Description
As we enter the new millennium, some of us
may be wondering who really will inherit the
earth. Traditionally, the meek have had the inside
track. But if we take Michael Power (1999) ser-
iously, auditors may be the ones who win out in
the end. In the public sector, new models of gov-
ernance and the privatization of government ser-
vices have created an explosive growth in the use
of audit as a mechanism for control.
Will auditors take over the world? Program, technique and
the veri®cation of everything
Brian T. Pentland
School of Labor and Industrial Relations, Michigan State University, 407 South Kedzie Hall, East Lansing, MI 48824-1032, USA
1. Introduction
As we enter the new millennium, some of us
may be wondering who really will inherit the
earth. Traditionally, the meek have had the inside
track. But if we take Michael Power (1999) ser-
iously, auditors may be the ones who win out in
the end. In the public sector, new models of gov-
ernance and the privatization of government ser-
vices have created an explosive growth in the use
of audit as a mechanism for control. Power (1999)
documents this phenomenon in detail, primarily
with examples from the UK. Foreign readers, like
myself, are likely to be a little overwhelmed by the
profusion of unfamiliar acronyms and institutions.
And for some readers, the focus on the UK may
raise questions about the generality of Power's
argument. But I believe the same trend is apparent
anywhere that the public calls for ``accountability''
in government. In the United States, for example,
we have just undertaken our ®rst audit of the
Internal Revenue Service (ironically, the IRS's
own internal controls were found lacking). In the
private sector, quality standards, such as ISO
9000, have created a parallel need for making
work processes auditable. Power argues that we
are experiencing ``the explosion of an idea'' (p. 4)
and a movement along a continuum from a
society that trusts everything and audits nothing
towards a society that trusts nothing and audits
everything. It's not quite a biblical prophecy, but
it is a compelling argument that raises a variety of
questions about the current and future status of
auditing, trust and accountability in modern
society.
In this essay, I have chosen to highlight a few of
the questions that arise from Power's argument.
The ®rst set of questions concerns the boundaries
of auditing as a practice. Phrased narrowly, this
question could be seen as merely de®nitional:
when is an audit really an audit? But the broader
and more provocative question concerns the limits
of applicability of audit, both as a program of
control and a set of techniques. Should everything
be ``made auditable'' and subjected to that pecu-
liar brand of rationalization? A related set of
questions concerns the relationship between var-
ious kinds of audits. There is a family resemblance
among environmental audits, educational audits,
medical audits, energy audits, value for money
audits, and so on, yet they diverge from the pro-
totypical ®nancial audit in a variety of ways. This
observation opens a range of inquiry into the
roots and consequences of the diverse practices
now being called ``auditing.'' Consequences
deserve careful investigation, because auditing is
not just a neutral rendering of the facts. Despite
the ideology of independence, ``making things
auditable'' tends to change the underlying activity
being audited. Auditing clearly has an e?ect, but
what kind of e?ect, and for whom? Power raises
these questions, among others, but argues that the
0361-3682/00/$ - see front matter # 2000 Published by Elsevier Science Ltd. All rights reserved.
PI I : S0361- 3682( 99) 00047- 1
Accounting, Organizations and Society 25 (2000) 307±312
www.elsevier.com/locate/aos
E-mail address: [email protected]
answers may prove elusive because of the ``essen-
tial epistemological obscurity'' of auditing. Audits
are conducted and working papers are written, but
what is really produced? This question turns out
to be the best and hardest of all, so we will save it
for last.
2. The boundaries of auditing
If the idea of auditing really is exploding, and
taking on new forms as it disperses across the
institutional landscape, how can we locate and
recognize the fragments? Power suggests that we
begin looking in places where there is a relation of
accountability between principal and agent (Flint,
1988). In a sense, the explosion of auditing is driven
by an increased desire for accountability. All kinds of
stakeholders are demanding accounts Ð veri®able
accounts Ð that their interests are being upheld.
To the extent that Power is right, auditing can lay
claim to any situation where there is a relationship
of accountability between two parties. This would
seem to represent a new and remarkably permeable
kind of boundary on professional jurisdiction.
Most professions are bounded by expertise in a
particular subject matter. For example, the medi-
cal profession has an apparently natural boundary
on its activities which is de®ned by the human
body and its ailments. The legal profession's
activities are bounded, more or less, by the inter-
ests and obligations created within the legal sys-
tem. But auditing can settle quite naturally in any
situation where there is a relation of account-
ability. The classic principal-agent relationship is
between corporations and their shareholders, but
if doctors are accountable to their patients (or
insurance companies, or government regulators),
and lawyers are accountable to their clients, then
the operating room and the courtroom can also be
subject to audit.
This phenomenon manifests itself in a remark-
able variety of contexts. For example, Llewellyn's
(1998) ®eld study of cost accounting in the social
services provides a detailed example of ``value for
money'' auditing encroaching on social workers'
traditional role as care givers. Demands for
accountability have in¯uenced relations between
law ®rms and their clients, as well. Insurance
companies, for example, are demanding extremely
detailed accounts of attorney billing and insisting
that this information be made available for audit
(Wall Street Journal, January 1999). If any work
process with a client is up for grabs, then the
boundaries of auditing can be drawn quite broadly.
