principles 5 and

191

Q and principles 7 and 8c into a form like principle l,

these principles may be added to the list given above. Sum—
marizing, then, the claim to development of principles through
the use of deductive logic is not supported by the form of the
principles or by the discussions in the monograph. In fact,
the form of the principles as stated is more in keeping with
some type of observation and inductive argument.

ngparison of Observed and

Proposed Methodology

The inductive methods apparent in the monograph for the

 

development of postulates are limited to the generally unre-
liable arguments of self-evidence, or, by broad interpre-
tation, the testimony of experts (the authors themselves).
With respect to the methods used in developing concepts or
principles, there can be no comparison made, since the
monograph offers no indication of what methods these were.
The structure envisioned by the Study Group corresponds
in some other important respects with that proposed in
Chapter V. The importance placed by the authors on formu—
lating adequate concept definitions coincides with the sug~
gestions of this writer. Moreover, it is interesting to
note that while the Study Group claims to have a general the—
ory of accounting, its authors recognize that their theory
may not be adequate for all areas of accounting. They be—
lieve that some specialized areas of accounting can be

covered only by developing additional postulates and prin—

 

—

is suggested i

ciples.

mics oi varying scope
‘ \

mssihie. The writer

:i the Study Group in
limits to the use of
general theory can b

derable investigati<

Structure Termino‘.
Professor Ct

oi general ideas
maybe based“ (1
sent the source
Tractices.“ (p
assumptions, 2
istics of the
Which the en'
m“titles a
accounting,

in vi

W0 categc
aetountin

cm the
(heating

M Simp

the w
y

A

rvgggfiw 192 1'!!!J

ciples. As suggested in Chapters III and V, several the—

ories of varying scope and on various subjects are, indeed,

possible. The writer‘s proposal differs from the position a
of the Study Group in that the writer, recognizing the f
limits to the use of logical methods, holds that an adequate

general theory can be developed (if ever) only after consi—

derable investigation on a much smaller scale.
Towards a General Theory

Structure Terminology
Professor Chambers states that he is pursuing "a body

of general ideas on which the practice of accounting is or

 

may be based" (p. 3), ”a network of ideas which can repre—
sent the source of all specific prescriptions for accounting
practices." (p. 4) The network consists of postulates or
assumptions, and principles. Postulates are ”the character—
istics of the world of action," which "represent the way in
which the environment of accounting is visualized.” (p. 29)
Principles are ”statements relating to the general nature of
accounting.” (p. 44)

In View of the author's description of his goal, to the
two categories specified above must be added another for

accounting practices. Professor Chambers notes that '

'a gen~
eral theory or framework of concepts is sought” (p. 46), in~
dicating that for him, concepts are not a separate category
Of simple terms-but rather propositions, and specifically,

the propositions of the category he calls principles.

 

—

homology Claimed and

in a sense, frofe‘

claims to the use of l
:tly that he believes
salvation or introspé
id have used any log
have exhibited such
actually claim that
leit to the reader
examine his own ti
“for support of ti
attempt to iusti
hung with hi:
he‘d afeuassun
The princ
inundation or
bi" “defilled”
left impeci
writer can t
Sttidies dis
Research 8
“Thrash
the made

DrinCipb
\

l
of “int

 

 

I

193

Mewmdology Claimed and Observed
In a sense, Professor Chambers is very modest in his

claims to the use of logic. Of his postulates he states
only that he believes them ”realistic and verifiable by ob-
servation or introspection.” (p. 39) But, he makes no claim

to have used any logical method in their development or to

have exhibited such argument in his monograph; nor does he
It is

 

actually claim that his postulates are self—evident.

left to the reader, then, to make his own observations or to

examine his own thought processes and sensory experience,
for support of the postulates. That the author makes no
attempt to justify these assumptions in the monograph is in
keeping with his claim. He simply states them for what

they are——assumptions.
The principles are said to "have emerged from" the

 

foundation of postulates (p. 44), and are ”deduced" (p. 3)
or ”derived” therefrom (p. 43). The method of deduction is

left unspecified, however. With respect to this claim, the
writer can only repeat those comments made concerning the
studies discussed earlier in the chapter. Like Accounting
Research Study No. 3 and the Illinois Study, the Chambers
monograph provides no list of rules of inference which allow
the reader to move from postulates to principles. With each

principle, Professor Chambers offers references to one or

” lThis second alternative is based upon the definition
Of introspection" given in Webster's Seventh New Collegiate

£EEEEQQ§EX (S ringfield, Massachusetts: G. & C. Merriam
Comm my, 19655D .

IIIIIIIIIIIIIIIIIIIIIIIll--______

 

in post

(
ulates and one

sa '
aw references, he 3,

rounds on which the st

H 1
they are discussed. t

fight speculate on how

:tstwlates to the pri

"finances used in tl

themes) were actua
when support poin
3‘? the author‘s lit
sample of one pri
rate this situat

The rele
varies dire
ities ii.e.
6M at u

mks“? Shawn‘s

For a
tensity Q
eXwectati
0f: agtio‘

The anther a?
monograph in

F0

by g
able 6
abstr
Vafie
the (
SO Q
mens
more
Whe
0‘?

The 90s

‘.
1

A

19M E!!!1

nnre postulates and one or more paragraphs of discussion,

mnch references, he says, ”are rough indication of the .

grounds on which the statements depend and the place in which
they are discussed." (p. MA) The writer and other readers
might speculate on how the author might have moved from the
postulates to the principles, but cannot be certain that those \

inferences used in the speculation (or any other logical in—

 

ferences) were actually used by the author. Nor does the in—

 

tended support pointed out by the author give an indication

of the author’s line of argument or rules of inference. An

example of one principle and its support will serve to illus—

trate this situation. In support of principle D, which states:

The relevance of accounting information
varies directly with its correspondence with real—
ities i.e., market conditions, contractual relations,
etc.) at the time of actions. (p.

 

Professor Chambers offers only the following postulate:

For a given role choice depends on the in—
tensity of wants, on present knowledge and on
expectations of the outcomes of available courses

of actions. (p. #0)
The author also refers the reader to one paragraph in the

monograph for discussion of his principle.
Formal or processed information is conveyed

by signals which differ from the original observ—
able events and things. Signals are necessarily
abstractions. The probability of optimal adaptation
varies with the correspondence of the signals to
the observable real events and things; they may be
so conventionalized that they represent in one di-

mension what may only be described adequately in
more than one dimension. Correspondence is critical

Where the actor has no possibility of confirming
or testing the quality of the signals. (p. 17)

The postulate mentions neither the accounting information,

IIIIIIIIIIIIIIIIIIIIIIIll--______

 

~ - nor ti
the realities 5

find in the principle 3
zation paragraph neith
inner presents even i
1: me principle in 0,!
he example give
hearse it did not re
estuary, on the basi
fiancgraph, the writ
in logic to the (
haiessor Chambers
hoiessor Ch
2‘framework of Cr
various kinds ar
described.“ Kl)-

?” demo? pra

that falls am
Chambers is f
monograph, t'

W
team

As m

\

l

the stud
Pmessc
Winnip-
paper .3
a list:

 

A

195

  

nor the direct proportional relationship

nor the realities,
And the dis—

found in the principle it is said to imply.
cussion paragraph neither relates to the postulate, quotes

it, nor presents even the outline of an argument leading

to the principle in question.
The example given is typical and was selected primarily
because it did not require extensive quotations.1 In
summary, on the basis of the information contained in the
nwnograph, the writer cannot attribute any method of deduc—

tive logic to the development of the principles listed by

Professor Chambers.
Professor Chambers claims that the principles provide

a framework of concepts "by means of which accounting of
various kinds and of various entities may be explained or
described." (p. 46) No suggestions are given regarding how

to develop practices from these principles, or how to decide
what falls into the category of practices. Since Professor

Chambers is not concerned with accounting practices in the

monograph, there is no way of knowing what methodology he

Would use.

£2fl2§rison of Observed and
lineages Methodology
As noted in Chapters III and V, it is often informative

1A complete listing of postulates and principles from

the study has not been included in this paper. Altogether,
Professor Chambers proposes forty postulates and twenty—one
principles. The one principle discussed at length in this
paper is a typical example of the group. The inclusion of
a listing was deemed, by this writer, as not necessary.

—_

 

 

 

ppppose ’dSS

zeta con

umptions f0
sequences and W3
xas‘wptions W10? to thi
tribute this motive ta
enetlg claimed it. TY
assumptions to the rea
airetrospectioni su
sported by some met
they, as well as si
tin the suggestion
up for the develo‘
prisons oi observ

:a the developmen
illear to have be

1n all fair

“dashed that h

all human

tthepal Therm

 

that We stuc‘
diets that t
“vations,? (9
author (am
trading en
intended .
“it Write
Q“mete

“mic m

A

196 Q

to propose assumptions for the specific purpose of exploring
Hwir consequences and without attempting to justify the
assumptions prior to this exploration. The writer might ,
attribute this motive to Professor Chambers though he has not

exactly claimed it. That he leaves the evaluation of his

assumptions to the reader‘s personal experience (observation

and retrospection) suggests that he believes they can be

supported by some method of inductive argument. This method~

ology, as well as simple hypothesis invention, is consistent

with the suggestions in Chapters IV and V as to the methodol—

ogy for the development of postulates. There can be no com-

parisons of observed and proposed methodology with respect

 

to the development of principles inasmuch as there do not
appear to have been any methods of logic used.

In all fairness to Professor Chambers, it should be em~
phasized that he made only limited claims for the completeness
and applicability of his study. The title itself, Towards a
General Theory, indicates that the author did not intend
that the study be final; and, in the Introduction, he pre-
dicts that the study ”will have some defects and some limis
tationsf (p. 3) But the limitations pointed out by the
author (applicability to only natural persons or simple
trading enterprises) give the impression that the author
intended the theory to be sufficient for those two cases.
The writer suggests that as Professor Chambers seeks to
00mplete his theory the utilization of the informal axio~

matic method for the development of principles (as proposed

 

—

thapter ll) would serve
t the

its he predicted a

Account ing
hectare Terminology

rthe work of YTOEE
iiiierent interpretat
'asie assumptions of
ice basic assumptiox
which must be prese
he definition of
laws, when he sug‘
be is observing i
beiinition of at

ll some assumpt
some way. he (

at these assur

With.

his lis

and eight, w
3mmlltione

TEither it

Elven. 1

Etheral i

System C
\

l

l'

 

\

A

 

197 IIIIIII'III'IIII

hiChapter V) would serve to bring to light the possible de—

fects he predicted at the outset of the study.

Accounting and Analytical Methods

Structure Terminology

The work of Professor Mattessich indicates an entirely
different interpretation of what other authors call ”the
basic assumptions of accounting.” For Professor Mattessich,
the basic assumptions of accounting are those conditions
which must be present for any accounting system-~in effect,
the definition of an accounting system (p. 26 and p. 30).
Thus, when he suggests that there are eighteen assumptions,
he is observing that an accounting system, to be within his
definition of an accounting system, must have been preceded
by some assumptions dealing with these eighteen matters in
some way. He does not always suggest, however, in his list
Of these assumptions, how all these matters are to be dealt
with.

