@
Pergamon
Accoutatfng, organfzatfonsandSocfety, Vol.22, No.7, pp.691-712, 1997
~ 1997Etsevier ScienceLtd
PrintedinGreatBritain. AUrigbtsreserved
0361-3682/97$17.00+0.00
PII: S0361-3682(96)00039-6
THE UNBEARM3LE AMBIGUITY OF ACCOUNTING
BRENDANMcSWEENN
Universiy of Warwick
Abatract
The paper arguesthat neithercurrent, nor reformed,accountingcan makeunambiguousrepresentations,
but concludes that the pursuitof that unrealizableidealis nonethelessperfectlyreal and eminentlypro
ductive. IWo significanttexts in which the claim of judgement-freeaccounting(accrual or cash-flow)is
privilegedare amdysed.Theirattemptsto explain,as distinctfromsimplyassert,the possibilityarc shown
to requirea series of selfertcelling rhetoricalmoves. A numberof implicationsof the anatysisarc then
considered.In contmstwith someptior literaturewhich hasconcludedthat a generalabandonmentof the
myth of unambiguousaccounting representationsis both desirableand possiblethe article argues, in a
discussionof the notion of a “regulativeideal”,that there is no necessatylinkbetweentheircritiquesand
the action they advocate. ~ 1997 ElsevierScienceLtd
The idea that accounting representations are,
or could be, unambiguous,that is, unequivocal
re-presentations of an uninterpreted reality is
regularly asserted in both professional and
academic literatures. For example, the Research
Committee of the Institute of Chartered
Accountants of Scotland (ICAS) states “that all
financial reports ought to reflect economic
reality” (Research Committee of the Institute of
Chartered Accountants of Scotland, 1988, p.
18). Lee insists that the accounting “method of
measurement adopted [should] not distort the
underlying reality being reported. Otherwise
correspondence will be achieved inadequately”
@e, 1984, p. 13). Simmonds assumes that
accounting data can be uncontaminated.
Describing “strategic management accounting”, he states: “the data is factual [as
the]. . competitors’ plants, their products, their
sales forces, their outlets, their suppliers and
their customers exist physically. And the competitors’ actions are real and visible in the marketplace” (Stnmonds, 1989, p. 12). The late
David Solomons succinctly described the position when he stated in a paper addressedto the
UK and Irish Accounting StandardsCommittee,
that the “right test of the fitness of an accounting principle or procedure . . . is whether it
‘tells it like it is’” (Solomons, 1989, p. 8). Elsewhere, in multiple discourses about accounting, including those involving auditing,
corporate failure, and changes in public
management the notion of unambiguous
accounting representation is the normative
criterion which is said to distinguish between
what is true and false, acceptable and unacceptable (see for example, Banham, 1987;
McSweeney, 1994; Power, 1994).
The capability of current, or a reformed,
accounting to produce unambiguous
I am gratefulto the anonymousreviewersfor their encouragementand constructivecomments and to those who offered
adviceandsuggestionson earlierdrafts:RichardBolandJr., RobBryer,SheilaDuncan,AnthonyHopwood,andparticipants
at a researchcolloquiumat CaseWestern ReserveUniversity.

691
692 B. McSWEENEY
representations has been challenged (see for
example, Hines, 1991). The aim of this article is
threefold: first, to contribute to that critique
through a textual analysis of two significant
texts which privilege and seek to justify that
capability;secondly to question the implication,
drawn in some earlier criticisms, that a rejection of the possibility of unambiguousaccounting necessarily requires its abandonment as an
ideal for everyday accounting practice; and
thirdlyto suggest that its continuation is in part
explicable by the constitutive and demarcation
roles the myth plays in accounting practice at
variouslevels. I want to suggestthat one cannot
understand the geneml production of accounting representations outside of the appeal of the
possibility of “objectivity”. The paper is organized as follows. Fir8t, some prior literature is
discussed, then the two texts are examined,
and finally,some implicationsof the analysisare
explored.
PRIOR STUDIES
Before examining the texts, four characteristics which would be requiredby an unambiguous accounting representation are considered.
In this context, some prior studies which have
argued, or implled, that accounting representations are ambiguous are briefly discussed and
the additional support provided for those
claims by the analysis in this article is
suggested.
How would we know that an accounting
representation was a non-arbitrarytruth? It is
this author’s view that there are either: (a) no
such truths, or (b) that while there are such
truths they could only be known from a “god’seye view”. Since none of us occupies that
position the truth any of us tind compelling will
always be judgmental. But that is a contested
claim. Less controversially it possible to say
what an unambiguousaccounting truth would
not be like (Fish, 1994).
Its re-presentationwould not be influenced
by any imposition of historical,cultural,or
othervalues
Two types of studies are particularly noteworthy in suggestingthat such values do influence the makingof accounting representations,
First, there are the extensive number of studies
of international accounting differences which
have shown that accounting measurements of
corporations’ accomplishments is strongly
influenced, in part at least, by local, not universal, factors (see for example, Radebaugh,
1973; Belkaoui, 1991). This literature clearly
suggests that contingent values have a role in
accounting-representation-making.The “same
circumstances” are, or can be, represented in
multiple ways. The intluence of diverse values
in the production of representations is also
indicated in the growing number of genealogical studies of the significance in accounting
change of varioussocial and institutionalfactors
(Hopwood & Miller, 1994). As these studies
reveal processes of change to methods of
accounting calculation other than the creation
of, or evolution in, ever better ways of reflecting a wholly separate realitythey suggest that at
best it is inadequate to consider accounting
merely as a mechanism which reflects reality.
The textual analysis in this article of two
significant depictions and defences of accounting’s “correspondence” capability (Lee,
1984)2 exposes their failureto eliminate the use
of values. In doing so it adds to the argument
that not only is the criterion not currently
satisfied,but that it could never be attained.
Its re-presentationwould not be revisable
If accounting representations are, or could
be, non-arbitrarythen revisions would not be
necessary. Studies which have challenged this
characterisation of accounting’s actual, or ideal,
capabilityinclude those which have shown the
dependence of even ex post accounts on ex
ante information. Various accounting regulations overtly consider the revivability of
*FASB(1980) also uses the term “correspondence” in a tnannersirniIarto “representationalfaithfulness”(FASB, 1980,
pp. 28, 34).
THE UNBEARAB LE AMBIGUITYOF ACCOUNTING 693
accounts of “past” periods, for example,
accounting standards on contingencies (TASB,
1975; ASC, 1980a), post-balance sheet events
(ASC, 1980b), and prior period adjustments
(ICAEW, 1957; FASB, 1977; ASB, 1992). Judgments about possible conditions in a future
state of the world that does not currently exist
are not restricted to such special decisions but
pervade much of what might be regarded as
routine accounting calculations. For example,
historical depreciation requires an estimate of
the future “useful economic life” of the asset
being depreciated. Indeed, even the treatment
of something as an asset requires an assumption
that it will have future benefits (FASB, 1985;
ASB, 1995a). Only on the basis of belief in the
ultimate absolute predictability of all futures
relevant to the event or object being accounted
for can the ideal be said to be eventually
achievable.
The analysis below of the two texts gives
instances in which they acknowledge that
representations about the past or the present
require speculations about future occurrences.
It would not be dependent on its representation, that is it would have existedprior to,
and be unaffectedby itsrepresentation
This characterisation assumes that what is
represented in an account is wholly pre-estabIished so that accounting referents and
accounting representation are regarded as
entirely distinct (Hopwood, 1984; Latour,
1987), albeit there is a hierarchy. In that duality
a representation-freereferent is given the privileged position, while its representation is considered to be merely a derivative (Wood,
1991).3
Studies which have disrupted this sharp
demarcation include those which have examined specific types of representation and have
sought to demonstrate that representation
requires the imposition of categories, definitions, judgments (see for example, Thomas,
1969; McSweeney & Steele, 1995) where a dual
and inseparable process of reflecting and
imposing is revealed. There is also a literature
which has argued for this position by drawing
largely on criticisms of the pure correspondence ideal made in the literature ‘of other disciplines, principally that of the “philosophy of
science”, or the “sociology of knowledge” (see
for example, Hines, 1991, 1988; Tinker, 1988;
Hopper et al., 1987). But its method of analysis
is open to the criticism that it overstates the
extent of rejection of the idea of unambiguous
representations in those disciplines as it only
cites the idea’s critics and ignores its supporters. Furthermore, as it does not examine the
specifics of accounting representations we as
readers are expected to assume that what is
said to characterise some other representations
is also a property of accounting representations. In contrast, the analysis in this paper
does not rely on such circuitous, or analogous,
arguments. Instead it directly addresses the
issue by examining some significant texts
which have attempted to explain and defend
the possibility of unambiguous accounting
representations.
Its meaning in most instanceswould be so
self-suflcientlyperspicuous that it would not
have more than one meaningfor those with
suflcient technicalexpertise
Thiscondition requires that there be a single
meaningof an accounting representation, albeit
only for experts. It could be argued that this
condition is not necessary, or applicable, as it
involves the use and not the production of
accounting representations (Solomons, 1991).
But the studies of the use of accounting representations which show that multiple meanings,
readings, are possible, that meanings do not
exist as such, that is as freestanding and “natural” entities which are communicated, but are
produced through readings, interpretations,
also provide a challenge to the possibility of
unambiguous accounting (see for example,
Boland, 1989, 1993; McSweeney, 1989, 1995).
3Buttreatingaccountingrepresentationsas a meansto pristineclarityleads,perhapsinevitablyat times,to the confusion,or
conflation,of referentand representation.
694 B, McSWEENEY
By illustrating that the reading of accounting
representations is not the mere receipt of
“communications” but is a complex process of
interpretation they imply that the producers of
accounts are themselves involved in an interpretative act—that the act of representing is
co-extensive with the act of reading. Only by
assuming that the users and producers of
accounts belong to a different species—the
users as simpletons or idiosyncratic, and the
producers as sophisticates with transparent
access to the “real”—can we avoid this conclusion. The analysis in this article of the two
texts reveals that in places they both acknowledge that some accounting representations may
not be based on the features of the supposed
reality but instead must rely on imagined uses
to which a particular representation might be
put by an imaginedreader.