What we have, in a sense, may be a new dimen-
sion of professional con¯ict. Traditionally, sociol-
ogists have conceptualized professional
jurisdiction as an outgrowth of their control of
esoteric knowledge. Abbott's (1988) classic analy-
sis of the boundaries of professions focuses on
occupational groups who are ®ghting over who
should perform certain kinds of work. Control
over substantive knowledge is clearly important,
but the traditional emphasis on content can lead
us to overlook process. For example, Krause
(1996) was analysing the same general trend as
Power (1999) when he argued that traditional
professions like law and medicine are subject to
increasing bureaucratization and ``the advance of
capitalism''. While Krause was writing about the
US context, it is an interesting testimony to the
invisibility of auditors that he failed to note that a
key mechanism of bureaucratic control has been
the increased use of auditing. One explanation
would simply be that audit has been less widely
applied in the US than in the UK. But Power's
argument suggests that auditors exert control pri-
marily over process, not content, and so they enter
the professional battle®eld from a new and unex-
pected direction. Auditors are, in a sense, neutral
parties to the ®ght over who is allowed to write a
prescription (psychologists? nurse practitioners?
midwives?), and they have no legitimate basis for
an opinion about what drug to prescribe. But if
health care providers are to be held accountable
(to whomever), then auditors have important
things to say about how those prescriptions are
written and what records are kept, so that the
process can be made auditable. This is true when-
ever there is a relation of accountability. How then
can we draw the boundaries on this diverse and
growing set of practices?
There are several ways one might attempt to
proceed. One traditional approach would be to
look at the members of the profession: auditing is
308 B.T. Pentland / Accounting, Organizations and Society 25 (2000) 307±312
something that auditors do. The problem, of
course, is that it is increasingly dicult to pin
down who is an auditor. As professional accoun-
tants broaden their sphere of in¯uence, the pro-
blem of encroachment on professional jurisdiction
threatens them, as well. In the realm of environ-
mental auditing, for example, accountants must
compete with scientists and engineers for the right
to de®ne and control the relevant knowledge base
(Power, 1997). In this case, the con¯icts appear to
be over both the process and the content. Energy
audits, which are even less like ®nancial audits
than environmental audits, are usually conducted
by engineers and technicians. Since various kinds
of auditing are being done by individuals who are
not members of the accounting profession, it
seems unwise to de®ne the practice in terms of the
practitioners.
Another approach might be to focus on auditing
practice itself, regardless of the practitioners.
Building on Rose and Miller's (1992) distinction
between program and technique, Power notes that
there are at least two aspects that one might con-
sider. At the programmatic level, auditing can be
seen as a normative image of what an audit ought
to be. At the technical level, auditing can be seen
as a variety of speci®c operations, procedures and
practices. Thus, a technical de®nition might refer
to actual work practices, while a programmatic
de®nition might refer to ideals and goals of the
audit process. The distinction is important because
these dimensions are loosely coupled, at best, so
they can di?use at di?erent rates. The rhetoric of
accountability and veri®cation might spread well
before (or trail behind) actual technical procedures
for auditing. Clearly, if we are to understand what
auditing is, and what it is becoming, we need to do
some investigation.
3. Varieties of auditing
As auditing spreads, it takes on new forms that
resemble the prototypical ®nancial audit in some
respects, but may diverge in others. It is worth
remembering, of course, that ®nancial audit itself
is hardly monolithic. Especially when viewed over
time, or internationally, ®nancial auditing has
varied considerably in terms of both program and
technique (Carpenter & Dirsmith, 1993). Thus,
simply identifying the kinds of audit and compar-
ing them would be a fruitful line of research. It
seems to me that ethnographic studies of auditing
would be required for this undertaking. The ques-
tions at this stage would be primarily descriptive:
What programs are espoused? What techniques
are employed? What do auditors in various kinds
of settings really do? Is there a set of work prac-
tices that can be used to categorize them that
transcends the colloquial labels (tax audit, envir-
onmental audit, energy audit, ®nancial audit,
quality audit, etc.).
While this would be an ambitious project, it has
had some precedent. In response to the explosion
of diverse occupations that shared the label
``technician'', Barley (1996) undertook a series of
ethnographic studies to discover what commonal-
ities they had, if any. Barley used data on techni-
cians' work practices to generalize beyond the
traditional occupational titles, such as lab techni-
cian, computer technician, emergency medical
technician, radiological technician, automobile
technicians, and so on. Barley asked two very
simple questions: what do technicians do and what
do they know? He worked inductively, from eth-
nographic data, to create an ideal-typical ``techni-
cian'' that transcended the traditional job
descriptions. In a sense, current thinking about
auditing is rooted in a similar set of labels, but we
are so heavily in¯uenced by the familiar arche-
typal ®nancial audit that we may be blind to the
growing variety. If so, then a similar undertaking
may be desirable for the broad and growing cate-
gory of activities we loosely refer to as ``auditing''.