His listl consists of ten specifically stated assumptions,
and eight, what he calls "surrogate” or "place—holding”

assumptions in which no specific statement is given but

 

rather it is assumed that there are statements which could b3
given. In a sense, then, his assumptions are not so much a
general theory of accounting as a general outline for a

SVStem of accounting. In this way, his work differs sub—

 

l . .
The text of his assumptions is given in Appendix C.

 

 

azatially

rained earlier.

from, for exai

For e
at a basic assumption

rthere exists
the value assigne

"is position is that
lttbi rules (the tea
ieternine the value
by wouldn‘t be abl
‘rbwewer, that any i
in category of “a
tlioonitz sugge
with be that ar
it the risk of c
thiessor liatte
mm“lition or t
Wessor Moor

am“Pt'icns a
tgain (x
“mm what
they be inc
fined term
1“ AMEN?
assume-1C
rename
QeWary

“hiss

 

I

stantially from, for example, the AICPA postulate study,

198

 

examined earlier. For example, Professor Mattessich states
that a basic assumption (one of his surrogate assumptions) is:

There exists a set of hypotheses determining
the value assigned to an accounting transaction.

His position is that it must be assumed that there is some
set of rules (the terminology here is the writer's) which
determine the value to be used in a transaction, or account—
ing wouldn‘t be able to take place. He does not believe,
however, that any particular set of valuation hypotheses fits
the category of ”a basic assumption.” On the other hand,

Dr. Moonitz suggests one of the basic assumptions of account—

 

ing to be that accounting data are based on exchange prices.
At the risk of oversimplification, it might be said that
Professor Mattessich suggests the subject matter of the
assumption or what accountants make assumptions about, while
Professor Moonitz (and others) are suggesting specific
assumptions about those subjects.

Again unlike the other authors, Dr. Mattessich does not
SpeCify what form the basic assumptions must take~whether
they be incorporated in definitions or propositions or unde~
fined terms. His own attempt at formulating a system (given
in Appendix A of his book) shows that he does not equate his
assumptions with axioms; for, he uses all three of the above~
mentioned forms to meet the conditions he maintains are ne—-
cessary. For example, the matters of economic object and en~

tities are incorporated into his system in the form of an

 

—

 

 

.cceiined term and def ll
utter oi time, the aut
nitions. 0n the oth
are generally limiter
For Professor lie
"principles" (p. 30)
“ll. “Theorem" '
biased in any syst
assumptions. The
consistent meanin:
litre definitic
“he nature of m
Selection Of tin
gate assumptlm

tices may be I

)M

Technic

his “basic 2
Mounting
uses W mo
Wing an :
0“ these
States t
“beers
)9-3f

MSW

199 IIIII

cmdefined term and definitions, respectively, while for the
matter of time, the author utilizes both definitions and pro— ;
positions. On the other hand, the other studies examined
have generally limited themselves to propositions only.

For Professor Mattessich, the ”assumptions” are also
”principles" (p. 30), and the term ”postulate” is not used
at all. "Theorem” is used to refer to anything which can be
deduced in any system which incorporates the eighteen basic
assumptions. The term "concept” is given no specific and
consistent meaning through the study. The author does not
offer a definition of accounting practices, but in View of

the nature of his assumptions, one might consider the

 

selection of the empirical hypotheses to replace his surro—
gate assumptions as accounting practices, or accounting prac~

tices may be the actual working with a given accounting system.

Methodology Claimed and Observed
Technically, Professor Mattessich does not claim that

' are the basic assumptions of

his "basic assumptions'
accounting at this time in the sense in which this phrase is
used by most authors. For, he holds that he will call some—
thing an accounting system only in cases where assumptions

on these eighteen matters appear to be operative. Thus, he
states that the eighteen assumptions are the ”necessary and
(hopefully) sufficient conditions for an accounting model”

(p. 31), i.e., for an accounting system to exist. If one is

missing, an accounting system simply does not, by definition,

 

—

 

1 L

Mu .

lless technical
"action would suggest
'ately called thg b
“re author makes no 5

rearriwe at those t

ally. Nor does he
ii the conclusion
gaming the eight
teat the assumptic
actions (p. lll),
fall into which
his assumptions
licapter Li) is
1‘The Evolution
cents consider
"Mounting 3y
that “from a
- . all of
“3 the acco
abSence of
mm attr
the develw
E‘Iidehge
“9t cons

a“ther

A

I 200

exist.

A_less technical interpretation of the author's in—
tention would suggest that he believes his list to be appro—
priately called th§_basic assumptions at the present time.
The author makes no specific claim as to the methods he used
to arrive at those ten assumptions which he states specifi-
cally. Nor does he claim some particular method in arriving
at the conclusion that there must be some assumptions re-
garding the eight other matters on his list. He states only
that the assumptions deal with both g_priori and empirical
notions (p. Al), and he does not specify which assumptions

fall into which category. He does not, however, consider

 

his assumptions unsupported. An entire chapter of the book

(Chapter A) is devoted to what Professor Mattessich calls

 

”The Evolution of the Accounting Model,” in which he pre—
sents considerable discussion and examples of various
accounting systems of the past and present. He observes
that ”from a broad point of view or by careful reflection
all of these basic assumptions are prevalent in each
Of the accounting systems discussed.” (p. 138) Despite the
absence of any claim by the author, the writer would will—
ingly attribute the method of inductive generalization to
the development of the author's list of assumptions. The
evidence provided——examples, and discussions of them——did
not consist of written historical statements of the exact
assumptions in each accounting system discussed; but the

author pointed out those attributes of the systems which

 

—

 

the led him to believe
tie. Whether the indi‘
fierce sufficient to war
yestion arising with
ear the individual‘ s
which is intended to
its author does
he list of assumptf
referred to the wor
“:lidence of the co
it. connection wity
Earlier as appear
author has attem-
incorporates the
is? a S‘Jstem.
We is a lim
(Glace of the 5
Suggested) _
not the Only
“9'59“th it
For hi
a timer“ .
any Yeddey

SWEQpL
\

lSe

9' 3.

201

have led him to believe that such assumptions were opera—
tive. Whether the individual reader will consider the evi—
dence sufficient to warrant the conclusions drawn is a
mmstion arising with all inductive inferences; but what—
ever the individual’s answer may be, evidence is presented
which is intended to support the listed assumptions.

The author does not claim to have deduced anything from
the list of assumptions as such. When other authors have
referred to the work of Professor Mattessich as providing
evidence of the contribution of logic,1 it has been primarily
in connection with the specific axiomatic formulation noted

earlier as appearing in Appendix A of his book. There, the

 

author has attempted to formulate a general system which
incorporates the basic assumptions presented in his outline
for a system. It is a ”general" system only insofar as
there is a limited attempt to supply specific hypotheses in
place of the surrogate assumptions in his list (only one is
saggested). It is an attempt insofar as the formulation is
not the only possible way of incorporating the conditions
presented in his outline.

For his general system (the author, however, calls it
a general theory), the author claims, and the writer and
any reader can observe and identify, the use of an axiomatic

System. The axioms and primitive terms are explicitly

 

1See, for example, the comments of Professor Storey
Quoted in this chapter, and Moonitz, The Basic Postulates,
p' 30

 

—

 

 

cated and the deriv at

a‘clcuahle rules of in
statement of the autt

is his rules, examin

hmuurison of Obser
dressed Metho dolog

rl‘he author ‘ S
his assumptions it
n; the writer in
ratio method to
cific assumptioc
Forever, it shc
ihttessioh has
the sense '
rather the be
hofessor Ma
theory of a!
knowledge x

View of

 

1lis

matte'Ssic
am‘umet
W391, i
to the (
amount
Q0“Sis‘c
drama
minor
with:

A

 

202 Illlllllll

stated and the derivations of theorems are given. The
allowable rules of inference are incorporated by explicit ;
statement of the author so that any reader can, by referring
to his rules, examine the derivation of any theorem.1
Comparison of Observed and
Proposed Methodology

The author's use of inductive methods to arrive at
his assumptions is in keeping with the methodology proposed
by the writer in Chapter V. Likewise, the use of the axio—
matic method to deduce further propositions from some spe—
cific assumptions coincides with the methodology proposed.

However, it should be remembered that what Professor

 

Mattessich has constructed in his Appendix is not a theory
in the sense in which that term is used by this writer, but
rather the beginnings of a specific accounting system. And,
Professor Mattessich's outline of assumptions is not a
theory of accounting or an attempt to structure accounting
knowledge but, as he says, a meta—theory (p. 426)—not a

theory of general accounting, but a theory about accounting ,

 

 

1A8 a matter of minor interest, insofar as Professor
Mattessich's system incorporates the matters of elementary
arithmetic, in accordance with the well—known work of
GOdel, it cannot be both consistent and complete with respect c
to the class of true sentences which can be written in the
language of his system. This fact will perhaps bother
accountants who are striving for the goal of a complete and
consistent theory of accounting; but if this were the only
drawback to Professor Mattessich’s system, it would be a
minor problem since all other systems which utilize basic
arithmetic contain the same limitation.

 

 

 

mgeneral. in fact,
some set of hypothes
gmeassmmtions, a
cashle,for any E
he related to some
ymrd.theory of
d he day W111 b
theory; and the .
the Ethereal thec
it is inte
Studies have le
equal level.
he nature of
“rated body
matically or
$9530? Matt.
that each 2
eighteen m
DOSitiOn ,
3' Postuie
the idem
transam
or the
that 0
iiCati

actic

A

203

c i-

in general. In fact, given that his outline demands that ‘
some set of hypotheses fill the conditions of his surro-

gate assumptions, a general theory of accounting seems im—
possible, for any set of hypotheses which is proposed will

be related to some Specific accounting objective. Hence, a
general theory of the type being called for in the literature
of the day will be, for Professor Mattessich, a specific
theory; and the Specific system he has constructed is not

the general theory for which accountants are asking.

It is interesting to note that Professor Mattessich’s
studies have led him to place all of his assumptions on an
equal level. As noted in Chapter IV, there is nothing in
the nature of inductive argument which will produce a coor—
dinated body of general acc0unting propositions or auto-
matically create a hierarchy among these statements. Pro—
fessor Mattessich's observations led him to the conclusion
that each accounting system embodies some assumptions about
Eighteen matters, and he puts all on an equal basis-—a
position radically different from that of others who accept
a postulate—principle scheme. Thus, he does not envision
the identification, valuation and classification of economic
transactions as following from the nature of the transaction
or the entity or any other general matter; rather he proposes
that one must postulate identification, valuation and classi-
fication criteria at the same time as one postulates trans-

actions and entities. Again, though his assumptions may be

 

—

 

mmatiue in the sense
they are the “should

can vary from system

his other assumptior

from the positions

Structure rhennino'
hethodology Claim
The termino‘.
in many respects
though he propo
terms of accour
most often to
though he doe:
some basic te
all relation
id- 17 and 1
i0 describe

it 9i, he
Circles“ {E
“0i confi
arrears .
“County

ciples
Pt

hot (it

204 Ill'lll

normative in the sense of postulating what ”should be,”
they are the "should be" for a specific accounting system,
can vary from system to system, and are not dictated by

his other assumptions. This, too, is a very real departure

from the positions expressed by most other writers. .
Accounting Theory as a Logical System

Structure Terminology and
Methodology Claimed’

 

 

The terminology used by Professor Carlson corresponds,
in many respects, with that used by the writer in Chapter V

though he proposes no specific meanings for the often-used

 

terms of accounting. ”Concept” is the term he applies

 

most often to the primitive terms of an axiomatic system,
though he does sometimes use the word in connection with
some basic term outside any system (p. 49). Postulates are
any relationship asserted to hold in an axiomatic system
(p. 17 and p. 81). While the author uses the word ”theorem”
to describe the propositions deduced in an axiomatic system
(p- 9), he apparently would allow the use of the title ”prin-
Ciples” (p. 17) for the same propositions. However, he does
not confine the title ”principle" to such theorems, and it
appears that he would class those general statements in
Accounting Research Study No. 7 and any like them as prin-
Ciples also.