THE TWO TEXTS
The texts analysedare: US FinancialAccounting Standards Board’s Statementof Financial
Accounting ConceptsNo. 2: QualitativeCharacteristicsof Accounting Information (1980),
hereafter FASB (1980), and Tom Lee’s Cczsb
FZowAccounting (1984), hereafter Lee (1984).
The first, is in part, an attempt by a major regulatory body (the US Financial Accounting
Standards Board) to explain and defend the
criterion of “representationalfaithfulness”.The
second text was selected because it explicitly
accepts that “accrual based” accotinting representations are unavoidablyarbitrary,but argues
that cash-flow accounting can be non-judgemental. Both texts chase what Derrida has
called “presence” (Derrida, 1981): a wholly
self-sufficientand accessible point of reference:
“economic phenomena” (FASB, 1980, para.
68); the “underlying economic reality” (Lee,
1984, p. 13). In the analysisbelow the ideal of
non-arbitrary representations, privileged in
both of the texts, is contrasted with the mode
of attempted realisation.Attemptsin these texts
to elaborate, as distinct from simplyassert, the
achievement of non-conventional re-presentations are shown to be radically incomplete.
Theirinvocations of the criterion of unambiguous accounting gets repeatedly hedged and
qualifiedin a series of self-cancellingrhetorical
moves when they attempt to describe its
achievement. Notwithstanding both texts’
claims about accounting’s ability to “correspond” (Lee, 1984), to be a “faithful representation” (FASB, 1980), each gives us the
means to think againstthat possibility.
The two texts are now analysed.To be as fair
as possible to them the readings, in the main,
follow the sequential order in which the texts
were originallywritten.
Statement of financial accounting concepts
number 2: the qualitative characten”stics of
accountinginformation
FASB, 1980 is one of a series of publications
in the U.S. Financial Accounting Standard
Board’s attempt to develop and outline a “conceptual framework for financial accounting
and reporting” (FASB, 1976, 1980; Peasnell,
1982; FASB, 1985; Dopuch & Sunder, 1980;
Macve, 1981; Bromwich, 1985; Power, 1989;
Hopwood, 1990; Hines, 1991). The concep
tual framework, the FASB stated, aims to set
out:
“a coherent systemof interrelatedobjectivesand fundamentalsthat is expected to lead to consistentstandards.. the fundamentalsare the underlyingconcepts
of financial accounting-concepts that guide the
selectionof transactions,events,andcircumstancesto
be accountedfor, theirrecognitionandmeasurement,
and the means of summarizingand communicating
them.. .Concepts of that type are fundamental in
the sense that other concepts flow from them and
repeated reference to them will be necessary in
establishing,interpreting, and applying accounting
andreportingstandards(FASB,1980).”
The purpose of FASB(1980), the FASBstates,
“is to examine the characteristics that make
accounting information useful” (p. 26, emphasis added). “Relevance and reliability” are
described as “the two primary qualities” that
achieve this (p. 27 and para. 15). “Reliability”
stems, FASB (1980) states, “from two characteristics. . .representational faithftdness and
verifiability” (para. 62). The sections in FASB
THE UNBEAR4BLE AMBIGUITYOF ACCOUNTING 695
(1980) which primarily consider these attributes (paras. 58-89) are now examined.
Representationalfaithfulness. “Representational faithfulness” is described in FASB
(1980) as: ‘{correspondence or agreement
between a measure or description and the phenomenon it purports to represent. In accounting the phenomena to be represented are
economic resources and obligations and the
transactions and events which change those
resources and obligations” (para. 63) and the
“faithfulness of reported measurements lies in
the closeness of their correspondence with the
economic transactions, events, or circumstances that they represent” (pm. 86). Having
first defined “representational faithfulness”
FASB(1980) then attempts to describe what it
means, what distinguishesa faithfulrepresentation from something else. Maintaining the
purity of the demarcation criterion becomes
problematic.
Replacement. The possibility that a representation may not correspond with what is
intended to be re-presented is acknowledged.
This qualification is made during the use in
FASB (1980) of a number of analogies with
which it seeks to describe “representational
faithfulness” (paras. 68,69, 70).
The first analogyused is “scholastic aptitude
tests of verbal skills” which purport to measure
that aptitude. FASB(1980) first wonders about
the re-presentation ability of the test: “does the
test really measure verbal aptitude? Is it, in
other words, a valid test of verbal aptitude?”
FASB(1980) then questions whether that measure was the intended phenomenon—they
might be measures of something else: “what is
verbal aptitude?”, it asks. The Statement then
goes on to say:
“theproblemof definingintelligenceand of judging
whether intelligencetests validlymeasureit, may be
even more difficultbecause of the many different
manifestationsof intelligence,the problem of separating innate and acquisedabitities,standardizingfor
differences in social conditions, and many other
things(FASB,1980, pm. 68).”
Neither the existence of the type of phenomenon which the representation purports to
identify, nor the capacity of measurement to
faithfully represent such phenomena, is doub
ted, in these paragraphsof FASB(1980), but the
two may, it is accepted, not be matched.
Something different may also, or exclusively,
have been measured. An example is given in
para. 69 of a spelling test orally administered.
Some students, though they can usually spell
well, fail the test because they have hearing
difficulties.“The test score purports to measure
ability to spell, whereas it, in fact, is partly
measuring aural acuity.” However “[t]he test
score lacks true representational faithftdness”
@ASB, 1980, pant. 69). Representations may
not, FASB(1980) here acknowledges, represent
faithfully.The flow direction, as it were, of the
representation may not be fully, or at all, from
phenomenon to representation: “a representation may not be faithftd to the economic phenomena that it purports to represent” (pm.
70).
In the next paragraph there is an acknowledgementof, as it were, a “horizontal”mobility,
of “designed” (para. 71) representations as distinct from a “vertical” re-presentation of an
underlyingreality.
“The discussionin the precedingparagraphs[68-70]
illustratessome of the problemsthat may arisewhen
representationsof economic phenomenaarc used in
ditTerentcontexts than those for which they were
designed…Information that is rcprcsenrationafly
faithful in the context for which it was
designed..may not be reliablewhen used in other
contexts (FASB,1980, pare. 71). ”
How does FASB (1980) try to reconcile this
admission with “faithful representation”?
Whilst it acknowledges that a referent represented as existing may be absent from the
situation being accounted for, such referents
are assumed to exist elsewhere, rather than
being a hypothetical, or constructed, phenomenon. The categories and labels used, and to
which measurements are given, are still treated
as real, that is they exist representation-free
elsewhere, and the labels initially emerged
through recognition of such real phenomenon.
But a distinction is admitted here in FASB
(1980) between the thinking of accounts as
696 B, McSWEENbY
representing faithfully,and their actually doing
so. It is not necessary, it acknowledges, for
what is referred to—and measured-to exist
when represented in accounts. But this is an
admission that a representation may not be a
duplication of something outside itself, that it
might not be a re-presentation. What is absent
cannot be “faithfullyrepresented”.
In this qualification of the possibility of
representational faithfulness, FASB(1980) uses
the term “context” as the re-production situation, or setting. Further qualifyingmoves take
place when the notion of “context” is also used
as that of the interpreting reader. In discussing
the use of a general price index “as a measure
of price change of a specific asset, a purchase
of [sic] a specific consumer, or an acquisitionof
a specific enterprise.. .[t]he index”, FASB
(1980) states, “must be interpretedin the context of what it was designed to do and in view
of the limitations of any averaging process”
(para. 70, emphasis added). Here again the
dividebetween fact and evaluationis broken—
representational fhithftdness now, it is said,
requires the involvementof an imaginedreader.
The notion of the active reader is also used
when FASB (1980) tries to reconcile its
acknowledgement that accounting representations may exclude information with its view
that incomplete informationis biased:
“Freedom from bias, both in the measurer and the
measurementmethodimpliesthat nothingmaterialis
left out of the informationthat may be necessaryto
insurethat it validfyrepresentsthe underlyingevents
and conditions. Refinability implies completeness of
information (FASB, 1980, parz 79, emphasis
added).”
Notwithstandingthis condition FASB(1980)
admits that there may be exclusions so that a
representation may be incomplete. Omissions
mayoccur, it states through choice, because: (i)
it is unnecessary to include certain phenomena
(as they are not “material” enough), or (ii) it is
deemed too costly to do so. These are both
dilutions to the claim that accounting faithfully
represents, but not to its asserted capability of
doing so. However, FASB (1980) also admits
that exclusions may also occur because faithful
representation is not possible (not “feasible”-
pam. 79. It tries to resolve the conflict
between its assertion that completeness is a
necessary condition of representational faithfulness and its admissionthat it may necessarily
be impossible by temporarily switching the
issue away from representation to reader
competence. “A reasonably informed user”
(para. 64) will, FASB(1980) states, be aware of
absences in representations:
“Doesa balancesheet that showsgoodwitlas an asset
purport to representthe companyas havingno goodwill except that shown? An uninformedreader may
well thinkso, while one who is famikarwith present
generaflyaccepted accounting principleswill know
that non-purchasedgoodwill is not included (FASB,
1980, para. 64). ”
Here againthere is an appealto the existence
of informed readers, or users, who will know
that accounts are “relative”, are not “complete”. The concept of reader competence
used in FASB (1980) can open a range of
considerations, for example: how could
accounting ever succeed in anticipating readers’—even a single reader’s-psychic and social
context? But for the purposes of this article it is
sticient to note that FASB (l98O)’S explanation of “representational faithfulness” fails to
limit itself exclusively to an allegedrelationship
between a representation and that which it is
supposed to be re-presenting. It also had to
introduce the active role of an interpreting
reader.