The point is to cast our net widely and look at all
the examples of people engaged in veri®cation Ð
what are the commonalities and the di?erences?
What do they do? What do they know? And given
the programmatic aspect of auditing, we might
also ask: What ideals do they espouse?
Barley's (1996) analysis of technical work is
suggestive of the kinds of thing we might ®nd if we
pursued this line of inquiry. He derived two main
categories of technician: ``bu?ers'' and ``brokers''.
When technicians act as bu?ers, they mediate
between the material world (e.g. cells growing in a
B.T. Pentland / Accounting, Organizations and Society 25 (2000) 307±312 309
Petri dish) and the symbolic world (e.g. an inter-
pretation of what those cells mean). When techni-
cians act as brokers, they facilitate access to some
aspect of the material world (e.g., connecting your
personal computer to a network). To the extent
that auditors are technicians, they seem to act
more as bu?ers than brokers. They provide an
interpretation of accounting data, so that others
do not need to bother with the details. Some kinds
of auditors, such as environmental auditors, may
be closer to Barley's (1996) ideal-typical techni-
cian-as-bu?er: they interpret some aspect of the
material world. But most auditors mediate
between two symbolic worlds: a set of accounts
and an interpretation of those accounts. This may
contribute to what Power (1999, p. 28) refers to as
the ``deep epistemological obscurity'' of auditing,
a point we will return to in a moment.
4. E?ects of auditing
As the idea of auditing explodes, it is not just
that there is more auditing going on. The logic of
auditing can change the way we do things, because
an activity must be ``made auditable'' before an
audit can be conducted. Power (1997) argues that
as the use of auditing intensi®es, there are two
broad kinds of e?ects that one might predict:
colonization and decoupling. Colonization occurs
when an organization internalizes the values of the
audit process: ``when the values and practices that
make auditing possible penetrate deep into the
core of organizational operations'' (p.97). The
work becomes truly auditable, but the result may
be ``e?ective in unintended ways'' (p. 13). At the
other extreme, decoupling occurs when an audit
process is disconnected from what is really going
on. In this situation, audits are rendered ine?ec-
tive because they are reduced to ``rationalized
rituals of inspection'' (p. 96). The paradoxical
result is that more auditing may lead to less ver-
i®cation. In my research on tax auditing, for
example, there was a fear that audits could turn
good taxpayers into bad ones, and that more
intensive auditing might simply push taxpayers to
be more secretive (Pentland & Carlile, 1996). It
seems likely that whenever the relationship
between the principal and agent has an adversarial
aspect, auditing could become self-defeating.
To further analyze the dynamic e?ects of audit-
ing, it may be useful to adopt an organizational
learning perspective. Super®cially, auditing and
learning would seem to be complementary, since
they share an emphasis on gathering facts and
providing feedback. But if auditing can lead to
colonization or decoupling, the story is not quite
so simple. Furthermore, as the preceeding discus-
sion suggests, it is important to consider di?er-
ences between subsystems, each of which may
have di?erent goals and roles in the audit process
(principal, agent and auditor). Extreme decou-
pling obviously inhibits learning for the system as
a whole because information is hidden. Extreme
colonization may be just as dysfunctional, and
harder to diagnose and correct. As organizational
members strive to conform to the requirements of
auditability, they may tend to narrow their per-
spective. In the worst case, they may become more
concerned with generating the right indicators
than with actually doing a good job. To return to
the medical example, an auditable process does
not necessarily result in a more accurate diagnosis
or a better prescription. The distinction between
process and content is often blurry, but to the
extent that auditing can a?ect one without a?ect-
ing the other, attempts at increased control and
accountability may or may not provide straight-
forward ``improvements''. As a result, it is dicult
to say whether more extensive auditing will be
good or bad, and for whom.
5. Epistemological obscurity of auditing
The preceding discussion begins to hint at a
more fundamental issue that Power (1999) calls
the ``deep epistemological obscurity'' of auditing
(p. 28): ``What is the nature of assurance given by
audits? Can it be observed?'' This question is suf-
®ciently subtle and provocative that it may never
really be answered, but I think there some good
reasons why it needs to be asked.
Like many occupations, auditors adopt the
rhetoric and, to some extent, the routines of sci-
ence. For example, auditors take ``samples'' and
310 B.T. Pentland / Accounting, Organizations and Society 25 (2000) 307±312
perform ``tests'' to reach ``objective'' conclusions.
This would be a pretty good description of a
technician in a biological research lab, as well. As
Carpenter and Dirsmith (1993) argue, the rhetoric
of science (e.g, ``statistical sampling'') and the
analogy to scienti®c practice is a powerful legit-
imating device for practitioners. It seems to
permeate much of the academic research on
auditing, as well. Put simply, the analogy suggests
that while scientists illuminate natural truths,
auditors illuminate ®nancial truths. And if Power
is right, auditors may soon be illuminating nearly
every other kind of truth, as well.