Professor Carlson’s meaning of the term "theory" is

not quite clear. The term is often used to describe an

—

 

 

micmatic system is .g .

"a single theory may i
rich formalize it er
2o indicate that a d
ulation or formali:
is something more c
are not defined at
that there was a <
deductive systems
filling their id
in his stue
system which he
called TS‘, an
he first is _:
the second in
changes. Th

15 based mpg
Moonitz-spy
iDTincipye.
Proie

Mtuiace

W. m
metymds
tulates

it at

l- 2i:

A

205 ‘IIII

axiomatic system (e.g., pp. 17—18). But he also notes that
’h single theory may have any number of axiomatic systems
which formalize it entirely or in part” (p. 8), which seems
to indicate that a deductive axiomatic system is only a for—
mulation or formalization of a theory (p. 4), and the theory
is something more or less intangible. Accounting practices
are not defined at all, though the author must have believed
that there was a category of such things, since theorems of
deductive systems can hardly be acts of accountants ful—
filling their functions as accountants.

In his study, Professor Carlson constructs an axiomatic

system which he calls TS, modifies this system into another,

 

Called TS', and outlines still another which he calls MS.

The first is intended to deal with ”traditional accounting,”

 

the second incorporates the possibility of purchasing-power
changes. The last deals with current—value accounting, and
is based upon Professor Carlson's interpretation of the
Moonitz—Sprouse (hence, MS) principles study. No theorems
(principles) are presented for the last two systems.
Professor Carlson makes no suggestions regarding how
postulates and definitions of concepts are to be devel-
Oped. Nor does he make any claims to the use of logical
methods in the development of his own definitions or pos—
tulates, though one postulate is said to be ”obvious”
(p- 94), and two others "intuitively obvious." (p. 158 and

p. 212) The theorems are, by definition, claimed as de—

 

—

 

heed from the postul:

i.ethodology Ob serv e d

Though as state

i logic in arriv in
has argue that his
eat of traditiona
argument does not
of his system are
tic assumptions
on showing that
the same as som
did, logically
herd not he ti
irrsons handi
Carlson himsc
the same gey
Of the
KQ‘E, and e
QOhCep’gS E
06ml eno
‘ura‘yg W
“-1, (3.

Q7113 a

i
Emery

 

nee .
d not be the assumptions made or eve

‘Q-ll) are offered as obvious or with

206
duced from the postulates.

Methodology Observed

T
hough as stated above, no claims are made to the use

Oflo‘ - ..
$10 in arriVing at the postulates, Professor Carlson

does
argue that his system TS does formalize a significant

ar - -
p t 0f traditional accounting (pp. 216—220). However, the

ar u
S ment does not support the contention that the postulates

ccounting or

___.——-—

of '
hlS System are therefore the assumptions of a

the -
___ assumptions of accountants. For, the argument rests

on '
ShOWing that the theorems of his system are essentially

the sa
me as some AICPA pronouncements on several matters.

An
d: logically, the postulates or assumptions of his system

n considered by those

persons handing down the AICPA pronouncements. As Professor

Calf’1son himself noted, there may be many systems to express

the same general position.

0f the twenty—three postulates in system TS,l six

) deal with simple mathematical

(Q-E, and Q—6 through Q-lO
concepts and may be considered as needing no support. (Yet,
Oddly enough, Professor Carlson chooses to present alge—

postulates of TS
Q—A, Q—5 and

braic proofs for some.) Twelve of the

(0-1, 0—2, 0—3, 0—5, 0—6, M—2, M—4, Q—l. Q—3,

out supporting argument,

1 -
A list of the postulates used by Professor Carlson ls

given in Appendix D.

 

 

rm appears in heepi
interest in the theore
system. Four of the
accompanied by refer
the by other accour
9.1m. Consistent
writer has conside‘
pm the testimony
matting, postulate
het'ive argument

5? systems TS‘ e
715, are Profess
“fade in the Mm
“tic was 9133

The meth

Meloment c

Dfistulates 1

10?;in Efi‘

EQEEEher wf

adopted.

Stated in

“Oh afffl
\

1 TY

“Newly
pfi the
eTTOTS
SYStEm
Son \ S
they p

207 IIIII

which appears in keeping with the author‘s expressed
interest in the theorems rather than the postulates of the 1
system. Four of the postulates (M—1, M—3; M~5 and M—6) are '
accompanied by references to somewhat similar statements
made by other accounting writers of pronouncements of the
AICPA. Consistent with the treatment of other studies, the
writer has considered these as inductive arguments based
upon the testimony of experts. The argument for the re—
maining postulate (C—A) can be identified as a simple de—
ductive argument of the traditional logic. The postulates
of systems TS' and MS, insofar as they differ from those of

T3, are Professor Carlson’s interpretation of statements

 

made in the Moonitz—Sprouse monograph, and no method of
logic was claimed or observed for their formulation.

The method of logic claimed and observed for the actual
development of the principles presented as deduced from the
postulates in System TS was the axiomatic method of deductive
logic. Primitive terms and axioms were explicitly stated,
together with a complete list of the rules of inference
adopted. Proofs (arguments in support) of principles were
stated in full. That there were errors in some proofsl does

not affect the ability to identify the method of logic used.

 

1These errors have been discussed with the author and
acknowledged. At the present time, the writer is not aware
Of the methods of their subsequent correction, if any. The
errors affected the proofs of almost all of the theorems of
System TS; and for this reason, a list of Professor Carl—
son's principles has not been included in this paper since
they have not been proved.

 

—

Comparison of Observe
imposed Methodology
Shwe Professor
it. supporting his pc
rhrhfles, the fac
:htpestulate deve
theses the testi
iwmisihg nor a (
ht. 2t8-25Qy the
@fistulates fmm
ht,inductive (
EU-tg‘ssted by th
0? POSLbulates,
the pTifiCiplg
me by the W
1“ some
%SSOT Carl:
YWessor c

SQECif-lc E

 

New“

 

idea t

M: DP;

AL

f' 208 .4

Comparison of Observed and
. ,Proposed Methodology

Since Professor Carlson was not primarily interested
in supporting his postulates but rather in deriving his
principles, the fact that the methodology attributed to
most postulate development was either inductive (argument
based on the testimony of others) or nonexistent is neither
surprising nor a criticism. The author himself suggests
(pp. 248—250) that it was his intention to find or invent
postulates from which he could deduce the desired theorems.
Yet, inductive methods and hypothesis invention were both

suggested by the writer as legitimate for the development

 

of postulates. The use of the axiomatic method to investi-
gate principles is, of course, in keeping with the suggestion
made by the writer in Chapter V.

In some important ways, however, the proposals of Pro—
fessor Carlson differ from those made by this writer.
Professor Carlson does not envision the use of axiomatic
method in the presentation of a simple argument for any
Specific position, but rather for problems of such scope as
all of traditional financial accounting theory, cost account~
ing, etc. Making no claims to the completeness of his
theory (in a logical or a nonlogical sense), he nevertheless
maintains that it contains a ”substantial part of currently
accepted accounting theory” (p. 248); thus he holds to the
idea that a general theory of accounting should be the goal

Of present research efforts employing the axiomatic method.

—

 

 

its system he has co

perhaps his claims a
system does not all
that accounts recei
memem assets, wh
necessity of inter
iomer of cost or
the system allows
cambe used to p
it does not prom
M some other 1
stantive (1.3
Only Eight are
the system es
the derimat'1c
point out th
““10“ assumps
his rostuia
M the def:
This fact
is perhap
stage p01

209

The system he has constructed provides some evidence that
perhaps his claims and such goals are premature. For, the
system does not allow the proof of such simple matters as
that accounts receivable and inventory are current assets,
or even assets, while it attempts to provide a proof for the
necessity of inter—period income tax allocation and the
lower of cost or market valuation of inventory. And while
the system allows a proof for the possibility that assets
can be used to provide other assets (the costs attach idea),
it does not provide for the replacement of liabilities

by some other liabilities. Moreover, while seventeen sub—
stantive (i.e., nonmathematical) postulates have been offered,
only eight are operative in his system as presented. Nor was
the system established as consistent. The major emphasis in
the derivations is upon definitions, but the author does not
point out that his definitions for nonprimitive terms are as
much assumptions, subject to acceptance or rejection, as are
his postulates. And it is the formulation, in logical terms,
Of the definitions, that led to the errors mentioned earlier.
This fact is further evidence that the scope of the project
is perhaps beyond reasonable expectations at this early

Stage (or perhaps even any stage) in the formalization of

accounting ideas.

General Conclusions

The terminology used in the five studies examined does

 

 

 

not completely coinc
and in each study 5
sent accounting voc
Ber example, in th
mocks of Professor
term"conoept" an
Kpostulate, prim
portant, in none
definition of av
that constitute
related to the
ressect, the p
'1“ Chapter \I
offer an ad“
in gene
Possible to
assumptions
Statepxents,
mine. Inc.
as 3991109
But, the
Statist'l
mm ll
MattesE
argume

“gums

A

210

not completely coincide with that proposed in Chapter V,
and in each study some very common structural term of pre—
sent accounting vocabularies was unexplained or ignored.
For example, in the Accounting Research Studies, and the
works of Professors Carlson, Chambers and Mattessich, the
term ”concept" and how it relates to the other terms used
(postulate, principle, practice) are not explained. More im—
portant, in none of the studies examined is there a workable
definition of an accounting theory or any suggestion as to
what constitutes an accounting practice, or how such are
related to the other terms which are discussed. In this
respect, the proposed interrelated structure of terminology
in Chapter V differs from all the studies and appears to
offer an advantage.

In general, the methodology used (when it was at all
possible to identify it) in the development of postulates or
assumptions was inductive, though the range of accounting
statements covered by the title of ”postulate” was amazingly
Wide. Inductive argumentation was suggested in Chapter V
as appropriate for developing assumptions for a theory.

But, the specific methods recommended (viz., inductive and
Statistical generalizations and analogy) and the methods
found in the studies (with the exception of Professor
Mattessich‘s work) were basically not the same. The major
arguments found were the much less acceptable methods of

arguments claiming self—evidence or arguments based upon the

 

 

 

testimony of others

methods rested heav
peetation of what i
or on the adjustme
mood with that us
With the exc

and Mattessich,
ments from the 2
ance with the r
methods could i
rest of the st
“Set, the con
With the mic
able differe
their termir

Mattessich

fie accoun.

t0 aXiOmat
ing.“
Thre
1atively
and Wei
Wench;
my me
the Si

haVe

— A

 

I____________—::::JIIIIIIIIIII!IIII

testimony of others. And, the identification of even those

211

methods rested heavily either on the writer's broad inter—

pretation of what constituted the presentation of evidence,
or on the adjustment of the proposed terminology to corres—
pond with that used in the study under examination.