Estimates. Even when the absence of a
phenomenon, or uncertainty about which
phenomenon to measure, is not considered
problematic, that is when it is assumedthat the
phenomenon “is there” to be measured, representational faithfulness may still, FASB (1980)
suggests,be elusive. The information “provided
by financial reporting often results”, it states,
“from approximations, rather than exact,
measures involvingnumerous estimates, classifications, summarizations, judgments, and
allocations” (para. 64). Amongthe examples of
these approximationsgiven is “fungibleassets”.
Rather than being representationally faithful it
“becomes more difficult”, FASB(1980) states:
THE UNBEARAB LE AMBIGUITYOF ACCOUNTING 697
“to substantiatethat the chosen amount does represent the economic phenomenon. since what is
shown as the assets’cost is onty one of severalalternatives.. .if there have been severalpurchasesat different prices and a number of disposalsat different
dates,onlyby the adoptionof some convention(such
as tirst-in,firstaut) can a cost be attributedto the
assets at a particular date (FASB, 1980, para. 66
(emphasisadded)).”
Referringto some other approximation,FASB
(1980) states: “the allocations required by
those procedures inevitably cast at least some
doubt on the representational faithfulness”
(paras. 66-7).
It may not even be certain, FASB (1980)
states, that the cost attributed to an acquired
“asset in the enterprise’s records does faithftdly
represent its cost” (para. 65):
“if a collectionof assetsis boughtfor a specilied
amount, the cost attributableto each individualitem
maybe impossibleto ascertain “ Theacquisitioncostmay
also be dir%cukto determineif assetsare acquiredin
exchange for assetsother than cash, by issuingstock,
or in transactionwith relatedparties.If assetsare converted into other assetswithinan enterprise,as when
raw materialsare converted into other assets witbin
an enterprise, as when raw materialsare converted
into ftttishedproducts, or buildingsor equipmentare
constructed by an enterprise for its own use, the
mtdtiplicityofcostingconventionsthatcan be used,all
within the boundariesof genemllyaccepted accounting principles,make it impossibleto attach a unique
cost to the finishedasset (FASB,1980, para. 65).”
Unavoidable. FASB (1980) acknowledges
that accounting representations may be imprecise (paras. 73-75). Such imprecision is consideredto be either a consequence of choice or
to be unavoidable. An “estimate” may be
appropriate, FASB (1980) states, because an
“accurate measurement” is not necessary. An
analogy is used of an “ordinary wristwatch”
where arterror in timekeeping of a few seconds
a day will usually be acceptable, but not from
“a chronometer used for navigation, scientific
work and the like” (para. 73). Materialityis the
choice criterion: “Fortunately, that is well
understood by accountants. They recognise
that a difference between an estimate and an
accurate measurement may be material in one
context and not materialin another” (para. 74);
“the question that accountants must face continually is how much distortion is acceptable”
(para. 76).
The achievement of representational faithftdness would require the re-presenter to function
merely as a conduit, a neutral someone, albeit
usuallyan expert one, who would record without biases or predilections. But here FASB
(1980) acknowledges that “distortions” occur
but argues that they are “acceptable” as the
“accountant” remains neutral and in control
knowing when it is appropriateto estimate and
when not. This is a retreat from the claim that
accounting alwaysfaithfullyrepresents, but not
from the idea that it is capable of doing so.
Speciftc representations may not have that
property or attribute, but, the deviation is presented as a choice—faithful representation
could have been attained. However, even this
position also is not solidly maintained in FASB
(1980). It also treats “imprecision” and “estimate” as an unavoidable quality of some
accounting representations because the “attributes”, the “transactions” may be inherently
surroundedby “uncertainties” (paras. 72, 94).
Necessarily, such accounting representations
cannot faithftdlyrepresent.
Verificationas consensus. Is it possible to
known that a representation is a faithful one?
For such a condition to be knowable, confirmation should be attainable that a representation is matched with its referent—the
“underlying condition”. But FASB (1980)
accepts that the only assurance that can be
gained from verification is that which relies on
replication of the representation, not one that
can confirm a match between referent and
representation. The purpose of “verification”
FASB(1980) states, is to “provide a significant
degree of assurance that “the accounting measures represent what they purport to represent” (para. 81). But “[verification does not
guarantee the appropriateness of the method
used, much less the correctness of the resulting
measure” (para. 83). Verification merely
“implies consensus” (para. 84), it does not
guarantee that a faithful representation has
occurred:
698 B. McSWEENEY
“Even though several independent measurers may
agree on a singlemeasurementmethod and apply it
honestlyand skilfully, the resultwillnot be reliableif
the method used is such that the measure does not
representwhat it purportsto represent. verifiability
meansno more than that severalmeasuretsarc likely
to obtain the same measure (FASB,1980, paras. 86,
89),”
Verification it seems is not a pure checking of
representation against referent. Even if a representation is a faithfulone, there is, it seems, no
way of knowing with certainty that it is. The
“assurance” that verification brings can only,
FASB(1980) states, be an assuredopinion not a
guarantee that what is represented is a faithful
representation. As UK audit reports state: “in
our opinion the financialstatements give a true
and fair view” (emphasis added).4If verification
cannot be a stronger confirmation than measurer consensus that faithfulrepresentation has
been achieved, how can we know that a representation, verified or unverified, is a faithful
representation? We may decide to believe that
it is, or that it is likelyto be so, or have to act on
the assumption that it is so, but we cannot be
certain. A faithful representation may seem to
be “within our grasp”, but it is never, FASB
(1980) eventuallyacknowledges, quite there.
“Representational faithfulness” commences
in FASB(1980) as a descriptive and prescriptive
term for accounting representations in general,
but the process of explaining that alleged general property and applyingit to specific examples required qualificationsand retreats. We are
left with an aspiration that FASB (1980)
acknowledges is unachievablefor at least some
accounting representations, and is unverifiable
in the sense of comparing it with what is
represented with its representation.
There is no escape it seems from what FASB
(1980) itself calls “human judgment”:
“Thoseunfamiliar withthenatureofaccounting are
often surprisedat the large number of choices that
accountants are requiredto make. Yet, choices arise
at everyturn. . . . There are thosewho seemto barbor
the hope that somewhett waiting to be discovered
there is a comprehensive scoring system that can
providethe universalcriterionfor makingaccounting
choices. Unfortunately,neitherthe Boardnor anyone
else has such a system at the present time.. .
Consequently,those who must choose among alternativesare forced to fall back on human judgement
@ASB,1980, p~. 10).”
Cczsb-jlows. But it could be argued that
such judgement is not required in making all
accounting representations. Perhaps the qualifications and desertions in FASB (1980) from
faithful representation as achievable, and verifiable, was because some accounting representations, but not all, are as FASB(1980) describes
them: surroundedby uncertainty (para. 72).
Some at least of the ambiguityacknowledged
in FASB (1980) might be caused by allocation
“the assignment of a total to one or more locations or categories” which Thomas (1975, p.
314) states: “must alwaysbe incorrigible-that
is to say, they can neither be refuted nor verified.” “Our attempt”, Thomas argues, “to
match costs with revenues must almost always
fail”. But he goes on to claim that: “funds
statement reporting that defines ‘funds’ as net
quick assets”, that is cash or near-cash, are
“allocation free” (Thomas, 1975, p. 318). The
second text to be analysed,below, was selected
because whilst it accepts that matching or
“accrual-based”accounting cannot attain “correspondence” (lee, 1984, pp. 10, 12, 36, 118),5
it claims, like Thomas (1975), that cash-based
accounting can do so. Similarviews about the
nature of cash accounting have been made
elsewhere, for example “the only unambiguous
and objective economic measurement is the
cash paid out in a past transaction. The remainder of economic realitywhich western accourttants have to contend with is to a greater or
lesser degree an abstraction” (Stamp, 1980, p.
79). The cash-flow type of accounting Lee
advances is probably one of the strongest
possible arguments that some accounting
representations at least can be affirmed as
4Self<haracterisationof an auditreport as an “opinion”is not uniqueto the UK and is also the practice in the USAand
elsewhere.
5Lee(1984, fn. 3) attributesrecognitionof “correspondence”as a “fundamentalof accounting”to ch~hem (1966).
THE UNBEARAB LE AMBIGUITYOF ACCOUNTING 699

tion to cash-flow accounting, its main defence
must fall.
T.A. Lee’scasbfiow accounting(Lee, 1984)
Tom Lee describes his book7 as a challenge
to the “conventional wisdoms of financial
accounting”, that is an “accrual-based”
accounting. He explicitly advances “correspondence” as a “key fundamentd” of accounting.
Initially Lee presents “correspondence” not as
an aspired for property of a reformed accounting-which he soon argues for in his book—
but as a current characteristic. “It is. . . important that the surrogate [’accounting information’] corresponds with what would be gained
from direct experience” (p. 11). And that the
“correspondence of information to the economic reality” (which he treats as self-evidently
real and representable) does “not just involve”
its identification. “M also means that the
accounting data are measured according to
generally accepted accounting rules that do
not distort the surrogate approach to direct
experience” (p. 12). But there is movement in
the book. Correspondence as an aspiration
rather than an actuality soon begins to
emerge.
Qual&ication. Immediately following the
description of accounting’s alleged representational capability described in the preceding
paragraph,Lee follows with an example.
“Assume IJ seflsinsurancepolicieson a commission
basis.Hisprofitfor the current periodon a cash basis
is S5 000 (no allowancebeingmade for the depreciation of his car). The latter asset cost S6000 at the
beginningof the period, and has an estimatedusetirf
fifeof six periods,with a negligiblescrap price. It is a
generafly accepted rule of accounting that some
deductionbe made for depreciationwhen determin-
~g P~fit. ThUSit could be arguedthatIJ’s profitfor
the current periodis:
&
TradingProfit 5000
Less:depreciation 1000
Profit 4000
But the question is whether the depreciation
provision aids the correspondence of profit
with the economic reality underlying it. The
above depreciation practice does not contravene generallyaccepted accounting at the present time. But doubts must exist as to whether
or not it reasonably describes the economic
reality (Lee, 1984, pp. 12, 13).”