The analogy between auditors and lab techni-
cians breaks down quite rapidly, however. In an
audit, the samples, tests, and interpretations are
all highly contextualized (conditioned strongly on
last year's numbers, among other things). If the
date and client name were removed from a typical
audit working paper or accounting report, for
example, it would be transformed from a working
paper into scrap paper (and vice versa) (Pentland,
1993). And unlike a laboratory, there is never a
control group. Each audit is a kind of uncon-
trolled experiment, and there is no way to know
what would have happened if an audit were not
performed. No wonder that audits are epistemo-
logically obscure Ð auditors have adopted the
rhetoric of scienti®c methodology without really
being able to adopt much of the substance.
A more profound diculty is that, unlike lab
technicians, auditors generally do not mediate
between the material world and the symbolic
world. Rather, with some exceptions, auditors act
as bu?ers between two symbolic worlds. They
interpret accounting systems, which are them-
selves interpretative products, and they do so by
following a variety of rules which are also open to
interpretation. While Power attributes the ``expec-
tations gap'' to the loose coupling between the
programmatic idealization of auditing and the
actual technical procedures, the analogy to tech-
nical work reveals a potentially much larger gap.
The rhetoric and procedures of auditing imply an
analogy to scienti®c practice, but the actual prac-
tice has only super®cial similarities. We are led to
expect empirical science, when the best we can
hope for is hermeneutics.
If so, then it may be equally illuminating to
compare auditors to movie or television critics.
They also act as bu?ers between a lay public and a
symbolic world. They also use evidence to create
opinions and issue reports so that we, the viewing
public, can allocate our viewing time wisely. This
comparison may strike some readers as ¯ippant, but
it is not. In the US, for example, recent dramatic epi-
sodes of violence by teenagers have led to increased
concern about the portrayal of violence in the
media. There is a growing movement to hold televi-
sion and movie producers accountable to the view-
ing public and society at large. Is that not exactly the
kind of situation where we should expect auditing
to thrive? Film criticism might never be called
``auditing,'' but the analogy is certainly provocative.
6. Some possible directions
Given the enormous breadth of possibilities
suggested by Power's (1999) book, it seems rea-
sonable to suggest some directions one might take
in empirical research. These questions deserve
attention because, as Power points out, auditing
occupies a much more signi®cant role in modetn
society than has previously been recognized. First,
as I have suggested already, there is a need for
description and comparison. What kinds of prac-
tices are emerging under the general category of
``audit'' and who is performing this work? I
believe that inductive, ®eld-based studies of audit
practice are likely to reveal some interesting com-
monalities and di?erences. And from a historical
perspective, if Power is right about the trend, it
would be valuable to document the state of a?airs
now, so that future scholars might have a baseline
for comparison. Second, there is a need to assess
the e?ects of these diverse practices. For example,
one might investigate the conditions under which
colonization or decoupling are most likely. The
key point here, is that we ought to investigate
e?ects, understood broadly, not just ``e?ective-
ness''. Finally, because Power's work focuses on
the UK, it suggests a fruitful opportunity for
comparative research. Given the diversity of audit
practices that seem to emerging, it would be an
enormous challenge to develop valid and reliable
B.T. Pentland / Accounting, Organizations and Society 25 (2000) 307±312 311
quantitative indicators of audit activity that could
be used across national borders. Alternatively, one
might approach the problem as a set of case stu-
dies. Regardless of the speci®c direction one
chooses to take, the kinds of research suggested by
Power's book should provide a great deal of
insight into a set of practices that have generally
been taken for granted.
References
Abbott, A. (1988). The system of professions an essay on the
division of expert labor. Chicago: Chicago University Press.
Barley, S. R. (1996). Technicians in the workplace: ethno-
graphic evidence for bringing work into organization studies.
Administrative Science Quarterly, 41, 404±441.
Carpenter, B., & Dirsmith, M. (1993). Sampling and the
abstraction of knowledge in the auditing profession: an
extended institutional theory perspective. Accounting, Orga-
nization and Society, 18(1), 41±63.
Flint, D. (1988). Philosophy and principles of auditing. London:
McMillan.
Krause (1996). Death of the guilds: professions, states, and the
advance of capitalism, 1930 to the present. New Haven: Yale
University Press.
Llewellyn, S. (1998). Boundary work: costing and caring in the
social services. Accounting, Organizations and Society, 23,
23±47.
Pentland, B. T., & Carlile, P. (1996). Audit the taxpayer, not
the return: tax auditing as an expression game. Accounting,
Organizations and Society, 21(2/3), 269±287.
Pentland, B. T. (1993). Getting comfortable with the numbers:
auditing and the micro-production of macro-order. Account-
ing, Organizations and Society, 18(7), 605±620.
Power, M. (1999). The audit society: rituals of veri®cation (2nd
Ed.). Oxford: Oxford University Press.
Power, M. (1997). Expertise and the construction of relevance:
accountants and the environmental audit. Accounting, Orga-
nizations and Society, 22, 123±146.
Rose, N., & Miller, P. (1997). Political power beyond the state:
Problematics of government. British Journal of Sociology,
43(2), 173±205.