With the exception of the works of Professors Carlson
and Mattessich, the methods used to develop further state~
ments from the assumptions or postulates were not in accord—
ance with the recommendation in Chapter V, since no logical
methods could be attributed to their development in the
rest of the studies. In both exceptional cases, the method

used, the construction of an axiomatic system, corresponds

 

with the writer's recommendation. Yet, there is consider—

able difference in the intentions of the two authors,

their terminology and the results of their work. Professor

Mattessich has constructed, by his own definition, a speci—

fic accounting system, while Professor Carlson attempted

to axiomatize the whole of ”traditional financial account—

ing."
Throughout this chapter, the writer has adopted a re—

latively broad interpretation of the notions of evidence

and even argument. Nevertheless, the writer found it ex—

tremely difficult to find evidence of and to identify

any method of logic as having been used in many areas of

the studies examined. That gil_the works examined here

have been considered by their authors and/or others as log—

 

—

 

I

ical studies provide
iieation of the natt
desirable addition
the studies examine
rigorous area of l
interesting to noi
available to Prof
quoted at the he
studies can be (
accounting ough
Work done by r}
to these same
tehtion that -
tonerai them

iZat ion in a

212 1

ical studies provides grounds for suggesting that clari—
fication of the nature of logic and its methods would be a
desirable addition to current accounting literature. Of
the studies examined, only two exhibited the use of the more
rigorous area of logic——some method of deduction. It is
interesting to note that neither of those studies was
available to Professor Storey at the time of his comment
quoted at the beginning of the chapter. That some logical
studies can be carried on relating to some aspects of
accounting ought not to be disputed in view of the type of
Work done by Professors Carlson and Mattessich. Reference

to these same two works does not, however, support a con—

 

tention that logical studies have, or even can, produce a

general theory at this stage in the development of formal—

ization in accounting theory.

 

   

 

 

This study
intense interest
very foundation
is something '1‘

mal practice 8
specific situ:
whatever the

said to be p‘

accounting i

being adopt
similar sii
results on
That
practices
vincing

a methoc'
found w
search

for er

I

of ai

A

CHAPTER VII
I /
RESUME

This study was undertaken because of the present
intense interest in investigation and research into the
very foundations of the accounting process. That there
is something in accounting which transcends the individ—
ual practice adopted by a particular enterprise in a
specific situation is seldom, if ever, disputed. But
whatever the name or names given to that elusive something
said to be prior to, and to provide the foundation for,
accounting practices, different practices are continually
being adopted—«practices which, when followed in apparently
similar situations, produce substantially different
results on financial statements.

That what purport to be justifications for these
practices appearing in professional journals are not con—
vincing to all accountants raises the question of whether
a method of justification or position presentation can be
fOund which would be accepted by all as convincing. The
Search for such a method most often produces suggestions
for an application of the methods of logic.

The use of ”logic” is suggested for investigations
Of all kinds of things in aCCOunting——principles, postu-

213

 

 

 

lates, concepts , et

mpcfl.techniques,
becmntrncted whi.
nation of these c2
ahierarchy of ac
pnamnbly show \
may current dis
bsisfor resol‘
current work. of
endence that '
mama general
haSbeen to c
fictions and

been meth

logical men

1‘5 sh
“loge“ b!
at least
lOg‘lc an

To Clarf
an Ham

“9ma

thEse

“\Ethg

made

‘-‘IIIII‘|.__»

214

lates, concepts, etc.——predicting that, with the help of
logical techniques, a general theory of accounting will

be constructed which would present an interrelated combi—
nation of these categories in its structure so as to produce
a hierarchy of accounting statements, Such a theory would
presumably show what should be done and bring an end to the
many current disputes, while at the same time providing a
basis for resolving future problems as they arise. The
current work of several researchers has been heralded as
evidence that logical studies have been made and can produce
such a general theory. The overall purpose of this study
has been to critically examine these suggestions, pre—
dictions and presumptions regarding the utilization of
logical methods in the light of the nature of logic and

logical methods and the state of accounting argumentation.

”Logic” Clarified

 

It should be recognized that a simple suggestion that
HlOgic” be applied to ”accounting” is, if not meaningless,
at least far from precise. There are several methods of
logic and there are many aspects to consider in accounting.
To clarify the first of the vague terms in the suggestion,
an examination of the nature of logic and its several
methods was undertaken in Chapters II, III and IV. In
these chapters, the attributes or characteristics of logical
methods were presented. Though limited comments were

made relating to possible utilization in accounting, for

 

 

 

:‘nenost part the d

limitations or pro‘
be recognized apar
Methods of a
for logical analy
those which invo‘
deductive) and t
tinction was sh
tation of certe
the word “logi
logic itself,
“ti-“E cert ai‘
Shims iwhic
“led through
‘mlilstified‘

0f logical
cOmlnsiong

lations.
01‘ u
90Ssib1y
hlze the
manner:
maths“
0P0Dosfi
vichm

215

the most part the discussion was organized so that the
limitations or problematic aspects of each method could
be recognized apart from any specific application,

Methods of argumentation (which is the subject matter
for logical analysis) were effectively segregated into
those which involved conclusive reasoning (properly labeled
deductive) and those which did not (inductive). The dis-
tinction was shown to be extremely important, for the conno—
tation of certainty, which, for some, may be associated with

the word ”logic,”

was shown to arise not from the nature of
logic itself, but rather from some of its methods. Attri—
buting certainty or irrefutability to scientific propo—
sitions (which are commonly understood to have been devel—
oped through the use of inductive methods) was shown to be
unjustified. What certainty is to be found through the use
of logical methods is attached to deductively derived
conclusions, and even this certainty is subject to reser-
vations.

Of the deductive(conclusive) methods, only two could
possibly be of interest to the accountant wishing to orga—
nize the body of accounting knowledge in some systematic
manner: truth functions of propositions and informal axio-
matics. The former, dealing as it does, only with whole
propositions, can be of no benefit in the study of indi~
Vidual terms, and its usefulness is limited primarily to

comparisons of relatively simple and clearly stated argu-

 

 

ments. rBhe latter

of varying scope am

of rigor, and has

attempts at formai
we has often gen
omate definition:
encouraged incre
assumptions and
Propernse allc
be placed on s
In Chapte
methods of re
Wily limite
"1318 to do s

for them, to

“We accura

for believi

those grow
Despite t?

the first

SnWort
Telat'h“

any re

as him

 

“timed .

216

ments. The latter method has proved useful for problems
of varying scope and complexity, allows of varying degrees
of rigor, and has been found appropriate for initial
attempts at formalization in a particular subject. Its
use has often generated much—needed interest in more ade—
quate definitions of basic terms of a discipline, and has
encouraged increased precision in the formulation of basic
assumptions and complete arguments. Most importantly, its
proper use allows specific and understandable meaning to
be placed on such claims as ”this logically follows,”

In Chapter IV it was shown that any discussion of the
methods of reasoning inconclusively (induction) is neces—
sarily limited by the fact that there are perhaps as many
Ways to do so as there are people to reason and matters
for them to reason about. Such ways of reasoning may be
more accurately described or classified as general grounds
for believing or holding a given proposition to be true,
those grounds being self—evidence,testimony and experience.
Despite the suggestions of many, accountants should dismiss
the first of these grounds, self-evidence, as a means to
Support their beliefs. Self—evidence was shown to be a
relative notion. It is a method of justification continu—
ally rejected in scientific investigations, and at least
as inconclusive a ground for belief as any of those exa—
mined°

The second ground, belief supported by reference to

 

 

 

the testimony of
of scientific im
accounting Wad"
by accountants .
is related to C
documented, am
that of relied
sciences, argi
tant part in
out that no 1
determinatio
expert. The
Those
were shown
ential grc
and stat'r
Proved we
time. T?
use of g
tation.
of evit
0? an
Clan (
uatir
bad,

Ev al

217

the testimony of others, was shown to be a valuable part
of scientific investigations, though the present state of
accounting practices indicate it to be generally rejected
by accountants. If and when expert accounting testimony
is related to conclusions based upon specific, well—
documented, and possibly quantitative experiments, as is
that of relied-upon experts in the physical and social
sciences, arguments in this form may also play an impor—
tant part in accounting investigations, But it was pointed
out that no rules or methods of logic can be utilized in a
determination of the acceptability of the testimony of an
expert. There is no logical necessity for its acceptance,
Those ways of reasoning popularly called inductive
were shown to be essentially the variations of the experi—
ential ground for belief. The methods available, inductive
and statistical generalization, and analogy, have all
proved valuable to researchers in other disciplines at one
time. The characteristic common to these methods is the
use of statements of evidence as the premisses of argumen—
tation. Evidence is the key to induction, and evaluation
Of evidence is what allows the determination of the strength
of an argument. But here it was emphasized that the logi—
Cian does not offer inviolable rules for gathering or eval-
uating evidence, for separating the good evidence from the
bad, or for drawing conclusions on the basis of evidence,

Evaluation of the evidence is not, in fact, the logician's

 

 

 

interest at all, “r

particular discip’.
the absence of ru
aresearcher frov
evaluation by ot
the evidence is
the evidence so

tased upon indw

accounting lit

The sim]
We? clarifi
“accounting
usage of th

“Practice ,'
have PTObe

‘IOcabular.

sistem-16

SEVEral

of these

identif
Substa}
wiringU
H310001
aimE

ab 0“

218

interest at all, but rather that of the researcher in the
particular discipline to which the evidence relates. But
the absence of rules for evaluating evidence does not excuse
a researcher from its presentation, for there can be no
evaluation by other than the original researcher unless

the evidence is available for review. Failure to report
the evidence supporting what appear to be conclusions
based upon induction was shown to be not uncommon in

accounting literature.

”Accounting” Clarified, A Proposal

 

The simple suggestion made above is in need of fur~
ther clarification for the second of its vague terms,
”accounting.” What is there to be investigated? Continued
usage of the terms ”concept," ”postulate,” ”principle,”

' and ”convention,” indicate that all

"practice,” ”theory,'
have probably become a permanent part of the accountant‘s
vocabulary. In Chapter V it was shown that serious incon-
sistencies exist in the use of these terms, and that the
several attempts made to explain the meanings of some

Of these terms have failed to consider the others or to
identify relationships between them which would give some
substance to the suggestions for the use of logic. Yet,
Without some clear understanding of at least what an
Haccounting theory” is supposed to be, it is quite impos—
sible to intelligently evaluate or make any predictions

about what the discovery of a ”general theory of account—

 

 

 

 

mg“ might OT might

The selection
ticular investigat
the thing being i
inconsistencies ,
defined interrel
process. Conse<
Chapter V which
listed terms i'
cabillty gene}
specific logi
these distinc
Which appear
matters fal?
basic unit
theory, wh
Presentat-j

lates and

basic he

were dej

Princip

nition

ventic

to ma

 

St at

219

ing” might or might not do for accountants°

The selection of a logical tool to apply in a par~
ticular investigation should be dictated by the nature of
the thing being investigated; and these terminological

inconsistencies, together with the absence of adequately v

 

defined interrelationships, seriously hamper the selection
process. Consequently, a terminology proposal was made in
Chapter V which allowed for a distinction between the above—
listed terms in such a way that, though all retained appli—
cability generally coincident with present usage,
Specific logical interrelationships were established. Given
these distinctions, those methods of logic were suggested
which appeared appropriate for investigations related to
matters falling in each of the categories defined. The
[basic unit of the proposed interrelated structure was a
theory, which was defined as any axiomatic system for the
presentation of a position or argument. Concepts, pogtu_
lates and principles, then, were equated with the system's
basic terms, axioms and theorems, respectively. Practices
were defined simply as actions in conformity with the
principle of some theory by adopting some set of defi—
nitions for the terms in the principle; the word ”con~
vention” was suggested to apply to some definition common
to many theories.