“Generally accepted accounting” initially
presented as sufficient to ensure recognition of
underlying economic reality is now described
as not necessarily achieving it.
Circularity. Following this qualification, an
alternative “profit statement” is presented substituting “fall in the price of car” (S2 500) for
the “depreciation” in the first statement resuking in a profit calculation of 42500. Additionally, a “cash flow statement” is also provided.
Lee re-labels “trading profit” as “trading surplus” and deducts A%000 for the “purchase
price for car” (instead of the “depreciation” or
“fall in price of car”) giving a calculated “cash
deficit” of S1 000. Lee’s commentary on the
second and third statements is:
“Argumentscan be madethat either of the latter two
statements corresponds better with the underlying
economic realityof the acquisitionandholdingof the
car than does generafly accepted practice, which
represents an arithmeticalallocation of a past cost
datum over the vehicle’s useful life. Whatever the
approach, it is essentialthat the method of measurement adopteddoes not distortthe underlyingreality
beingreported(Lee, 1984, p. 13).”
That accounting is correspondence is here
replaced by can and should be. The aspiration
for correspondence is treated as a description of
what “accounting”, as something fundamental
‘Other terms which have been used in variousliteraturesto try anddescribesirniIarnotionsinclude:tnirmr; picture; copy;
act as a window; mimic; mimeticallyrepresent; imitate;replicate; represence; duplicate;he the same as; tit directly;be
isomorphicwith; be congruentwith; maketransparent.
7Thisbook was chosen as an exemplarof claimsthat cash-flowaccountingis unambiguous.That argumenthas also been
madeby others. The analysisundertakenis of Lee (1984) and is not concerned with its author’sother work.
700 B. Mc~ENN
or essential, is capable of doing: “Whatever the
approach, it is essential that the method of
measurement adopted does not distort the
underlyingreality beihg reported” (p. 13). No
means of determining this aspiration is offered
other than the aspiration itself. Elsewhere in
the book a circukir test is also offered: the
account that corresponds should be chosen—a
choice that can made by choosing the account
which corresponds. For example:
“CERattempts toavoidtime-period allocation ofcosts
thathaveno real-world referent. Onoccasions, this
hasbeentakento implythatCmadvocates aresuggestingthatthesystemisfreefromallallocative judgments.Thisisnotthecase…. Inthesesituations in
Cm, the criterionof correspondence should dominate(Lee, 1984, p. 118).”
Lee’s examples could be subjected to a range
of other considerations, for example, what is
meant by “economic reality”, but in this article
the purpose is to illustrate privilegingand yet
the postponement of correspondence in
accounting representations of an unambiguous:
“underlyingeconomic reality”.
Substitution. Havingtit presented “correspondence” as a property of “generallyaccep
ted accounting” Lee then states that: the
“traditional system of accounting. . based on
cost allocation generally and historic costs particularly” and concludes that it is a:
“man-made financial indicator of considerable variety.basedusuaflyon a mixture of factuaf transactions and subjective accounting judgments. The
latterare unavoidable,but can be minimized.Amajor
problem is findinga systemthat maximisesfact and
minimizesjudgementinorderthatsubjectivitydoesnot
unnecessarilydistottreality(Lee, 1984, pp. 47, 48). ”
Cash flow accounting or cash flow reporting
(CFR), “a complete system of reporting” (p.
123), is adv~ced as the solution. Lee cltis
that using CFR “the accounting characteristic
of correspondence is adhered to, and the
required accounting data are determined without value judgments being imposed by the
producer of the financial statements” (p. 51);
“What CFR provides is a factual basis” (p. 82);
it is “clear, unambiguous and objective” (p.
90); “realised cash-flowsare certain and objective” (p. 109).
Lee’s outline of a “Statement Of Realised
Cash Flow” one of his “four main financial
statements advocatedfor a total CFR system” is
as follows (pp. 53, 60, 96):
&
Casbinflows

x
z=
Lee’s CFR could be criticised for avoiding
many measurement problems by considering
cash flows of a comparatively simple activity
only. Among the issues avoided by this presupposition are: boundaries—chronic problems in defining the boundaries of an entity
whose cash flows are to be identified are
ignored; interdependences—within the
imposed boundaries an oveti view only is
considered so that the possibility of shared
flows (between products, services, segments,
departments, divisions, and so forth) are by
definition excluded and unambiguous aUocation can be assumed; multiple currencies—
flowsare assumedto take place in one currency
only so that judgments about appropriate
exchange rates and methods for conversions or
translations are ignored; non-cash-all transactions are assumedto be cash-based,whilst in an
entity this maynot be so (FASB, 1980, patz. 65).
But even within these restrictions-cash-flow
accounting of a very narrowly defined activity-Lee camot maintain an unqualified claim
of correspondence. There are explicit qualifications both in general, and of specific aspects of
CFR. Descriptions of some of these qualifications and retreats are given below.
Z.Dilution. Accounting as correspondence
changes in the text from achieved or achievable
to not fullyyet (not even in CFR):
“HOUS chaptershaveattemptedto outlinethe
mainfeaturesof, andarguments for, a systemof
external reporting based on a combination of cash
flows and sale prices.. .it would be wrong to imply
THE UNBEARAB LE AMBIGUITYOF ACCOUN’ITNG 701
that such procedures do not involve problems of
accounting and disclosure that require &tber
tbougbt and attention..the CFR system has problemssimilarin terms of generalprincipleto those of
other forms of reporting-that is, measurement, disclosure, usage and Jea.rfbflity(Lee, 1984, p. 117,
emphasisadded).”
CFR is “reasonably objective” (p. 92,
emphasis added); “The measurementprocess is
not entirelyfreefrom personal judgement(and,
arguably,may never be)” (p. 117).
IZ.Acknowledgement. Lee initially, and at
times subsequently, treats “operating cash
flows” and “capital expenditures” as judgement free. They exist for him pre-objectified,
pre-established-separate, autonomous, recognizable and measurable in accounting representations (p. 101). But later he concedes that
classifications may be an imposition rather
than a corresponding. There are acknowledgements, of varying strengths, of constraints, of
unachievability: “A similarproblem arises with
respect to the classification of individual
intlows and outflows, and appearsto have roots
in the long-lived issue of distinguishing
between capital and revenue” (p. 117); “there
are considerable problems of classification in
CFR” (p. 117); “the analyst will be relying on
the reporting accountant’s judgments concerning the classification of expenditure
between capital and expenditure. . for example, expenditure on research and development,
renovation and advertising could be classified
as operational rather than investment cash
outflows (and thus would diminish realized
operating cash flows” (p. 121, emphasis
added); “There are considerableproblems of
classificationof data in CFR.. .Classificationsare
a form of allocation, and require subjective
judgments to be applied” (pp. 117, 18,
emphasis added); “CFR attempts to avoidthese
allocations. It cannot be said to ignore them”
(p. 118).
Here Lee acknowledges that classifyingcash
flows is not a presuppositionless recognition,
but an imposed order. The cash paid for a
dinner, for example, is not inherently, say,
“subsistence”, it could be classified as “entertainment”, or “sales costs”, or “employee
emoluments”, or “research and development”,
and so forth. Payroll taxes could be incorpo
rated in operating cost, or spreadacross various
cost divisions with some, or none, included
in capital expenditure, or Iabelled employee
tax, or included in a more extensive tax paid
figure which might, or might not, also include
local or sales taxes, and so forth.a A further
qualification is revealed when Lee acknowled@ that CFR may require aggregations of
cash inflows and outllows. As discussed
above he had already admitted that the individualcomponents in each aggregationare ftrst
the products “judgements” (p. 121). But he
also concedes that the aggregation itself (to
“net” or “combine”) is also judgmental (pp.
53, 119).
111Replacement. Lee’s CFR requires the
classification of assets as “realized, readily realizable, not-readily realizable, and not realizable.. .In this way the accounting characteristic
of correspondence is adheredto” (p. 51). But it
is suggestedlater in the text that there is more
than one realizable reality: “CFR depends to a
considerable extent on the choice of sale price
by the reporting accountant.. .There are often
many different prices to chose from” (p. 118).
Additionallythere is no consensus even about
the basis of choice:
“Afurther problemiswhether touseasalepricethat
assumes animmediate liquidation andtheexitingof
theassetfromtheentityorasaleinthenormal cow
of trading. ..Thistext takesthe latterviewof sale
pricesbecauseit appears to correspond bestwith
the reatityof whatthe entitydoeswithits assets.
But othersmightarguethatpureexit pricesare
themorefactual presentation ofthatreality (Lee,1984,
p.119).”
CFR thus requires projections about what
might happen (and what has happened).
[ndeed, expectations about what the future
‘AS Ekxu-dieuet al. state: “Measurementand measuringinstruments. . . are so many theories in action [and] . . . every
taxonomyimpliesa theory” (Bourdieuet al., 1991, pp. 37, 46).
702 B, McSWEENEY
might be are acknowledged by Lee as unavoidable: “CFR aims to report on both actual and
potential cash flows.. .CFR should.. contain
statements relating to realizedcash flows. . and
realizable cash flows. . .Unrealized earnings
constitute potential cash flows”, (pp. 50, 62,
emphasis added)..An accounting representation
cannot correspond with what has not yet
occurred. Such representations are not about
the real as conceived in correspondence: nonarbitrary,discrete, re-presentationsof what is in
the here and in the now, but evaluations.
IV Supplemented. Cash flow statements
refer to a specific time period(s). Lee comments
on a particular problem that this presents for
cash flow accounting:
“Anobviousfeatureof reafizedcash flow data is that
they lack the “smoothness”of accmat and allocation
baaed accounting measurements…[this causes] pm
blemsfor the report user ifhisexaminationof realized
cash flows is limited to a series of individualyears
particularlyif an individualyear is intlurmcedsignitlcantlyby the incidenceof a particulartransaction
(lee, 1984, p. 102).”