312 B.T. Pentland / Accounting, Organizations and Society 25 (2000) 307±312
doc_745512272.pdf
As we enter the new millennium, some of us
may be wondering who really will inherit the
earth. Traditionally, the meek have had the inside
track. But if we take Michael Power (1999) ser-
iously, auditors may be the ones who win out in
the end. In the public sector, new models of gov-
ernance and the privatization of government ser-
vices have created an explosive growth in the use
of audit as a mechanism for control.
Will auditors take over the world? Program, technique and
the veri®cation of everything
Brian T. Pentland
School of Labor and Industrial Relations, Michigan State University, 407 South Kedzie Hall, East Lansing, MI 48824-1032, USA
1. Introduction
As we enter the new millennium, some of us
may be wondering who really will inherit the
earth. Traditionally, the meek have had the inside
track. But if we take Michael Power (1999) ser-
iously, auditors may be the ones who win out in
the end. In the public sector, new models of gov-
ernance and the privatization of government ser-
vices have created an explosive growth in the use
of audit as a mechanism for control. Power (1999)
documents this phenomenon in detail, primarily
with examples from the UK. Foreign readers, like
myself, are likely to be a little overwhelmed by the
profusion of unfamiliar acronyms and institutions.
And for some readers, the focus on the UK may
raise questions about the generality of Power's
argument. But I believe the same trend is apparent
anywhere that the public calls for ``accountability''
in government. In the United States, for example,
we have just undertaken our ®rst audit of the
Internal Revenue Service (ironically, the IRS's
own internal controls were found lacking). In the
private sector, quality standards, such as ISO
9000, have created a parallel need for making
work processes auditable. Power argues that we
are experiencing ``the explosion of an idea'' (p. 4)
and a movement along a continuum from a
society that trusts everything and audits nothing
towards a society that trusts nothing and audits
everything. It's not quite a biblical prophecy, but
it is a compelling argument that raises a variety of
questions about the current and future status of
auditing, trust and accountability in modern
society.
In this essay, I have chosen to highlight a few of
the questions that arise from Power's argument.
The ®rst set of questions concerns the boundaries
of auditing as a practice. Phrased narrowly, this
question could be seen as merely de®nitional:
when is an audit really an audit? But the broader
and more provocative question concerns the limits
of applicability of audit, both as a program of
control and a set of techniques. Should everything
be ``made auditable'' and subjected to that pecu-
liar brand of rationalization? A related set of
questions concerns the relationship between var-
ious kinds of audits. There is a family resemblance
among environmental audits, educational audits,
medical audits, energy audits, value for money
audits, and so on, yet they diverge from the pro-
totypical ®nancial audit in a variety of ways. This
observation opens a range of inquiry into the
roots and consequences of the diverse practices
now being called ``auditing.'' Consequences
deserve careful investigation, because auditing is
not just a neutral rendering of the facts. Despite
the ideology of independence, ``making things
auditable'' tends to change the underlying activity
being audited. Auditing clearly has an e?ect, but
what kind of e?ect, and for whom? Power raises
these questions, among others, but argues that the
0361-3682/00/$ - see front matter # 2000 Published by Elsevier Science Ltd. All rights reserved.
PI I : S0361- 3682( 99) 00047- 1
Accounting, Organizations and Society 25 (2000) 307±312
www.elsevier.com/locate/aos
E-mail address: [email protected]
answers may prove elusive because of the ``essen-
tial epistemological obscurity'' of auditing. Audits
are conducted and working papers are written, but
what is really produced? This question turns out
to be the best and hardest of all, so we will save it
for last.
2. The boundaries of auditing
If the idea of auditing really is exploding, and
taking on new forms as it disperses across the
institutional landscape, how can we locate and
recognize the fragments? Power suggests that we
begin looking in places where there is a relation of
accountability between principal and agent (Flint,
1988). In a sense, the explosion of auditing is driven
by an increased desire for accountability. All kinds of
stakeholders are demanding accounts Ð veri®able
accounts Ð that their interests are being upheld.
To the extent that Power is right, auditing can lay
claim to any situation where there is a relationship
of accountability between two parties. This would
seem to represent a new and remarkably permeable
kind of boundary on professional jurisdiction.
Most professions are bounded by expertise in a
particular subject matter. For example, the medi-
cal profession has an apparently natural boundary
on its activities which is de®ned by the human
body and its ailments. The legal profession's
activities are bounded, more or less, by the inter-
ests and obligations created within the legal sys-
tem. But auditing can settle quite naturally in any
situation where there is a relation of account-
ability. The classic principal-agent relationship is
between corporations and their shareholders, but
if doctors are accountable to their patients (or
insurance companies, or government regulators),
and lawyers are accountable to their clients, then
the operating room and the courtroom can also be
subject to audit.
This phenomenon manifests itself in a remark-
able variety of contexts. For example, Llewellyn's
(1998) ®eld study of cost accounting in the social
services provides a detailed example of ``value for
money'' auditing encroaching on social workers'
traditional role as care givers. Demands for
accountability have in¯uenced relations between
law ®rms and their clients, as well. Insurance
companies, for example, are demanding extremely
detailed accounts of attorney billing and insisting
that this information be made available for audit
(Wall Street Journal, January 1999). If any work
process with a client is up for grabs, then the
boundaries of auditing can be drawn quite broadly.