Under this proposal, the methods of inductive and

statistical generalization and analogy all would appear

 

——___

 

 

appropriate in the

ieiinitions of conc
tered as appropria
has a premium is

and inspiration ,
fronted with mati
factor. frincip
63 the deductiv‘
theory (axiomat
though clearly

to the other 5

tenerally not

ktic .

A “ger
distinct-10.
could W e:
hopes and

emii‘ely
devour,E
time to

to (inc:

w

the

220

appropriate in the investigations of postulates and the
definitions of concepts; hypothesis invention was consi~
dered as appropriate also for suggesting postulates.

Thus a premium is still placed on creativity, ingenuity
and inspiration, and the accountant continues to be con~
fronted with matters requiring exercise of the judgment
factor. Principles would be the result of the application
of the deductive rules of inference designated in a given
theory (axiomatic system). And, practices and conventions,
though clearly and understandably related, by definition,
to the other structural categories, would, nevertheless,
generally not be subject to investigation by methods of

logic.

A General Theory

A ”general theory,” using the above definitions and
distinctions, has some meaning. Whether such a theory
could ever be constructed and whether it would satisfy the
hopes and claims of those who urge its construction are
entirely different matters. And the advisability of
devoting the bulk of our research efforts at the present
time to the construction of such a theory is surely open

tO question.

An Ambitious Objective
In View of the profession's desire for progress in

the study of accounting problems, the directing of our

 

 

 

 

etiorts to the pure
almost a contradic
vigorously analyze
more progress the
large or small.
that all the cur
can he attributc
Though the
ticians of the
basis might be
study within
theory in any
alone in env
in Other dis
mailing-10“
lying, uni-
knowledge
seems are
WHY, at
depeyopm
e“gab'll’xst
the ott
a Seme-
the e
that

Sth

A

221

efforts to the pursuit of the elusive general theory seems
a . .

lmost a contradiction. A problem of secondary importance
rigorously analyzed and brought to solution represents

more progress than the failure to resolve any problems,

large or small. Furthermore, no proof has been offered

that all the currently controversial issues in accounting

can be attributed to the absence of a general theory.

Though there have always been hopes among the theore—

ticians of the more advanced physical sciences that some

baSiS might be found for unifying the seVeral branches of

StUdy Within a giVen science, there is, as yet, no general
theofy in any of them. Thus, While accountants may not be

alone in envisioning and pursuing such an ideal,researchers
in Other disciplines have not limited their efforts in for—
malization of their theories to the search for this under—

lying, unifying general theory. Without the formalized
knOWledge in the specific branches of each discipline, it

gest that there would be little to

seems appropriate to sue
at a stage of

unify, and the theories in accounting are

zation which strongly discriminates

development and formali
against the success of a search for a general theory. On
the Other hand, the development of formalized theories on
a SOmewhat less grand scale may, in fact, contribute to
the eventual construction Of a general theory by revealing
'that a particular series of axioms appears in each case.

e evidence that those assumptions

Such a result would b

 

 

 

core the postulates

analysis was
studies pointed t(
the help of logic
examination of t'
that the use of
a general theor
malization in :
Generally
the selected '
hers and the
with the bro
Dresentatio-
limited use
testimony
tation wav

ing (1'1ch

ever evi

rThe Matt

theory

acCOun

Sistey
them
seem

pl. 8,:

Spa

A

222

were the postulates of a general theory.
Analysis was undertaken, in Chapter VI, of the several

StUdieS pointed to as evidence that a general theory with

the help 0f lOgic is a reality or near reality. Critical

examination of these studies fails to support a contention

that the use of logical methods has, or even can, produce

a general theory at this point in the development of for—

malization in accounting theory.

Generally, no use of logical methods was discerned in

the selected works of Professors Moonitz, Sprouse and Cham—

bers and the staff of the University of Illinois, though

With the broadest interpretation of what constitutes the
to identify some

presentation of evidence it was possible

limited use of arguments claiming self—evidence or the

testimony of experts. The fact that such broad interpre—

tation was necessary underscores again the need,in account—
ing discussion, for more care in the presentation of what—
ever evidence is intended to Support reported conclusions.
The Mattessich study was reVealed to be not a general
theory Of accounting at all but rather a theory about
accounting in general, or a definition of an accounting
SVStem. Following his outline for a system, a general

g discussed at the present time

theory of the type bein

for, any hypotheses proposed to fill the

seems impossible;

place of his surrogate assumptions must be related to some
Specific accounting objective and hence the system becomes

 

 

 

specific. The Carl

cf the difficulty <
the subject matter
the possibility t'
ditional i‘inanciz
expectations at

accounting idea:

it is pos
commons m
be 00nstructe
QWm, indee
principles (
related sti
a hierarch
circumstax

a theory
The

What s_h<

of 9mm
enahie
numbe:

of h

\\ '
Qua

be

A

223

specific. The Carlson study serves to provide evidence

of the difficulty of adequately defining the boundaries of

the subject matter for a theory. It also seems to support

the possibility that even a theory encompassing only tra—
ditional financial accounting is probably beyond reasonable
expectations at the present stage in the organization of

accounting ideas in a formal manner.

Some Unfulfilled Predictions

It is possible to make some observations about the
conditions that will exist if such a general theory should
be constructed.

Its basic terms, axioms and theorems

could, indeed, be called the basic concepts, postulates and

principles of acoounting. And, it would present an inter—

related structure of these categories which would produce

a hierarchy of a sort. But much more can be said about the

circumstances which would surround the construction of such
a theory and the problems it would produce.

The claim that a general theory will in some way show
what should p§_done is without support. No amount or method
of examination and observation of what i§_(induction) will

enable a determination of what should be. And no finite

number of applications of presently known deductive rules
of inference will result in a conclusion involving an
"Ought to” without an assumption of the same nature.

The theory could not be said to be right or wrong, to

be true or false. Resting on the truth of its postulates

 

 

 

islms), which, i‘
hepostulates of
desye reasoning
is merely useful.
Nor, if mor
mud mm rules
esopposed to a
sh anomatica
mportantly, 8
RS 90stul ate
fie subject t
Ks rules of
Practical v2
Even i e the
Lthese Words
Order that
into Prac'
Sistent v
this qua
meOhsig
filthns
0? def
else a
must
the,

this

224

(axioms), which, in turn, rests on either the truth of
the postulates of other theories or some method of incon—

clusive reasoning (induction), the theory isn‘t true; it
is merely useful.

Nor, if more than one general theory were proposed,

would any rules of logic demand that one theory be accepted

as opposed to another. That is, ”a” general theory need

not automatically become ”the” general theory, or, more
importantly, a generally accepted theory of accounting.
Its postulates, and the definitions of its nonbasic terms
are subject to acceptance and logically open to question°
Its rules of inference may be challenged. To be of any
practical value, its basic terms require interpretation.
Even if the interpretation utilizes seemingly simple words,
these words will undoubtedly require clear definitions in
order that the principles of the theory may be translated

into practices. Thus, while the principles will be con—

sistent with each other if the system is proved to exhibit
this quality, practices could be, in a manner of speaking,
inconsistent by virtue of the adoption of different defi—

nitions by different practitioners. Hence, a single set

Of definitions also will have to be generally accepted,

else alternative practices will exist, And accountants
must not expect to settle such problems by searching for
the right or true definitions, for definitions are neither

true nor false, but only useful and workable.

 

 

 

The very Que)
mother in most C
acceptance: canfi‘
the advent 0f ar
increased use 0'

ing the various
attention t0 t7
the controveTE

ferehces of 0'

with either E

or the more
The ave
the usefuln
special the
problems,
the more ?
general t
iution 0‘
other, if
he of m
dispute
busihe
in th<
defir
becoi

ing

A

225

The very question which appears in some form or

another in most current accounting discussions, that of
acceptance, cannot, then, be expected to disappear with

the advent of any general theory of accounting. An

increased use of formalization in the method of present—
ing the various sides of issues in dispute may draw

attention to those areas which represent the center of

the controversy. But a once—and—for—all-time end to dif—

ferences of opinion is highly improbable in that most deal
with either some important general or specific assumption

or the more difficult area of definitions.

The availability of a general theory would not destroy

the usefulness, or even the need for, what might be called

special theories, or theories designed to cover special

problems, such as, for example, cost accounting, or even

the more limited problem of deferred income taxes. The
general theory may provide some assistance in the reso~
lution of as yet unborn problems, but reference to those
other, necessarily less abstract theories, would doubtless
be of more value in connection with any specific future
disputes. And the possibility will always exist that new
business facts may exhibit characteristics not observed
in those used in the formulation of currently workable
definitions, so that re—examination of these definitions
becomes necessary, bringing with it new conflicts regard—

ing definition selection. It can be seen, then, that if

 

 

and when any gene?

1
the account ant S i

There is so
the hope of first
search need not
be immediately

pietely unf ami

matters relat‘.‘
evidenced by
sections poii
able practic

the use of .

hroblem are
about whic'
A the
construct
CEductiom
meets mo
set of ,
t0 deri
Study
would
term
tere:

ant s

 

226

and when any general theory of accounting is proposed,

the accountant‘s problems will hardly be over.

The Beginning
There is no need for the profession to abandon

the hope of finding a general theory of accounting. The

search need not be discontinued. Nevertheless, it should

be immediately recognized that most accountants are com-

pletely unfamiliar with many of the most fundamental

matters relating to logical inference. This fact is

evidenced by the several misunderstandings and miscon-

ceptions pointed out in this paper. There is consider~

able practical advantage to be gained by introducing

the use of any logical method in rather simple and limited

problem areas, perhaps even in connection with positions

about which there is no apparent disagreement at present.
A theory is not necessarily a complex thing. The

construction, within the confines of a known natural

deduction system, of a proof of a single proposition

meets most of the qualifications for membership in the

set of theories. The experiment mentioned in Chapter VI,

to derive one of the principles suggested by the Illinois
StUdy Group from selected postulates and definitions,
would constitute a somewhat elementary theory as that

term is used in this paper. That the principle was unin-

teresting and would probably be accepted by most account—

ants without any proof at all does not negate the value of

 

 

its demonstration .

appeared to be ne
iirectly proportf
pleteness of acc
might not so ea:
theory might be
sell—written p1
recording of 6
one of the ma
involve segre
held from th
some of the
example, if
aifgument i:
may also i
logical a

in this s

to expre
Positim

Be

method

the St

(must

 

the p
A CE

Writ

227

its demonstration. On the contrary, the postulates which
appeared to be necessary in the proof (e.g., assuming a
directly proportionate relationship between the com—
pleteness of accounting records and their reliability)

might not so easily be accepted. A more interesting

theory might be proposed after a careful analysis of a
well—written professional article in opposition to the
recording of deferred federal income taxes. Undoubtedly
one of the major problems in any such analysis would

involve segregating the reasoning for the position being

held from the reasoning for or evidence in support of

some of the premisses used in the main argument. For
example, if one of the assumptions necessary for the

argument is'that taxes are not an expense, the article
may also include considerable space devoted to an ana_
logical argument in support of this assumption. Even
in this theory, a single proof would probably suffice
to express the reasoning followed in support of the final
position.