How does Lee try to reconcile this with correspondence? He addswhat Derrida(1981) has
called a “supplement” (p. 41). Something initially alleged to be complete has to be added to
as it is subsequently conceded as Iacking, as
insufficient. Lee implies that there is a deeper
level of economic reality which can be corresponded with: “the natumltradingcycle” which
can be identifiedby accounting (p. 102).
The “user” is urged to:
“avoid using only twelve monthfyreported realiaed
cash flows,and instead,construct multi-periodaggregatesto providea seriesof cash flowresultsthat is not
atYectedundulyby those of a singleperiod. In doing
so it is arguedthat the user shouldbe reflectingin his
mufti-periodanalysisthe mrural tradingcycle of the
reportingentity.. .Muchwilldependon the reporting
entity’s operating cycle as to what multi-periodis
chosen for analysis(Lee, 1984, p. 102).”
“The natural trading cycle”, Lee states, “is
something which cannot be prescribed” (p.
137). It must depend on the “nature” (p. 103)
of the entity and its operations. It is not knowable through, or from, CFR accounts, but
requires the analyst to know about “the operatingcycle” (pp. 103, 137). The “natural” is to
be known through knowledge of the entity’s
nature. Assessment is again to be achieved
through the circular process of appealingto an
aspiration. The cycles, it is assumed, are real
and repetitive and identifiablefrom aggregated
past cash flow data.But correspondence here is
sub~quently treated as not quite achievable.
Instead of continuity, it is acknowledged that
there may be changes “in the structure of the
entity.. .[there may be] gradual adaptationof
the entity,9 [and]from one form of operation to
another” (pp. 109, 110, emphasis added).10
‘Such changesare not necessarilygradual.Anillustrationof thisis that in 1988, andto a limitedextent earfierin 1987, Reed
InternationalPlc, which was mainlyinvolvedin printing,soldits manufacturingbusinessesfor some41.2 billionand used
the proceeds to relocateitselfin “publishingandinformation”.The percentageof total assetsrepresentedby tangiblefIxed
assets feffdramaticallyand were largelyreplaced by intangiblefixed assets (includingtitles, exhibitionrights, and databases).The legalname remainedthe same,but was it the “same” entity?
1’%Vhensetting accounting standardson cash-flowsregulatorshave had to chose between alternativedefinitionsof “an
amorphousconcept known as ‘cash flow’” (AICPA,1963) (where, for e=mple, does “cash” end ~d a “cash equi~ent”
begin?);to decidewhether, andifso inwhat circumstances,to atlownettingof intlowsandoutflowsof cash, andto define
anddifferentiatebetween,typesof cash-flow(for example,betweendevelopmentandmaintenanceexpenditures).Thatthe
choices made are necessarilyarbitraryis evidencedby a numberof significantditTerencesbetweenthe cash-flowstandards
of, for instance, the FinancialAccounting StandardsBoard (FASB, 1987, 1989), the Accounting SGUI@ Bo~ (AsB~
1995u, 1995b) andthe IntemationatAccountingStandardsBoard(IASC,1992). The contestablequalityof each of the three
types of cash-flowjudgement(definition,aggregation,classification)is indicatedby the historyof the AccountingStandards
Board’scash-flowstandard(ASB,1991). The standardwas tirstissuedin September1991. It replacedan earliersrmiard on
“statementsof source andapplicationof funds”(ASC,1975). InMarch1994 the Boardannouncedthat it proposedto revise
its 1991 standardandissueda draftrevisedstandardinDecember1995 which differssignificantlyfromthe 1991 standardin
respect of each of three judgementtypes (and in other ways) (ASB,1995b). The draftstandardsummarkesthe comments
the AccountingStandardsBoardreceived on the original1991 standard(see ASB,1995a AppendixIII).These comments
also demonstratea diversityof views and the contestabilityof choices ultimatelymade by the regulatorybody, choices
which would not have to be madewere cash-flowa mere “correspondencewith . . . direct experience” (1.-ee,1984).
THE UNBEARABLEAMBIGUITYOF ACCOUNTING 703
IMPLICATIONS
The discussion above of FASB(1980) and Lee
(1984) demonstrated a variety of retreats from
the assertions of correspondence. It exposed
those blind spots or moments of self-contradiction where they involuntarilyreveal the tensions between what they aspire to achieve and
what nonetheless they are constrained to say
(Norris, 1987, p. 19). In those texts unmediated
access to a primary, foundational, representation-free reality through accounting representation was shown to be privileged yet deferred.
The pure grounding of accounting representations in the extra-representationalwas sought,
but never attained. Pure presence, the reflection of pre-existing objectivity, remains postponed it seems. Speakinggenerallyof attempts
to cleanse representation of judgement, Merleau-Ponry states that “[t]he most important
lesson. . reduction teaches us is the impossibility of complete reduction” (Merleau-Ponty,
1962). There are unresolved tensions in the
texts between the initial claim about correspondence: as a condition which accounting
representations are in; what representations
could be—an achievable condition; and a regulative ideal a hypothetical condition, towards
which representations should aspire.
Some implications of the analysisof the two
texts for our understandingof the properties of
accounting representations are now considered.
Accountingrepresentation
Representation, accounting and any other,
can be distinguishedfrom other modes of symbolization through its concern with the natural
and social world in so far as it can be cognitively objectified @dlinikos, 1992). Acco~ting
representations can be separated from, for
example, fictional literature not on the simple
grounds that one is true and the other is not,
but rather that an accounting representation
explicitly seeks to re-present that beyond itself.
The analysisof the texts reveaJed limits in that
capabilitybut it does not imply that there is no
reality outside of representations or “stupidities
of that sort” (Derrida, 1981). As Rorty argues
there is no inference from an assertion that
“one cannot give a theory independent
description of a thing to there are no theory
independent things” (Rorty et al., 1980, p.
279). As Merleau-Pontystates:
“Thesamereasonsthatkeepusfromtreatingperceptionasanobjectalsokeepusfromtreating itasan
operation ofa “subject”, inwhatever senseonetakes
theterm. Mthe “world” upon which it opens, the
ambiguousfield of horizons and distances, is not a
regionof the objectiveworld, it resistsas much being
tied on the side of “fact of consciousness” or
“spiritualacts” (Merleau-Ponty, 1968, p. 23). ”
The argument made here is not that “reality
does not exist independentlyof accounts of it”,
as Hines (1988, p. 258) puts it, but rather that
there is no self-manifestation,no pure re-presentation. Accounting representations do not
naturallyoccur nor, as it were, present themselves. Referents and representations (in so far
as one may use such language) are inexorably
intertwined, indeed indistinguishable.ll
Accounting representations are indeed constructed, but they cannot be reduced to the
social.
This depiction of accounting representations
does not imply that they cannot be influenced
by changes in the extra-representationalsuch as
metaI fatigue, climatic changes, traffic jams, a
chief executive’s illness. But rather it suggests
that the effects do not occur in an unmediated
way. There is no pure flow from the extrarepresentational to representation. Wittgenstein (1953) arguesthis when he writes that we
do not hear “sound” but “motorbikes”. That is,
no matter how “simple” the stimulus, or
“pure” the form, awareness and representation
ll~is Ctimtic Poflmyd of ~pmsentition ~ tti article is consistentwith that advancedelsewhere in the accountfig
literature.AnthonyHopwood speaks of representationsand referents being, as it were, intertwinedor interweavedor
interrneshcd(for example in Hopwood, 1978, 1979, 1984, 1986, 1987, 1990). See also Btdand,1993; Miller& O’Leary,
1987; Christenson,1980.
704 B. McSWEENN
require conceptual processes. Accounting
representation may be disciplinedby the otherness with which it engages, but is not determined by it.
An alternativeapproacb?
Some earlier studies which have also argued
against the possibility of judgement-freerepresentation have additionallyconcluded that the
practice of accounting, and not merely research
into the practice of accounting, would be
better if the criterion were abandoned. John
Nelson, for example, states that
‘,a post-modern practice of accounting
[could]..(re)generste the(pre-modem) narration and
acknowledgement thatlurkin thepoeticoriginsof
thepracticeaswemight nowreinvent them..Apostmodemaccounting willin-formaudiences withnsrrstives, morethancontrol themwithrepresentations.
It will tell stories withmanycharacters andplots
@Ielson, 1993,pp.209,226).”
Ruth Hines suggests that rejecting “representational faithfulness” can “aid us in becoming new beings” and “may help society to
break from outworn vocabulariesand attitudes,
a task of social significance in the post-modem
era of anomie and alienation” (Hines, 1991, p.
328). Peter Manicas states that the “elimination] of false consciousness (or in Derridian
terms, ‘the false grounding of textuality in a
metaphysics of presence)’;.. .is the first and
essential step towards emancipation.. the first
condition of self-liberation”(kkanicas,1993, p.
158). But such headyprogramsdo not necessarily follow from critiques of the possibility of
unambiguousaccounting.
We can distinguish between ambiguity as
that which accounting lives out (which
researchers should, and are, exploring increasingly) and ambiguity as a programme it might
be able to live by. Whilst arguablythe “interpretative act is a universal condition” (Bolartd,
1993), a position this article has supported
through its analysisof the two texts, it does not
necessarily follow that accounting practice
would be “better”, or could even exist, if its
practitioners recognised, or better recognised,
and acted on that understanding.To do so is to
draw an illegitimateinference from a legitimate
thesis. Without assuming a direct causal Hnk
between the richness of a description of a
practice and the quality of the practice’s outcomes (a highly problematic assumption) can
such a universalconsequence be deduced.
In opposition to the advocates of the general
abandonmentof the criterion, there is the view
of Stanley Fish, and others, who speaking of
naturalscience representations, argue that such
rejection would result “only in confusion and a
loss of focus” (Fish, 1994, p. 24). Fish approvingly cites Collins (1992) that: “for science
would not make sense as an institution unless it
were normallythe case that acting scientifically
meant acting as though the sociology of science
were not true”.