What we have, in a sense, may be a new dimen-
sion of professional con¯ict. Traditionally, sociol-
ogists have conceptualized professional
jurisdiction as an outgrowth of their control of
esoteric knowledge. Abbott's (1988) classic analy-
sis of the boundaries of professions focuses on
occupational groups who are ®ghting over who
should perform certain kinds of work. Control
over substantive knowledge is clearly important,
but the traditional emphasis on content can lead
us to overlook process. For example, Krause
(1996) was analysing the same general trend as
Power (1999) when he argued that traditional
professions like law and medicine are subject to
increasing bureaucratization and ``the advance of
capitalism''. While Krause was writing about the
US context, it is an interesting testimony to the
invisibility of auditors that he failed to note that a
key mechanism of bureaucratic control has been
the increased use of auditing. One explanation
would simply be that audit has been less widely
applied in the US than in the UK. But Power's
argument suggests that auditors exert control pri-
marily over process, not content, and so they enter
the professional battle®eld from a new and unex-
pected direction. Auditors are, in a sense, neutral
parties to the ®ght over who is allowed to write a
prescription (psychologists? nurse practitioners?
midwives?), and they have no legitimate basis for
an opinion about what drug to prescribe. But if
health care providers are to be held accountable
(to whomever), then auditors have important
things to say about how those prescriptions are
written and what records are kept, so that the
process can be made auditable. This is true when-
ever there is a relation of accountability. How then
can we draw the boundaries on this diverse and
growing set of practices?
There are several ways one might attempt to
proceed. One traditional approach would be to
look at the members of the profession: auditing is
308 B.T. Pentland / Accounting, Organizations and Society 25 (2000) 307±312
something that auditors do. The problem, of
course, is that it is increasingly dicult to pin
down who is an auditor. As professional accoun-
tants broaden their sphere of in¯uence, the pro-
blem of encroachment on professional jurisdiction
threatens them, as well. In the realm of environ-
mental auditing, for example, accountants must
compete with scientists and engineers for the right
to de®ne and control the relevant knowledge base
(Power, 1997). In this case, the con¯icts appear to
be over both the process and the content. Energy
audits, which are even less like ®nancial audits
than environmental audits, are usually conducted
by engineers and technicians. Since various kinds
of auditing are being done by individuals who are
not members of the accounting profession, it
seems unwise to de®ne the practice in terms of the
practitioners.
Another approach might be to focus on auditing
practice itself, regardless of the practitioners.
Building on Rose and Miller's (1992) distinction
between program and technique, Power notes that
there are at least two aspects that one might con-
sider. At the programmatic level, auditing can be
seen as a normative image of what an audit ought
to be. At the technical level, auditing can be seen
as a variety of speci®c operations, procedures and
practices. Thus, a technical de®nition might refer
to actual work practices, while a programmatic
de®nition might refer to ideals and goals of the
audit process. The distinction is important because
these dimensions are loosely coupled, at best, so
they can di?use at di?erent rates. The rhetoric of
accountability and veri®cation might spread well
before (or trail behind) actual technical procedures
for auditing. Clearly, if we are to understand what
auditing is, and what it is becoming, we need to do
some investigation.
3. Varieties of auditing
As auditing spreads, it takes on new forms that
resemble the prototypical ®nancial audit in some
respects, but may diverge in others. It is worth
remembering, of course, that ®nancial audit itself
is hardly monolithic. Especially when viewed over
time, or internationally, ®nancial auditing has
varied considerably in terms of both program and
technique (Carpenter & Dirsmith, 1993). Thus,
simply identifying the kinds of audit and compar-
ing them would be a fruitful line of research. It
seems to me that ethnographic studies of auditing
would be required for this undertaking. The ques-
tions at this stage would be primarily descriptive:
What programs are espoused? What techniques
are employed? What do auditors in various kinds
of settings really do? Is there a set of work prac-
tices that can be used to categorize them that
transcends the colloquial labels (tax audit, envir-
onmental audit, energy audit, ®nancial audit,
quality audit, etc.).
While this would be an ambitious project, it has
had some precedent. In response to the explosion
of diverse occupations that shared the label
``technician'', Barley (1996) undertook a series of
ethnographic studies to discover what commonal-
ities they had, if any. Barley used data on techni-
cians' work practices to generalize beyond the
traditional occupational titles, such as lab techni-
cian, computer technician, emergency medical
technician, radiological technician, automobile
technicians, and so on. Barley asked two very
simple questions: what do technicians do and what
do they know? He worked inductively, from eth-
nographic data, to create an ideal-typical ``techni-
cian'' that transcended the traditional job
descriptions. In a sense, current thinking about
auditing is rooted in a similar set of labels, but we
are so heavily in¯uenced by the familiar arche-
typal ®nancial audit that we may be blind to the
growing variety. If so, then a similar undertaking
may be desirable for the broad and growing cate-
gory of activities we loosely refer to as ``auditing''.