Beginning the utilization of appropriate logical
methods with such modest objectives corresponds with
the suggestion made in this paper that attempts at theory
construction on a small scale would be appropriate at
the present time, as well as more likely to succeed.

A certain amount of resistance has confronted those

Writers who have utilized complex mathematics as part

 

 

of the present atic
not as widely kno
to expect that ti
position present
tance. As a pr:
of experimentat
0f the profess
selves of the
Preparation w
gently analy:
might be pro
logic in cox

issues .

 

228

of the presentations of their ideas. Since logic is
not as widely known as mathematics, it is reasonable

to expect that the use of logical methods to assist in
position presentation will meet the same type of resis—
tance. As a practical matter, then, during this time
of experimentation with simpler theories, the members
of the profession could take steps to convince them—
selves of the acceptability of the methodology. This
preparation will enable the accountant to more intelli-
gently analyze and more readily accept whatever results

might be produced by the future use of some method of

logic in connection with more controversial and basic
issues.

 

 

PO

Postulates
I,—
h—l Quail

making rationE
song alterna‘
cmsequences.

h—‘Z @39
produced are
directly cor

h—3 lg
Economic ac
entities.
up partic

n-t
period) l
periods 0
clearly t

t—S
monetary
which g
and cap
which n

B
proces
finan<
rest‘

gene
actu

exp

 

APPENDIX A

POSTULATES AND PRINCIPLES OF THE

ACCOUNTING RESEARCH STUDIES

Postulates

A—l Quantification Quantitative data are helpful in
making rational economic decisions, i.e., in making choices

among alternatives so that actions are correctly related to
consequences.

A-2 Exchange Most of the goods and services that are
produced are distributed through exchange, and are not
directly consumed by the producers.

A—3 Entities (including identification of the entity)
Economic activity is carried on through specific units or
entities.

Any report on the activity must identify clearly
the particular unit or entity involved.

A—A Time period (including specification of the time
period) Ewconomic activity is carried on during specifiable
periods of time.

Any report on that activity must identify
clearly the period of time involved.

A_5 Unit of measure (including identification of the
monetary ufiit7__N5heyIis~the common denominator in terms of
Which goods and serviceS. including labor, natural resources,
and capital are measured. Any PGPOPt must clearly indicate
which money (e.g., dollars, francs, pounds) is being used.

3-1 Financial statements The results of the accounting
pPOCess are expressed in a set of fundamentally related
financial statements which articulate with each other and
rest upon the same underlying data.

B~2 Market prices Accounting data are based on prices
scherated by past, present or future exchanges which have
aCtually taken place or are expected to.

 

B—3 Entities The results of the accounting process are
expressed in terms of specific units or entities.

229

 

 

Tentative
relatively short pe
allocations betweer
are required.

—

C-l Continui
limited life In
the entity should
indefinitely. In
has a limited lif
in operation inde

0—? Object
and the related
retained earning
mal recognition
lime at which t

0-3 Consf
TOT 3 given en
ment of its pc
followed consi

 

0-” an
on 3 Stable (y

C-S Pl
“at which i

W
In em
Chapter VI
my thee.
or Val—hat

A
buSlRESE
aSSignS

introdu
aSSlgnE

B
the an

230

B—A Tentativeness The results of operations for
relatively short periods of time are tentative whenever

allocations between past, present, and future periods
are required.

C-l

Continuit (including the correlative concept of
limited life) In the absence of evidence to the contrary,

the entity should be viewed as remaining in operation
indefinitely. In the presence of evidence that the entity

has a limited life, it should not be viewed as remaining
in operation indefinitely.

C—2 Objectivity Changes in assets and liabilities,
and the related effects (if any) on revenues, expenses,
retained earnings, and the like, should not be given for—
mal recognition in the accounts earlier than the point of
time at which they can be measured in objective terms.

C—3 Consistency The procedures used in accounting
for a given entity should be appropriate for the measure—
ment of its position and its activities and should be
followed consistently from period to period.

C-A Stable unit Accounting reports should be based
On a stable measuring unit.

C-5 Disclosure Accounting reports should disclose
that which is necessary to make them not misleading.

M

In accordance with the writer's limitation noted in
Chapter VI, the list of principles which follows includes

only those not related to specific suggestions for pricing

or valuations.

A) Profit is attributable to the whole process of
business activity. Any rule or procedure, therefore, which
aSSigns profit to a portion of the whole process should be
continuously re-examined to determine the extent to which it
introduces bias into the reporting of the amount of profit
assigned to specific periods of time.

 

B Changes in resources should be classified among
the amounts attributable to
1) Changes in the dollar (price—level changes)
which lead to restatements of capital but
not to revenues or expenses.

 

 

2) Chax
bel
whi

C) All as
investments of
should be reco:
financial stat
pendent of the

a) All
in the accoun

G) In 2
classified i‘
learned surp
classified 2
underlying 1
Capital,

ll) A
reveal the
“lit compar
the data E
Denses, g2

231

2) Changes in replacement costs (above or
below the effect of price—level changes)
which lead to elements of gain or of loss.

3)

Sale or other transfer, or recognition of
net realizable value, all of which lead to
revenue or gain.

4) Other cases, such as accretion or the dis—

covery of previously unknown natural
resources.

C) All assets of the enterprise, whether obtained by
investments of owners or of creditors, or by other means,
should be recorded in the accounts and reported in the
financial statements. The existence of an asset is inde—
pendent of the means by which it was acquired.

. E) All liabilities of the enterprise should be recorded
1n the accounts and reported in the financial statements.

G) In a corporation, stockholders' equity should be
classified into invested capital and retained earnings
(earned surplus). Invested capital should, in turn, be
classified according to source, that is, according to the

Underlying nature of the transactions giving rise to invested
capital.

H) A statement of the results of operations should
reveal the components of profit in sufficient detail to per—
mit comparisons and interpretations to be made. To this end,
the data should be classified at least into revenues, ex-
penses, gains, and losses.

 

 

 

POSTl

1) hocoun‘fi":
usefulness for w

2) Economi
enterprises, 3“
accountability
analysis and re

3) Acccu‘
on an enterpr'l
enterprises ox
results essen‘

A) Tran
consummated "
this money p
measurement

An
change of e
suasive evf

6) i
prise engz
within it

hincipl‘
1)
prise e)
maintai
Changes

APPENDIX B

POSTULATES AND PRINCIPLES OF THE

ILLINOIS STUDY GROUP
Postulates

1) Accounting data and reports have validity and
usefulness for widely differing purposes.

2) Economic activity is engaged in by identifiable
enterprises, and these enterprises constitute units of

accountability and centers of interest for accounting
analysis and reports.

3) Accounting is primarily concerned with the effect
on an enterprise of its exchange transactions with other
enterprises or individuals and with events which produce
results essentially the same as exchange transactions.

A) Transactions in which an enterprise engages are
consummated in terms of a stated or implied money price, and

this money price provides an appropriate basis for accounting
measurement and analysis.

5) An enterprise will continue without significant
Change of environment and activities unless there is per—
suasive evidence to the contrary.

6) The flow of economic activity in which an enter~
prise engages can be related to specified time periods
within its_life on a meaningful basis.

Principles

1) To provide reliable information concerning enter-
prise economic activities: accounting should develop and
maintain a complete record of all events which result in
changes in enterprise assets and equities.

a All enterprise transactions (exchange trans—
actions and equivalent events) should be
recorded in terms of their effect on enter~
prise progress and status.

 

232

 

‘0) TO av
matur
effe(
ente
pone

traY

hav<

2) Enters)?i
tecy terms as W1]

interests in an ‘
a) En

tt
me

a.
b) T
t
l

3) The :
sity the fina
their use for

h) on
proper ident
prise, and v
dissimilari‘
prise progr

5) Tc
prise stat
may be red

6) ‘
ficat ions
evident i

7)
E)Qpropri
progres:
periodi

 

233

b) To avoid erroneous conclusions based on pre.

mature analysis, recognition of the ultimate
effect of some enterprise transactions on
enterprise progress and status may be post_
poned until, but only until, the series of

transactions of which they are integral parts
have been completed.

2) Enterprise data should be expressed in such mone~
tary terms as will facilitate their use by the various
interests in an enterprise.

a Enterprise transactions should be recorded at
their money price in order that these data
may be recorded in a reasonably homogeneous

) and objective manner.

b

To achieve reasonable comparability and com-
pleteness, enterprise transactions in Which
necessary quantification is not apparent
should be recorded by means of reasonable
approximation of exchange prices.

_ 3) The accounting records should accumulate and clas-
Sify the financial data in such'a manner as will facilitate
their use for analysis, interpretation, and reporting.

4) Classifications should be used which will permit
proper identification of assets and equities of the enter-
prise, and which will recognize significant similarities,

dissimilarities, and interrelationships relevant to enter-
prise progress and status.

5) To reflect enterprise progress and changes in enter~

prise status, modification of enterprise transaction data
may be required.

. 6) The nature, extent, and effect of any recorded modi-
fications of enterprise transaction data should be clearly
evident in the accounting records.

7) To provide enterprise interests with information
appropriate to their needs, reports reflecting enterprise
progress and status should be prepared and made available
periodically.

a) All enterprise transactions should be rel
ported to reflect their effect on enterprise
progress and status.
.Information reflecting events and deve10p_
ments other than enterprise transactions

may be required to supplement or complement
reports of basic transaction data.

8)

To provide enterprise interests with information in

 

 

 

aform approprii
arrayed as to r
guidance 0f the

a) E

,
C

b)

234

a form appropriate to their needs, reports should be so
arrayed as to reflect managerial success or failure in
guidance of the enterprise economic endeavors.

a Significant similarities, dissimilarities
and interrelationships in the reported data
should be clearly indicated in the report“

b) Reporting practices should be followed
which facilitate comparisons over time and
among enterprises.

c) A clear distinction should be maintained
in accounting reports between basic enter-
prise transaction data an

d supplementary
or complementary information.

 

 

l) Monet
values, expre
phic to the s

2) Time
(or minim—2K)—

3) is
reflecting E

no lie
that a valu
sisting of

5) h
ordered pa
and the at
period.

6)
We,
to change

to enter

Lesa a
Change 3

81)
w
Comment
t0 eco

 

9
the f-

l
“PM
Wes
Seas
and

 

APPENDIX C

THE BASIC ASSUMPTIONS OF

PROFESSOR MATTESSICH

1) Monetary Values There exists g set g§_additive

values, expressed in a monetary unit; this set is isomor—
phic to the system of integers plus the number zero.