These contrary arguments:that the aspiration
for representational faithfulness can (and
should) be generally abandoned, or akernativelythat its continuity is essential, are both
united in assumingthat such a singularbelief is
fundamental to everyday accounting practice.
But this characterisation of representation-making is speculative. At present we do not know
enough about the practices of making accounting representations to be able to make general
judgments about its current single, or mukiple, epistemologicalbasis. There are, as it were,
differentlevels of accounting activity(including
the making of accounting representations) of
which we have inadequate descriptions, and
about whose inter-relationshipswe know little.
We cannot adequately deduce what influence
the impact (current or prospective) epistemo
logical theories have on everyday accounting
fkomreadingnormativeliterature, such as FASB
(1980). That is not to suggest that such literature does not influence, perhaps even significantly, everyday accounting. But different
levels of accounting whose characteristics and
inter-relationships we still know little about
should not be conflated.
It is, perhaps, significant that FASB (1980)
and Lee (1984), examined above, were fully
“objectivist” only when they tried to function
as “philosophers” articulating the general
properties of accounting representations, but
THE UNBEARABLEAMBIGUITYOF ACCOUNTING 705
much less so when trying to practice, to apply
that aspiration to more specific applications
(1-Iopwood,1990). ,1’hereare parallelsherewith
some literature from the natural science disciplines which point to similar differences
between scientific aspirations and descriptions
of science in action (Iaw, 1992; Pickering,
1992; Latour & Woolgar, 1986; Canon, 1986).
This suggests that the makers of accounting
representations may not be as sin@e minded
during that process as either Hines, and others,
have assumed. Rather, it is suggested that we
can as scientists, accountants, or whatever else,
in makingrepresentations have inJames Joyce’s
apt phrase at least “two thinks at a time”; having different “thinks” at different times, or, to
use Derridian language, to think “under erasure”, that is, with a tentativeness and reserve
in keeping with awareness of their judgmental
constitution (Bruner, 1986).
In some practices the relationship between
explanatory and performWce skills varies. It is
not unreasonable to suggest that it is likely to
be closer for accounting practice than it is for
riding a bike, and even closer in formulating
prescriptions for what accounting practice
should be. I am not drawing the spurious distinction between practices informed by theory
and practices innocent of theory, but between
practices in which ji particular theory-tilk, for
instance notions of “representational faithfulness”, seem to be significant, including some
theories about accounting practice, and those,
at the level of everyday accounting, about
which we still know little (1-Iopwood, 1990).
The practice of articulating the conditions of
accounting practice, however well or badly
done, is not the same as the practice of
accounting. The perspective within which
those who do a job, includingmakingaccounting
representations, may not be able to produce a
theoryabout what theypnwtice but nevertheless
they operate on the basis of some (not necessarily consistent) theoretical principles. When
pmctitioners describe the principles that they
believe do, or should, determine practice there
is no ~tee tit this acco~t, even when
made, as it usuallyis in good faith, is an accurate
description of the theory or theories in action.
Foolishnessas an explanation
The advocates of the abandonment of the
aspiration for correspondence assume that not
only should it, but also that it could be, discarded. Some reconcile their rejection of the
possibility of unambiguous accounting representations with their acknowledgement of
ongoing support for it (at least within the
accounting community) by suggesting that the
belief is foolish (see for example, Hines, 1991).
But foolishness is not seen by them as a significant obstacle to fundamental change they
seek as they also presume the societal conception of knowledge is now radicallyin transition
away from “outworn” (Hines, 1991, p. 328)
notions of representational faithfulness.
Appealsare made in abandonment literature to
a particular understanding of “post-modem”
(Hines, 1991, p. 328; Nelson, 1993, p. 209).
The prefix “post” is taken to mean the belief in
absolute objectivity has been, or has almost
been, overcome, surpassed, by most people,
albeit accounting/accountants are treated as
amongst the slowest to adapt. A chronological,
and currently occurring, ~PlaCerneflt of one
theory by a better one is assumed. But this criticism of the ideal of representational faithfulness as outdated relies on an unfounded
periodisation.12
The depiction of abandonment as progress
has relied on a common rhetorical move. It is
arguedthat as the possibilityof representational
faithfulnesshas been largely,if not completely,
rejected in other disciplines it should therefore
be forsaken by accounting. Don Iavoie, for
example, treats “objectivist philosophical
assumptions [as] completely out of step with
modem philosophy of science” (Iavoie, 1987,
IZ~e ~bmd~~ent ~wents ~, fio~cwy, madewithina modernistnotion of progress:the post-modem(or whatever
description) as a replacement of the modern. For criticismsof the claim that the post-modem is a replacement of the
modem see for example Lyotard(1984); Vattbno(1992); Ruccio(19X).
706 B. McSWEENN
pp. 579, 591) and Ruth Hines states that “[i]n
the social sciences, the underminingof realism
has been even more complete [than in the
‘physical sciences’]” (Hines, 1991, p. 317).
Their error has been to treat what is only a factional position in the literature of other disciplines as a consensus view. Using quotations
from, or referencing some of the works of,
emblematic authors opposed to the possibility
of representationalfaithfulness,such asJacques
Derrida, Paul Feyerabend, and Richard Rorty is
not evidence of the degree of support for that
view within, or outside, the academy. References from supporters of representational
faithfulnessare as readilyavailable.Whilst there
has been a considerable growth in accounting,
and other disciplines, of the volume of criticisms of the possibility of unambiguousrepresentations, that does not necessarily indicate
that the notion has even been weakened. In
pamllelwith the growth of anti-representational
criticisms there has also been an increase, in
the quantity of accounting literature which has
argued in support of the possibility of unambiguous representation (see for example, Watts
& Zimmerman, 1978, 1990). Furthermore,
there is some evidence at least that outside the
academy the aspiration for unequivocal
accounting, and other, representations remains
robust and may indeed even be intensifying
(McSweeney, 1994, 1996; Power, 1994).
A regulativeideal
If representational faithfulnessis an unattainable state why is there not methodological and
procedural anarchy?To argue from an analysis
of FASB(1980) and Lee (1984) againstthe possibility of representational faithfulnessis not to
conclude also that all that is left are the vagaries
of random individualviews. Accounting representations are contingent(in the sense of being
the function of multiple, uncertain and potentially disputed variables) rather than subjective
(that is, fundamentallypersonal, and often from
the standpointof others, arbitrary(Smith, 1991;
Porter, 1994). But these variables are usually
limited and often regular: they norrtudlyoccur
within ranges and exhibit patterns. Those ranges and patterns are not the product of the
unclouded perception sought by FASB (1980)
and Lee (1984), but are constitutively connected to some disciplinary or community
vocabulary, or procedures, for representing.
Within the accounting community, views have
sometimes been conspicuously divergent, for
instance over inflation accounting. But even in
the controversies over that issue, every advocate has claimed that their favoured method is
that which achieves representational faithfulness (or equivalent criterion) (Tweedie &
Whittington, 1984; Thompson, 1987). There is
much that is convergent and variations usually
occur within a narrow range (and from a common commitment to, even if disputed achievement of, unambiguous representations). The
producers of accounting representations do not
start each time anew nor in isolation: the
objects they represent and the representational
procedures they employ are always, to some
extent, pre-classifiedand predesigned. There is
a disciplinaryhistoricity, or narrativeaccrual, of
accounting practice. Accounting representations are simultaneously collective and constructed. Although difference has, perhaps,
been disproportionately emphasized in
research, a strikingfact about the techniques of
accounting-representation-makingis that there
is such a high degree of consensus.
I want to argue here that the consensus,
albeit incomplete, is, in part, the product of the
ideal of representational faithfulness. Rather
than being the result of mere foolishness, as
Hinesand others have suggested, the pursuit of
the unattainable is nonetheless real and productive. The ideal of representational faithfulness functions regulatively.13 The argument
here is that without the ideal of “representational faithfulness” there would be excessive
instability, diversity, and randomness, in the
production of accounting representations.
Excessive in the sense that without the albeit
imperfect but constraining effect of the
131~ ~tefil t. one ~fthe ~on~ous reviewersfor alertingme to the notionOf“@atiVe ideal”.
THE UNBEARAB LE AMBIGUITYOF ACCOUNTING 707
regulative ideal, the discipline of accounting
could not exist. The aspiration, whilst not
achievable, nonetheless is conceived within
this explanation as havingan essential function.
Knowledge, unless we assume that only warranted certainty is knowledge, does not require,
and arguably cannot have, certainty (Weintrattb, 1991). But knowledge cluims require
justifications. Justification is a marking out of
those beliefs we consider that we, and others,
ought to hold onto (or adopt) and of those we
should modifyor reject. We largelyaim to have
true beliefs. But justification is the only guide
one has to truth. A known and actual state of
“representational faithfulness” may be t.dtimately elusive, but the desire for it as a fotmdation may be necessary.
The ideal of representational faithfulness
does not separate accounting from other practices which seek to construct themselves as
representational, but it does from those that
seek to be something else: fictional, halhtcinatory, or whatever. Accounting representations
can be distinguished from, for example, fictional literature not on the dichotomous ground
that one is true and the other is not, but rather
that accounting usuallyexplicitly seeks to construct itself as representational. That is not to
argue that texts from practices which do not
construct themselves as representational have
nothing to do with knowledge, but that their
knowledge claims are not advanced as being
representational.
Kant in the CritiqueofJudgement discussed
the role of “regulative”ideas, principles, ideals.
These he stated are mind-made “subjective
principles which are derived, not from the
constitution of an object but from the interest
of reason in respect of a certain possible perfection of the knowledge of the object” (in
Applebaum, 1995, p. 50). His argumentis that a
hypothetical, an imaginary, can be employed
regulativelyand thus become mediatelyrelated
to what is deemed to be true or untrue.Nicholas
Rescher, a pragmatic philosopher, also argues
for the utilityof unrealizableideals.Responding
to claims that ideals which cannot be realised
are pointless he states that this judgement:
“failsto reckon properly with the facts of the situation. For whileidealsput beforeus situationsthat, in a
way, arc mere fictions, these fictions nevertheless
ditect and canatizeour thought and action. .To be
sure, an idealis not a goalwe can expect to attain.But
it serves to set a direction in which we can strive.