The point is to cast our net widely and look at all
the examples of people engaged in veri®cation Ð
what are the commonalities and the di?erences?
What do they do? What do they know? And given
the programmatic aspect of auditing, we might
also ask: What ideals do they espouse?
Barley's (1996) analysis of technical work is
suggestive of the kinds of thing we might ®nd if we
pursued this line of inquiry. He derived two main
categories of technician: ``bu?ers'' and ``brokers''.
When technicians act as bu?ers, they mediate
between the material world (e.g. cells growing in a
B.T. Pentland / Accounting, Organizations and Society 25 (2000) 307±312 309
Petri dish) and the symbolic world (e.g. an inter-
pretation of what those cells mean). When techni-
cians act as brokers, they facilitate access to some
aspect of the material world (e.g., connecting your
personal computer to a network). To the extent
that auditors are technicians, they seem to act
more as bu?ers than brokers. They provide an
interpretation of accounting data, so that others
do not need to bother with the details. Some kinds
of auditors, such as environmental auditors, may
be closer to Barley's (1996) ideal-typical techni-
cian-as-bu?er: they interpret some aspect of the
material world. But most auditors mediate
between two symbolic worlds: a set of accounts
and an interpretation of those accounts. This may
contribute to what Power (1999, p. 28) refers to as
the ``deep epistemological obscurity'' of auditing,
a point we will return to in a moment.
4. E?ects of auditing
As the idea of auditing explodes, it is not just
that there is more auditing going on. The logic of
auditing can change the way we do things, because
an activity must be ``made auditable'' before an
audit can be conducted. Power (1997) argues that
as the use of auditing intensi®es, there are two
broad kinds of e?ects that one might predict:
colonization and decoupling. Colonization occurs
when an organization internalizes the values of the
audit process: ``when the values and practices that
make auditing possible penetrate deep into the
core of organizational operations'' (p.97). The
work becomes truly auditable, but the result may
be ``e?ective in unintended ways'' (p. 13). At the
other extreme, decoupling occurs when an audit
process is disconnected from what is really going
on. In this situation, audits are rendered ine?ec-
tive because they are reduced to ``rationalized
rituals of inspection'' (p. 96). The paradoxical
result is that more auditing may lead to less ver-
i®cation. In my research on tax auditing, for
example, there was a fear that audits could turn
good taxpayers into bad ones, and that more
intensive auditing might simply push taxpayers to
be more secretive (Pentland & Carlile, 1996). It
seems likely that whenever the relationship
between the principal and agent has an adversarial
aspect, auditing could become self-defeating.
To further analyze the dynamic e?ects of audit-
ing, it may be useful to adopt an organizational
learning perspective. Super®cially, auditing and
learning would seem to be complementary, since
they share an emphasis on gathering facts and
providing feedback. But if auditing can lead to
colonization or decoupling, the story is not quite
so simple. Furthermore, as the preceeding discus-
sion suggests, it is important to consider di?er-
ences between subsystems, each of which may
have di?erent goals and roles in the audit process
(principal, agent and auditor). Extreme decou-
pling obviously inhibits learning for the system as
a whole because information is hidden. Extreme
colonization may be just as dysfunctional, and
harder to diagnose and correct. As organizational
members strive to conform to the requirements of
auditability, they may tend to narrow their per-
spective. In the worst case, they may become more
concerned with generating the right indicators
than with actually doing a good job. To return to
the medical example, an auditable process does
not necessarily result in a more accurate diagnosis
or a better prescription. The distinction between
process and content is often blurry, but to the
extent that auditing can a?ect one without a?ect-
ing the other, attempts at increased control and
accountability may or may not provide straight-
forward ``improvements''. As a result, it is dicult
to say whether more extensive auditing will be
good or bad, and for whom.
5. Epistemological obscurity of auditing
The preceding discussion begins to hint at a
more fundamental issue that Power (1999) calls
the ``deep epistemological obscurity'' of auditing
(p. 28): ``What is the nature of assurance given by
audits? Can it be observed?'' This question is suf-
®ciently subtle and provocative that it may never
really be answered, but I think there some good
reasons why it needs to be asked.
Like many occupations, auditors adopt the
rhetoric and, to some extent, the routines of sci-
ence. For example, auditors take ``samples'' and
310 B.T. Pentland / Accounting, Organizations and Society 25 (2000) 307±312
perform ``tests'' to reach ``objective'' conclusions.
This would be a pretty good description of a
technician in a biological research lab, as well. As
Carpenter and Dirsmith (1993) argue, the rhetoric
of science (e.g, ``statistical sampling'') and the
analogy to scienti®c practice is a powerful legit-
imating device for practitioners. It seems to
permeate much of the academic research on
auditing, as well. Put simply, the analogy suggests
that while scientists illuminate natural truths,
auditors illuminate ®nancial truths. And if Power
is right, auditors may soon be illuminating nearly
every other kind of truth, as well.