2) Time Intervals There exists 3 set 9: elementary

(or minimal), additive time intervals.

3) Structure There exists a structured set of classes
reflecting significant categories of an entity.

4) Duality For all accounting transactions, it is true
that a value is assigned to a three—dimensional concept con—

sisting of two accounts and a time instance.

5) Aggregation Every balance assigns a value to an
Ordered pair; the latter consists of the pertinent account
and the above stated period which starts with the accounting

period.

6) Economic Objects There exists g_set 9£_economic
apiegts, whose values and physical properties are subJect
to change.

7) Inequity of Monetary Claims There exists a custom
tO enter debts with the understanding pg redeem them ip
l§E§l_tender at face value-—whether meanwhile price—level
Changes vis—a:Vis this legal tender have occurred or not.

8) Economic Agents There exists a set of economic
§E§E£§_Who set specific goals to an accounting system,
Command resources, and make plans and deCiSions With regard
to economic actions.

Entities There exists §_set 9£_entities setting

__._____. .
the frame for economic actions.
There exists a set of gm:

10) Economic Transactions . . __h f
ElElEél_phenomena, called economic transactions. Eac to

these transactions assignS, by means of empirical hypo hf )
eses, a value to an ordered pair of transactors (categories

and a time instance.

235

 

 

 

ll Valuati
mining the value

12) Realiz:
specifying which
are exercised by
Such a change ef
current income .
ovmers‘ equity
0? t3) affects
rent income of

13) Clas
required to es

I it) an
Teddired to d
0‘3 aggregacic

formulated.

15) m
“‘9 EXpectet
deration, a]
or Bub-peri

it) i
filing the '
accounting
more 00mp-I

l7)
tcriteria
0T Pelate
m.

determ—h
01‘ f1 OW

NOTE;

 

236

ll) Valuation There exists a set of hypotheses deter—
mining the value assigned to an accounting transaction.

12) Realization There exists a set of hypotheses,
specifying which of the following mutually exclusive effects
are exercised by a chan e g£_gp_entity‘s economic object(s).
Such a change either: ll) affects the value assigned to the
current income of the entity; or (2) does not affect the
owners' equity of this entity (within the specified period);
or (3) affects the owners' equity without affecting the cur-
rent income of the entity.

13) Classification There exists a set of hypotheses
required to establish a chart of accounts.

14) Data Input There exists a set of hypotheses
required to determine the form of data input and the level
Of aggregation for which accounting transactions are to be
formulated.

15) Duration There exists a set of hypotheses about
the expected life of the entity (or entities) under consi-
deration, and the duration of individual accounting periods
or sub-periods.

16) Extension There exists a set of hypotheses speci—
fyins the empirical conditions under which two or more
accounting systems can be consolidated and extended to a
more comprehensive system.

. 17) Materiality There exists a set of hypotheses
(Criteria) determining if_and when an economic transaction
Ortrelated event is to be reflected py 33 accounting trans—
w. I“ ‘—

 

18) Allocation There exists a set of hypotheses
determining the allocation of an entity's economic objects
Or flows of services to subentities and similar categories.

NOTE: The emphasis in the above assumptions originated with
Professor Mattessich.

 

 

 

ProfessC
notations.
of the assur
easily tran
either no ‘
ever a lOC
included
lation w'I
thor giv
to prov
g.

C

 

 

APPENDIX D

THE BASIC ASSUMPTIONS OF

PROFESSOR CARLSON

Professor Carlson formulates his assumptions in logical
notations. In most cases, he provides a loose translation
Of the assumption, but on occasion, the assumption is not
easily translated into good English, and the author offers
either no translation at all, or a very literal one. When—
ever a loose translation has been given by the author it is
included in the listing below; otherwise, the literal trans—
lation will be given. For Q—8 and Q—lO, for which the au—
thor gives no English equivalent, the writer has attempted
to provide one.

C—l Every Economic Service Inflow has a Source.

0—2 A thing sold expires because of the Source
generated by the sale.

C~3 The Source generated by the sale of an item is
received because that item was foregone.

0—4 For all x, and for all y, if x is received as a
result of foregoing y, then y expired in the
current fiscal period.

0-5 The Economic Service Inflow itself can always
be distinguished from the Source by which it
was provided.

0-6 If a thing obligates assets, then it expired for
the same reason that it obligated assets.

237

 

 

Q-3

Q-A

Q-S

Q-6

Q“?

238

All expired costs are matched with some Source.

If an.Economic Service Inflow is Part of another
and neither has expired, then the one is matched

with the other.

Anything which obligates Assets because of Revenue

or Non—operating Revenues is aanxpired Cost.

If one thing expires because of another, and one
is a Source and the other is an Economic Service

Inflow, then the one is matched with the other.

Net Non—operating Revenues and Non—operating
Losses are matched with Net Income.

Net Income to Retained Earnings is matched with
Net Worth Before Transfer of Current Income to
Retained Earnings.

For all x and for all y, if x is the Gain on y,
then there exists a 2 such that z is the Source
from the sale of y.

For all x and for all y, if x is Greater than y

or if y is Greater than x, then there exists a 2

such that z is the Difference between x and y.

At acquisition both an Economic Service Inflow
and its Source are quantified at the Implied
Cash Cost of the Economic Service Inflow.

If the Net Book Value of an Asset is Greater than
the Expected Net Benefits to be obtained from it,

than the Difference between the Net Book Value
and the Expected Net Benefits expired during the
current fiscal period.

[If the Potential Margin is less than the Conserva-

tive Estimate of Next Period‘s Margin, then the
Difference between the two expired during the
current fiscal period.

For all x, y, and 2, if x is Greater than y, and
y is Greater than 2, then x is Greater than 2.

For all x, y, z, w, and v, if x is Greater than y
and z is Greater than y and either x is Greater
than 2 or z is Greater than x and w is the

Difference between x and y and v is the Difference
between x and 2,

then w is Greater than v.

 

 

 

Q—8 Given x:

Q—9 Two nun
tween i

Q-lO Given
Q—ll The S

but i?
liabi

 

 

Q-8
Q-9

Q-lO

Q-ll

 

239

Given x=y-z and w=y-r and p=x—w, then p=r—z.

Two numbers which are both the difference be—
tween the same two things are identical.

Given x=z-w, and y=z-x, then w=y.

The Source of anything which obligated Assets

but for which no assets.were foregone is a
liability.

 

 

 

 

 

 

irthur Afidersey
of the AC‘

19 2.
. IT
met-FEE:

Black, Max.

Inc” 1‘

Blanche , R01C
New ‘10

Bridgman, E
The M2

Carnap, RU
Intel
Numb

Uni
Chambers
Ne".

Ac

Church
P

Sofie}

Cohei

Con}

A.

 

 

SOURCES CITED

Books

Arthur Andersen & Co. Accounting and Reporting Problems
of the Accounting Profession. Second Edition. October,
1962.

 

 

. The Postulate of Accounting-What It Is, How It
Is Determined, How It Should Be Used. 1960.

 

Black, Max. Critical Thinking. New York: Prentice—Hall,
Inc., 1952.

 

Blanche: Robert. Axiomatics, Translated by G. B. Keene.
New York: The Free Press of Glencoe, 1962.

Bridgman, P. W. The Logic of Modern Physics. New York:
The Macmillan Company, 1928.

 

Carnap, Rudolf. Foundations of Logic and Mathematics.
International EECyclopedia of Unified Science, Volume I,
Number 3.) Chicago: University of Chicago Press, 1939.

 

. Logical Foundations of Probability. Chicago:
University of Chicago Press, 1950.

 

Chambers, R. J. Towards a General Theory of Accounting.
Melbourne, Australia: The Australian Society of
Accountants, 1962.

 

Church, Alonzo. Introduction to Mathmatical Logic, Volume I.
Princeton, New Jersey: Princeton University Press, 1956.

 

COffey, P. The Science of Logic, Velume II: Method, Sci—

fgcg and Certitude. London: Longmans, Green and Co.,
1 .

Cohen, Morris R. A Preface to Logic. Cleveland: The World
Publishing Company, 1944.

 

Copi, Irving M. Introduction to Logic. New York: The
Macmillan Company, 1961.

 

_.l______. Symbolic Logic. New York: The Macmillan Com-
pany, 1955.

240

 

 

 

Gt
,1931.

ieamside W. Ward
wood Cliffs,

Eaton, ORalph

iiegl, Herbert ax
of Science.
/

1953
inge,cottiob.
by J. L.

Goldberg, Loui
American

Grady, Paul.
Study N\
IIQG Pul

“Pale, Will
sity P-
Leouard, Re

Henry

Lewis) 0112
1932

Littletox
Uni

 

Ac

M366 en
Marple
Mat ’59

MOM

 

241

Eaton, Ralph M General L '
. #229- Ne : .
SOUS, 1931, g W York Charles Scribner's

Fearnside W. Ward and Holther, William B. Fallacy. Engle-
wood Cliffs, New Jersey: Prentice-Hall, Inc., 1959.

Fiegl, Herbert and Brodbeck, May, editors. The Philosophy
of Seience. New York: Appleton—Century-Crofts, Inc.,

1953.

Frege, Gottlob. The Foundations of Arithmetic. Translated
by J. L. Austin. New York: Harper & Brothers, 1950.

Goldberg, Louis. An Inquiry into the Nature of Accounting.
19 5.

American Accounting Association,

Grady, Paul. Inventory of Generally Accepted Accounting
s Enterprises. Accounting Research

Principles for Busines
Study Number . New York: American Institute of Certi-

fied Public Accountants, 19

bility and Induction. Oxford Univer-

Kneale. William. Proba
Sity Press, 1939.
ht Reason. New York:

Leonard, Henry S. Principles of Rig
Henry Holt and Company, 1957-

Lewis, Clarence Irving and Langford, Cooper Harold. Sym—
bolic Logic. New York: Dover Publications, Inc.,

1932.
Littleton, A. C. E§§§y§_on—Acgguntan%y. Urbana, Illinois:
University of Illinois Press, 19 1.

ccounting Theory.

Structure of A
1953.

Accounting Association,

American

 

Madden, Edward H., editor. The Structure of Scientific
Thought. Boston: Houghton Mifflin Company, 1960.

ting Philosophy.

Marple, Raymond P. Toward a Basic Accoun
‘ation of Accountants, 1964.

New York: National Ass001
ting and Analytical Methods.

Mattessich, Richard. Accoun

Homewood, Illinois: Richard D. Irwin, Inc., 196E.

Moonitz, Maurice. The Basic Postulates of Accounting.
Accounting Research Study Number 1. New York: Ameri—
can Institute of Certified Public Accountants, 1961.

 

 

 

wrath, OW)“ E):

national Enc
0f the Uni—JED
university ‘

ntou, William p
1922.

___,_..—— and Li
Accountim
19 O.

Hut chins

Quins, Will 8-3
York: ‘

Russell, 38}
New Yo

Robinson , I
Intro<
Editi

lncox
Salmon, N
Engl

196’;

Searles ,
Y0]

Sprouse
oi

I

1

Store

Stud

Sui

232

 

Nurath,t0tto. Foundations of the Social Sciences. (Inter-
na ional Encyclopedia of Unified Science, Foundation

of the Unity of Science Volume II N '
_ . umber 5. Chica o:
UniverSity of Chicago Press, 1933., ) g

Paton, William A. Accounting Theory. Ann Arbor, Michigan:

.__..__._,_._.___
1922.

and Littleton, A. C. An Introduction to Corporate
Accounting Standards. American Accounting Association,

1930.