Idealsare irrcalities,but they are irrcalitiesthat conditionthe nature of real thought through their intluence on humanthought.. .Ideals,though instruments
of thought are not mere myths. For there is nothing
falseor fictionalaboutidealsas such—ordyaboutthe
idea of their embodimentin concrete reality. Their
pursuitis somethingwhich can be perfectlyreal and
eminentlyproductive(Rescher, 1993, pp. 130, 138).”
A regulative ideal is a normative mechanism
within a communitywhich suppresses or reduces the variability of what is potentially an
open-ended process. It contributes to the
achievement of some degree of conformity and
closure. All human thinkirtg may be based on
certain common activities: they are unavoidably
interpretative @uner, 1986), or constructivist
(Smith, 1991), but disparate disciplines are
guided by different constraints. Within the
accounting community, or communities, the
ideal of “representational faithfulness” may
contribute to maintaininga broad consensus. It
is a way of asserting,at least over a limitedfield,
the unity and the validity of its knowledge.
Other disciplines or communities may have a
different ideal, or ideals, for instance, “reason”
in law, and have different degrees of explicitness and articulation. Even the literature which
has arguedfor abandonment of the criterion of
“representational faithfulness” in accounting
has necessarilyrelied on an alternativeidealas a
basisof knowledgejustification. Hines(1991, p.
328), for example, argues that instead of
“accurate representation” we shouldsee “truth
as ‘what is better for us to believe’”. The criterion of personal utility, she proposes, is certainly different from “accurate representation”,
but it is no less an ideal. Whether the abandonment (if possible) of the hope of “representational faithfulness” (FASB, 1980) or
“correspondence” (Lee, 1984) would result in
“confusion” (Fish, 1994) or alternatively the
eliminationof its “catastrophic effects” (L.avoie,
1987) is not answerable here, but it is
708 B. McSWEEIWW
suggested that what would then exist would be
a radicallydifferent or new discipline.
The ideal of representational faithfulness
does not prevent incursions into “accounting”
by some other disciplines, or expansions out by
“accounting” into some other disciplines,but it
does limit the “becoming what it was not”
(Hopwood, 1987) to that which aspires to be
representational. The ideal is, of course, not a
sufficient condition for making accounting
representations as this achievement also
requires historically developed mechanisms:
rules, procedures, and so forth (Brown et al.,
1993).Indeedit is possible to conceive of such
mechanisms, for example a computer program,
as exclusively capable of making accounting
representations. But I suggest that the ideal
plays an essential role in creation, availability,
and selection of those processes.
Abandonment or analysis?
The failure of FASB(1980) and Lee (1984) to
sustain their descriptions of pure verticality, a
discrete division between dependent representations and a determiningworld suggestsan
active role for accounting in constructing those
objects it appears to merely reflect. The
accounting endeavour, and other techniques
which aspire to standardise and decontextuaIise representations, may be false in the sense
that they cannot evade judgement. But there is
not a direct relationship between truth in some
literal sense, and usefulness. Representations,
accounting and other, are *not disinterested
characterisations of the world but rather are
arguablycreated in order to act on and master it
(Nietzsche, 1968; Heidegger, 1977; Hacking,
1983). Accordingto IanHacking“[w]e represent
in order to intervene, and we intervene in the
lightof representations”(Hacldng, 1983, p. 31).
The literature on accounting representationmaking, is smothered by speculative and normative claims, but, as a practice, it remains a
largely uninvestigated activity. The analysis of
FASB (1980) and Lee (1984) suggests that we
cannot come to understand importan~dimensions of accounting representation-making
within the bounds of belief in the actual
achievement of “representational faithfulness”
but that treating that criterion as a regulative
ideal may help explain its durability. It is not
possible to say with confidence what is the significance of the ideal of “representational faithfulness” within the various types and levels of
accounting representation-making. But the
discussion above suggests that whilst the quest
for “objectivity” is inescapably constrained, it
remains robust and is perhaps unavoidable.
Even in the abandonment literature there is
recognition of at least its current intluence.
Ruth Hines describes “representational faithfulness”, as “the central premise of our society”
(Hines, 1991, p. 327) and John Nelson states
that “late-modern people stay dominated and
mesmerized by it” (Nelson, 1993, p. 208).
Rather than seeking abandonment of the
aspiration for representational faithfulness “by
society” (Hines, 1991, p. 328), and by accounting in particular, an alternative, and it is
suggested more fruitful, activity is that we
should explore more extensively the
constitutive effects on accounting practice at
various levels of that regulative ideal. The failure of concept of “representational faithfulness” to eliminate differences in the
formulation of accounting representation-making regulations (Dopuch & Sunder, 1980;
power, 1989), reinforces the arguments that
the criterion is unattainable by suggesting that
accounting is always constructing the foundation upon which it is claimed to rest, but it does
not weaken the possibility that the idea of a
foundation, though mythical, is nonetheless
constitutive.
BIBLIOGRAPHY
AICPA(1963). AccountingResearchStudyNumber 2, ‘<Cash flow” analysisand thefixndsstatement.
New York: AmericanInstituteof CertifiedPublicAccountants.
THE UNBEARAB LE AMBIGUITYOF ACCOUNTING 709
A,SB(1991). Financi~ ReportingStandardNumber 1, CashjZowstatements.London: AccountingStandardsBoard.
ASB (1992). Financial Reporting Standard Number 3, Repotiing jkzancial pe~ormance. London:
AccountingStandardsBoard.
ASB(1995a). Statementofprfnciplesfor financial reporting.London:AccountingStandardsBoard.
ASB(1995b). FinanciatReportingExposureDraftNumber 10,Revfsion of FRS 1 “casbflow statements”.
London:AccountingStandardsBoard.
ASC(1975). Statementof StandardAccountingPractice Number 10, Statementsofsormce and application offunds. London:AccountingStandardsCommittee.
ASC (1980a). Statementof StandardAccounting Practice Number 18, Accounting for contingencies.
London:AccountingStandardsCommittee.
ASC(1980b). Statementof StandardAccountingPractice Number17, Accountfngforpost balancesheet
events.London:AccountingStandardsCommittee.
Applebaum,D, (1995), ?’bevisfonof Kant. RockportM.: Element.
B~, J. (1987).Doingitbetter: theunitedKin@mnLocaf Authorities Audit Gmunisskm. Australian
Journal of Public AdminWration, 126-130.
Befkaoui,A. (1991), Cultural studies and accounting research. Report of the Cuttural Studies and
AccountingResearchCommittee,Advances in InternationalAccounting.
Boland, R. J. (1989). Beyond the objectivist and subjectivist:leamitrg to read accounting as text,
Accounting, Orgatdzatiorssand Socie@, 14, 591-604.
l)ol~d, R.J. (1993). Accountingand the interpretativeact, Accounting, organizations and So~eW, 18,
125-146.
Bourdieu,P., Chamboredon,J-C., & Passeron,J-C. (1991). ZbeCraJ7of SocioZo~-EpistemologicaZ Preliminaries.Berlin:Walterde Gmyter.
Bromwich, M. (1985). Zbe economics of accounting standard setting, Englewood Cliffs:PrenticeHatl.
Brown, G. A., Coflins,R. & Thornton, D. B. (1993). Professionaljudgement~d accounting s~*rds,
Accounting, Organizationsand Socie&,18,275-289.
Bruner,J. (1986). Actual mind$,possible worids.Cambridge,MA:HarvardUniversityPress.
Canon, M. (1986). Some elements of a sociology of translation:domesticationof the scaflopsand the
fishermenof St. Brieuc Bay. InJ. Law (Ed.) Power, action and belief.a new sociology of knowledge?
London:Routledgeand KegaoPauf.
Chambers, R.J. (1966).Accounting,erxduationandeconomicbebatiour. EnglcwoodCliffs,NJ:Prentice-Hall.
Christenson,C. (1980). Discussionof interaubjectivity-the challengeand opportunityfor accounting,
Accounting, (hganizations and Society,5, 143-146.
Collins,H. M.(1992). Changingorder:replicationand inductionin scientijicpractice.Chicago:Chicago
UniversityPress.
Derrida,J. (1981). Positions.London:AthfonePress.
Dopuch, N., & Sunder,S. (1980). FASB’Sstatementon objectivesand elementsof tinartcialaccounting:a
review. 7beAccountingRevfew, January, 1-21.
FASB(1975). Statement of FinancialAccounting StandardNumber 5, Accounting for contingencies.
StamfordCT: FinancialAccountingStandardsBoard.
FASB(1976). Scope and implicationsof tbe conceptualframework project. StamfordCT: Financial
AccountingStandardsBoard
FASB(1977). Statementof FinancialAccountingStandardNumber16, Prioryear adjustments.Stamford
CT: FinancialAccountingStandardsBoard.
FASB(1980), Statement of FinancialAccounting Concepts Number Z Qualitatfue c~aract~~~ of
accountinginformation.StamfordCT: FinancialAccountingStandardaBoard.
FASB(1980). Statementof FinancialAccountingConcepts Number6, Elementsoffinancial statements.
StamfordCT: FinancialAccountingStandardsBoard.
FASB(1987). Statementof FinancialAccountingStandardNumber95, Statementof casbjlows. Stamford
CT: FinanciatAccountingStandardsBoard.
FASB(1989). Statementof FinancialAccountingStandardNumber 1~, Statement of casb~ows—net
reportingof certaincasb recieptsand cashpayments and ciassijicationof casbjlowsfrom bedging
transactions.StamfordCT: FinanciafAccountingStandardsBoard.
Fish,S. (1994). There’sno sucb tbing afiee speecband it’sa good tbing,too. Oxford: Oxford University
Press,
710 B. McSWRENEY
Flamholtz,E. G. (1980). The process of measurementin managerialaccounting: a psych~teclmical syatems perspective,Accounting,OrganizationsarsdSociety,5, 31-42.