The analogy between auditors and lab techni-
cians breaks down quite rapidly, however. In an
audit, the samples, tests, and interpretations are
all highly contextualized (conditioned strongly on
last year's numbers, among other things). If the
date and client name were removed from a typical
audit working paper or accounting report, for
example, it would be transformed from a working
paper into scrap paper (and vice versa) (Pentland,
1993). And unlike a laboratory, there is never a
control group. Each audit is a kind of uncon-
trolled experiment, and there is no way to know
what would have happened if an audit were not
performed. No wonder that audits are epistemo-
logically obscure Ð auditors have adopted the
rhetoric of scienti®c methodology without really
being able to adopt much of the substance.
A more profound diculty is that, unlike lab
technicians, auditors generally do not mediate
between the material world and the symbolic
world. Rather, with some exceptions, auditors act
as bu?ers between two symbolic worlds. They
interpret accounting systems, which are them-
selves interpretative products, and they do so by
following a variety of rules which are also open to
interpretation. While Power attributes the ``expec-
tations gap'' to the loose coupling between the
programmatic idealization of auditing and the
actual technical procedures, the analogy to tech-
nical work reveals a potentially much larger gap.
The rhetoric and procedures of auditing imply an
analogy to scienti®c practice, but the actual prac-
tice has only super®cial similarities. We are led to
expect empirical science, when the best we can
hope for is hermeneutics.
If so, then it may be equally illuminating to
compare auditors to movie or television critics.
They also act as bu?ers between a lay public and a
symbolic world. They also use evidence to create
opinions and issue reports so that we, the viewing
public, can allocate our viewing time wisely. This
comparison may strike some readers as ¯ippant, but
it is not. In the US, for example, recent dramatic epi-
sodes of violence by teenagers have led to increased
concern about the portrayal of violence in the
media. There is a growing movement to hold televi-
sion and movie producers accountable to the view-
ing public and society at large. Is that not exactly the
kind of situation where we should expect auditing
to thrive? Film criticism might never be called
``auditing,'' but the analogy is certainly provocative.
6. Some possible directions
Given the enormous breadth of possibilities
suggested by Power's (1999) book, it seems rea-
sonable to suggest some directions one might take
in empirical research. These questions deserve
attention because, as Power points out, auditing
occupies a much more signi®cant role in modetn
society than has previously been recognized. First,
as I have suggested already, there is a need for
description and comparison. What kinds of prac-
tices are emerging under the general category of
``audit'' and who is performing this work? I
believe that inductive, ®eld-based studies of audit
practice are likely to reveal some interesting com-
monalities and di?erences. And from a historical
perspective, if Power is right about the trend, it
would be valuable to document the state of a?airs
now, so that future scholars might have a baseline
for comparison. Second, there is a need to assess
the e?ects of these diverse practices. For example,
one might investigate the conditions under which
colonization or decoupling are most likely. The
key point here, is that we ought to investigate
e?ects, understood broadly, not just ``e?ective-
ness''. Finally, because Power's work focuses on
the UK, it suggests a fruitful opportunity for
comparative research. Given the diversity of audit
practices that seem to emerging, it would be an
enormous challenge to develop valid and reliable
B.T. Pentland / Accounting, Organizations and Society 25 (2000) 307±312 311
quantitative indicators of audit activity that could
be used across national borders. Alternatively, one
might approach the problem as a set of case stu-
dies. Regardless of the speci®c direction one
chooses to take, the kinds of research suggested by
Power's book should provide a great deal of
insight into a set of practices that have generally
been taken for granted.
References
Abbott, A. (1988). The system of professions an essay on the
division of expert labor. Chicago: Chicago University Press.
Barley, S. R. (1996). Technicians in the workplace: ethno-
graphic evidence for bringing work into organization studies.
Administrative Science Quarterly, 41, 404±441.
Carpenter, B., & Dirsmith, M. (1993). Sampling and the
abstraction of knowledge in the auditing profession: an
extended institutional theory perspective. Accounting, Orga-
nization and Society, 18(1), 41±63.
Flint, D. (1988). Philosophy and principles of auditing. London:
McMillan.
Krause (1996). Death of the guilds: professions, states, and the
advance of capitalism, 1930 to the present. New Haven: Yale
University Press.
Llewellyn, S. (1998). Boundary work: costing and caring in the
social services. Accounting, Organizations and Society, 23,
23±47.
Pentland, B. T., & Carlile, P. (1996). Audit the taxpayer, not
the return: tax auditing as an expression game. Accounting,
Organizations and Society, 21(2/3), 269±287.
Pentland, B. T. (1993). Getting comfortable with the numbers:
auditing and the micro-production of macro-order. Account-
ing, Organizations and Society, 18(7), 605±620.
Power, M. (1999). The audit society: rituals of veri®cation (2nd
Ed.). Oxford: Oxford University Press.
Power, M. (1997). Expertise and the construction of relevance:
accountants and the environmental audit. Accounting, Orga-
nizations and Society, 22, 123±146.
Rose, N., & Miller, P. (1997). Political power beyond the state:
Problematics of government. British Journal of Sociology,
43(2), 173±205.
312 B.T. Pentland / Accounting, Organizations and Society 25 (2000) 307±312
doc_745512272.pdf