POpper, Karl. The Logic of Scientific Discovery. London:
Hutchinson of London, 1959.

Quine, Willard Van Orman. Methods of Logic, Revised. New
York: Holt, Rinehart and Winston, 1959.

Russell, Bertrand. Human Knowledge, Its Scope and Limits.
New York: Simon M. Schuster, l9 .

RObinson, Daniel Sommer. The Principles of Reasoning, An
Introduction to Logic and Scien ific Me ho . Second
Edition. New York: D. Appleton—Century Company

Incorporated, 1930.

'c (Foundations of Philosophy Series)

Salmon, WeSley C. Logi .
ffs, New Jersey: Prentice—Hall, Inc.,

Englewood Cli

1963.
Searles, Herbert L. Logic and Scientific Methods. New

York: The Ronald Press Company, 1938.
Sprouse, Robert T. and Moonitz, Maurice. A Tentative Set

of Broad Accounting Principles for Business Enterprises.
Accounting Research Study Number 3. ew York: American
Institute of Certified Public Accountants, 1962.

 

ng Principles—

Storey, Reed K. The Search for Accounti
Today's Problems in Perspective. New York: American
Institute of Certified Public Accountants, 1963.

StUdy Group at the University of Illinois. A Statement of
Basic Accounting Postulates and Principles. Urbana,
Illinois: for International Education and Re—

Cen er
search in Accounting, 196 -

 
     

    

Suppes, Patrick. Introduction to Logic. (The University
Series in Undergraduate Mathematics) Princeton, New
Jersey: D. Van Nostrand, Inc., 1957.

 

 

T

Ushenko, A' ?'
Brothers, E

Ton Wright) G6.“
Will/153
1951.

voodger, 3' 3'

(Internal)C
Numb er 5-

merican IDS
of the
rise

Anderson , 1
Rep or

Redford, 1
Met‘n
sue.

Bl ough ,
A95

Carlson
31

T1
Chambe
Dein ,

Devi

Ed

233

Ushenko, A P T °
. . he Theory of Logic. New York: Ha er &
Brothers, Publishers, 19 6. Pp

Von Wright, Georg Henrik. A reatise on Induction and

T
Eggiability. London: Routledge and Kegan Paul, Ltd.,

WOOdger, J. J.. The Technique of Theory Construction.
(Internatidfiai’tfiayafafiédia'dedfiffifiiffififiifififii'volume II,

Number 5.) Chicago: University of Chicago Press, 1939.

Articles, Periodicals,
Reports and Other

Amerioan Institute of Certified Public Accountants. ”Report

of the Special Committee on Research Programs,”

Journal of Accountancy, CVI (December, 1958).

Anderson, David S. ”Communication Problems of Financial
Reporting," Journal of Accountancy, CXV (April, 1963).

Bedford, Norton M. and Dopuch, Nicholas. ”Research
Methodology and Accounting Theory-Another PerSpective,”

e Accounting Review, XXXVI (April, 1961).

Th

Blough, Carman ”Principles and Procedures,“ Journal of

Accountancy, CXI (April, 1961)-

”Accounting Theory as a Logical
lished Ph.D. dissertation, Commerce,

1963.

Carlson, Marvin Lee.
System,” Unpub
The University of Wisconsin,

Chambers, R. J. ”Blueprint for a Theory of Accounting,”
Accounting Research, VI (January, 19 5).

, Backward at Research in Account-
,” he Accounting Review, X XVI (January, 1961).
Methodology and Accounting
nting Review, XXXV (July,

Dein, Raymond D ”A Glance
ing

'“Research

Devine, Carl Thomas.
” The Accou

Theory Formation,
1960).
Accounting Principles,"

Editorial. ”The Approach to
CXIII (May, 1962).

Journal of Accountancy,

Flanders, Dwight P. ”Accountancy, Systematized Learning,
and Economics,” The Accounting Review, XXXVI (July,
1961)..

 

 

 

“'1
1 of 30

ha) Charles 3 '
I ourna

Cordon, Wm“ 3'

M1191, Carl G'

. H on
/ I
new)

Jennings, MW
Review ,
'..,

of Acco‘

Ileerekoper
J ourna
,_.

LaSalle, Bi
Accou

Li, David
Enti
(J at

 

 

Matt

Ne‘

233

Gaa, Charles J. "Uniformity of Accounting 'Principles',”

Journal of Accountancy, CXI (April, 1961).

Gordon, Myron J. ”Scope and Method of Theory and Research

in the Measurement of Income and Wealth,” The Account—

ing Review, XXXV (October, 1960

Hempel, Carl G. ”Geometry and Empirical Science,” The
American Mathematical Monthly, LII (January, 1935).

.. ”0n the Nature of Mathematical Truth;H $32
rican Mathematical Monthly, LII (December, 1935).

W

 

Jennings, Alvin R. ”Accounting Research,” The Accounting

Review, XXXIII (July, 1958).

Journal of Accountancy, ”Comments on 'The Basic Postulates

M
of Accounting', CXV (January, 1963).

Kleerekoper, I. ”The Economic Approach to Accounting,”
Journal of Accountancy, C V (March, 1963)-

arch in Accounting,” The

LaSalle, Brother. ”Basic Rese
(October, 1959).

Accounting Review, XXXIV

__.___________fl___

Li, David H. ”Alternative Accounting Procedures and the
Entity Concept," The Accounting Review, XXXVIII
(January, 19 3 .

r the Entity Concept,”

”The Funds Statement Unde
XXX (October, 19 3 .

The Accounting Review, VIII
__,______. ”The Nature and Treatment of Dividends Under
the Entity Concept,” he Accounting Review,

(October, 1960).

Lorig, Arthur N. ”Some Basic Concepts of Accounting and
Their Implications," he Accounting Review,
(July, 1963).

Mattessich, Richard. “Towards a General and Axiomatic
Foundation of Accountancy,” Accounting Research, VIII
(October, 1957).

”The ‘Basic Postulates‘

Metcalf, Richard W.
The Accounting Review, XXXIX (January, 1

in Prospective,”

We Need 'Postulates‘ and

Moonitz, Maurice. "Why Do
‘Principles‘?” Journal of Accountancy, CXVI (December,
1963). "’"

Park, 01in. ”Thought Processes In Creative Writing,” The
Accounting Review, XXXIII (July, 1958).

 

 

 

 

k

NY
We: 1418.001“ 1
ciples ) EH

Queen’om, 50““ V?"
ReseaTCh)

Schmidt, Leo A.
Logic in

W

Schradef, 1111‘
Theory ,"

Scott, 133- ‘
TechniO.
(Januai

Spacek, Lec
Accer)‘1

_____‘.
Revié

__.__———

Acc<

__——

Spencer,
1111‘

Spiller
o

l
Stead

Tiet

Vat

235

Pye, Malcolnh ”Reasons, Probabilities, and Accounting Prin—

Clples, The Accounting Review, XXXV (July, 1960).

Queenan, John W. ”Postulates: Their Place in Accounting
Research,' Journal of Accountancy, CXIV (August, 1962).

Schmidt, Leo A. ”Practical Uses of the Device of Formal
Logic in Accountants' Daily Work,” Journal of
Accountancy, LXXXVIII (November, 19 9 .

Schrader, William J. ”An Inductive Approach to Accounting
Theory,‘ The Accounting Review, XXXVII (October, 1962).
SCOtt, DR. ”The Influence of Statistics upon Accounting

V

Technique and Theory,” The Accounting Review, XXI
(January, 1939).

Spacek, Leonard. ”Are Accounting Principles Really Generally

Accepted?” Journal of Accountancy, CXI (April, 1961).

. ”Need for an Accounting Court,” The Accounting

ReView, XXXIII (July, 1958)-

 

"Solution to the Principles Dilemma," The

1963).

___.__——-———

Accounting Review, XXXIX (April,

Spencer, Milton H. ”Axiomatic Method and Accounting Science,”
The Accounting Review, XXXVIII (April, 1963).

Spiller, E. A., r. ”Theory and Practice in the Development
of Accounting,“ The Accounting Review, XXXIX (October,

1963).

Stead, Gordon W. ”Towards a Synthesis of Accounting Doc-
trine,” The Accounting Review, XXIII (April, 1938).
ples, Practices and

”Accounting Princi
cxv (April, 1963).

Tietjen A. Car
Methods," Journal of Accountancy,

Vatter, William J. ”Postulates and Principles,

of Accounting Research, I (AUtumU; l9 .

o

” Journal

Windal, Floyd. ”Legal Background of the Accounting Concept
of Realization,” The Accounting Review, XXXVIII

(January, 19 3 .

 

 

 

L

PIE; 381°01nh
ciples:

Queenan, J 0m.
Reseafc

Schmidt: Le(
Loglc'
Accou“
If,“

Schrader, 1
Theor

Scott, DR
Tech‘
(Jan

Spacek, T
Acc

__.———-

Re

—.

 

A

.—

Spence

Spill

Ste:
Tii

V2

 

235

”Reasons, Probabilities, and Accounting Prin-

Pye, Malcoln.
ciples,” The Accounting Review, XXXV (July, 1960).

Queenan, John W. “Postulates: Their Place in Accounting

Research,” Journal of Accountancy, CXIV (August, 1962).

Schmidt, Leo A. ”Practical Uses of the Device of Formal
Logic in Accountants' Daily Work,” Journal of
Accountancy, LXXXVIII (November, 19 9 .

”An Inductive Approach to Accounting

Schrader, William J.
ting Review, XXXVII (October, 1962).

Theory,” The Accoun

Scott, DR. ”The Influence of Statistics upon Accounting
Technique and Theory,” The Accounting Review, XXIV
(January, 1939).

k, Leonard. ”Are Accounting Principles Really Generally

al of Accountancy, CXI (April, 1961).

Space
Accepted?” Journ

. ”Need for an Accounting Court,” The Accounting
Review, XXXIII (July, 1958).

. ”Solution to the Principles Dilemma," The
Accounting Review, XXXIX (April, 1963).

Spencer, Milton H. ”Axiomatic Method and Accounting Science,“
The Accounting Review, XXXVIII (April, 1963).

E. A., Jr. ”Theory and P
Accounting,” The Accounting Review, XXXIX
1963).

ractice in the Development

Spiller,
of (October,

Stead, Gordon W. ”Towards a Synthesis of Accounting Doc-
' XXIII (April, 1938).

trine,” The Accounting Review,

Tietjen A. Carl ”Accounting Principle

Methods,” Journal of Accountancy,

Vatter, William J. ”Postulates and Principles,“ Journal

of Accounting Research, I (Autumn, 19 3). -

Windal, Floyd. ”Legal Backg
of Realization," The Ac
(January, 19 3 .

s, Practices and
CXV (April, 1963).

round of the Accounting Concept
counting Review, XXXVIII

 

 

 

 

 

 

 

 

 

 

 

 

 

 

, 111111111111111111111111.