Hacking,L (1983). Re@esentfng and intervening. Cambridge:CambridgeUniversityPress.
Hayes,D. C. (1980). An organizatiomlperspectiveon a psych~tecbnical systemsperspective,Accounting, Organizationsand Soci@, 5, 43-48.
Heidegger,M.(1977). Tbequestionconcerning technologyand otber essays.HarperTorchbooks.
Hines, R. D. (1988). Financialaccounting: in communicatingrealitywe construct reality,Accounting,
Organizationsand Society,13, 251-262
Hines,R. D. (1991). The FASB’Sconceptualframework,financialaccountingandthe maintemuce of the
socialworld,Accounting,Organizationsand Society,16, 313-331.
Hopper, T., Storey,J. & Wilfntott,H. (1987). Accountingfor account@: towardsthe developmentof a
dialecticalview,Accounting, organizationsand Socie~, 12,437-456.
Hopwood,A. G. (1978). Towardsan organizational perspectivefor the studyof accountingand information systems,Accounting,Organizationsand Society,3,3-13.
Hopwood, .4, G. (1979). Criteriaof corporate effectiveness. InM.BrodieandR. 8CMett @s) perspectives
on rnanageriate~fectiueness. ThamesVatleyRegionalManagementCentre.
Hopwood,A.G. (1984). Accountingandthe pursuitof efficiency.InA.G. HopwoodandC. Tomfdns(Eds)
issues inpublic sectoraccounting.London:PhiffipAflan.
Hopwood,A. G. (1986). Accounting and organisationalaction. LondonSchool of Economicsand Universityof ReadingDiscussionPaper in Accountingand FinanceNo. 10,
Hopwood, A. G., The archaeology of accounting systems, Accounting, Organizations and Socie~,
12(1987) 207-234.
Hopwood, A. G. (1990). Ambiguity,knowledgeand tetxitoriaiclaims:some observationson the doctrine
of substanceover form: a reviewessay. ZbeBritish AccountingReview, 79-87.
Hopwood, A. G., & Mifler,P.,(Eds) (1994). Accounting as socialand institutionalpractice.Cambridge:
CambridgeUniversityPress.
IASC(1992). InternationalAccounting StandardNumber 7, Casb Jlow statements,Rev edn. London:
InternationalAccountingStandardsCommittee.
ICAEW(1957). Events occurring after the balance sbeet &te. London: Instituteof CharteredAccountants in EngfandandWales.
minikos, J. (1992). Decontextuafisedaccomplishment:representationiSteCfUIOIOgy. Paperpresentedto
Technologiesof RepresentationConference,Universityof Warwick, 16-18 December.
Latour,B. (1987). The authorresponds:Latourto OMroyd.SocialEpistemology,347-352.
Latour,B., & Woolgar,S.(1986). Laboratoryl~e: tbe constructionof scient!!cfhcts, 2nded. PrittCetonNJ:
PrincetonUniversityPress.
Lzvoie, D. (1987). The accounting of interpretationsand the interpretationsof accounts: the communicativefunctionof the “languageof business”,Accounting, O?ganizatiorrsand Society,12, 579-
604.
bw, J, (1992), Representationand agency. Paper presented to Technologiesof RepresentationConference, Universityof Warwick,December 16-18.
Lee, T. A. (1984). Casbj’i’owaccounting.WokingharrxVanNostmndReinhold.
Lyotard,J. F. (1984). ZBepostmodern condition: a report on knowledge. Manchester:ManchesterUniversityPress.
Macve,R. (1981). A conceptuulframeworkfor~nancial accountingand reporting:tbepossibilitiesfor
an agreedstructure.London:ICAEW.
Manicas,P. (1993). Accountingand the interpretativeart,-4ccounffrsg Management, @Wfriz@Onsad
Society,18, 147-161.
McSweeney,B. (1989). The roles of accountingin organizatiottdSItSirtten~Ce~d chage: insightsfrom
a case-study.In R. Mansfield(Ed.) Frontiers of management researcb”andpractice (pp. 233-246).
London:Routfedge.
McSweeney,B. (1994). Managementby accounting.In A. G. HopwoodandP. Miller(Ecfs)Accounting as
socialand institutionalpractice(pp. 237-269). Cambridge:Cmbri@e Univemiqpress.
McSweeney,B. (1995). Accounting and organizationalaction: a subs~ing exPl~ation or sit~ted
explanations?Accounting,Management and Information Technologies,245-282.
McSweeney,B. (1996). The arrivalof an accountability:explainingthe impositionof managementby
accounting. In R. MunroandJ. Mouritscn(Eds)Accountubilify:power, etbos and tbe technologiesof
managing (pp. 201-224). London:InternationalThomsonBusinessPress.
THE tJNBEARAB LE AMBIGUITYOF ACCOUNTING 711
McSweeney,B., & Steele,A, (1995). RanksHovisMcDougalLInJ. A. Schweikart,S.J. Gray,&C. B. Roberts
(Eds) Internationalaccounting:a caseapproach.New York: McGraw-HiU,
Merleau.Ponty, M,(1962). Pbenomenology ofperception. London:Routledgeand KeganPauf.
Merleau-Ponty, M,(1968). Thevisibleand tbe invisibk, Evanston:NorthwesternUniversityPress.
Miller,P. & O’Leary,T. (1987). Accountingand the constructionof the governableperson,Accounting,
Organizationsand Socfe~, 12, 235-265.
Mifler,p. & CYIXW, T, (1990). Makingaccountancy pmcticaf,Accounting, (hganizatiorrsand Socfev,
15,479-498.
Morgan,G. (1988). Accountingas realityconstruction:towardsa new epistemologyfor accountingpractice, Accounting, Organizationsand Society,13,477-485.
Nelson, J. S. (1993). Account and acknowledge,or represent and control? On post-modempoliticsand
economics of collectiveresponsibility, Accounting,Organizationsand Society,18, 207-299.
Nietzsche,F. (1968), ?fbewill topower. Vintage,
Norris,C. (1987). Dewida. London:FontanaPress,
Peasnell, K. V. (1982). The function of a conceptual framework for corporate financiafreporting.
Accounting and Budness Researcb,243-256.
Pickering,A, (1992). Scienceas apracticeand culture.Chicago:ChicagoUniversityPress,
Porter, T. M. (1994). Objectivityas standardization:the rhetoric of impersonalityin measurement,
statistics,and cost-benefit analysis.In Me@l, A. (Ed.) RetbirskingObjectiuf~. London:DukeUniversity
Press.
power, M. (1989), on the idea of a conceptual frameworkfor financialreporting. Discussionpaper in
AccoundngandFinanceNo. 89/7. London:LondonSchoolof Economics,
Power, M. (1994). The audit society. In A. G. Hopwocd, & P. Mifler(Eds) Accounting czsso~’al and
institutionalpractice.Cambridge:CambridgeUniversityPress.
Mdebau@, L, H, (1973). Environmentalfactors infhsencingthe developmentof aCCOUSStiIIgobjectives,
standardsand practices in Puru. ZbeInternationalJournal of Accounting:Education and Research,
39-56.
Resches,N, (1993), A septum of pragmatic idafism, VOL H:The Validityof VdUeS. PCiIICetOn:pIifICetOII
UniversityPress.
Research Committee of the Instituteof CharteredAccountantsof Scotland(1988). Making corporate
reportsvaluable.London:ICAS/KoganPage.
Rorty,R. (1980). PbiZosopbyand tbe mirror of nature.Oxford:BasifBlackwelL
Ruccio, D. F. (1991). Postmodemismand economics.Journal of Post KeynesfanEconomics,495-510.
Simmonds,K. (1989). Strategicmanagementaccounting:the emergingparadigm.plenafyLecture at the
12th AnnualConferenceof the EuropeanAccountingAssociation,Universityof Stuttgart,5 ApriL
Smith,B. H. (1991). Beliefand resistance:a symmetricalaccount, Crftical1nguiry, 18, 125-139.
Solomons,D. (1989). Gufdelinesfor@nancfalreportjng:apaperpreparedfor tbe researcbboard of the
Instituteof CbarteredAccountantsin Englandand Walesandaddressedto tbeAccountingStandards
Committee. London:ICAEW.
Solomons,D. (1991). Accountingand social change: a neutralistview, Accounting, Organizationsand
Society,16, 287-295.
Stamp,E. (1980). Corporatereporting:ftsfutureevolution:a researcbstudy.Toronto: CanadianInstitute
of CharteredAccountants.
Thomas, A. L. (1969). The allocationproblem in finan~al accounting tbeow. studiesin Accounttig
Resemch Number 3. Sarasota, FL, AmericanAccountingAssociation.
Thomas,A. L. (1975). The FASBandthe auocationfafIacy. Journal of Accountant 65-68.
Thompson, G. (1987). Inflationaccounting in a theory of calculation,Accounting, Organizationsand
Socit?w, 12, 523-543.
Tinker,T. (1988). Panglossiantheories: the science of apologizingin style,Accounting, Otgatdzations
and Society,13,165-189.
Tweedie,D., & WhittingtonG. (1984). TbeinjZationaccountingdebate.Cambridge:CambridgeUniversityPress.
Vattirno,G. (1991). Z5e transparentsociety.Cambtidge:PolityPress.
Watts, R. L., & Zitmnertnan, J. L. (1978). Towardsa positivetheory of determiningaccountingstandards.
7%eAccounting Review.
Watts, R. L., & Zimmerman,J. L. (1990). Positive accounting theory: a ten year perspective. Zbe
Accounting Review, 131-156.
712 B. McSWEENN
Weimraub,R. (1991). Epistemologywithoutknowledge.Ru~o, 157-169.
Wittgenstein,L. (1953). Pbilosopbfculfnvestigatfons.Oxford: IMsflBlackwell.
wood,D.(1991).mea%wnszructfon of time. AtlanticH-ds: HUIIXUU “tiesPress.