'ACCOUNTABILITY': AN EVER-EXPANDING CONCEPT?


Richard Mulgan

Discussion Paper No. 72

September 2000

ISBN: 0 7315 3415 8

ISSN: 10302190



Abstract

The scope and meaning of 'accountability' has been extended in a number of directions well beyond its core sense of being called to account for one's actions. It has been applied to internal aspects of official behaviour, beyond the external focus implied by being called to account; to institutions that control official behaviour other than through calling officials to account; to means of making officials responsive to public wishes other than through calling them to account; and to democratic dialogue between citizens where no one is being called to account. In each case the extension is readily intelligible because it is into an area of activity closely relevant to the practice of core accountability. However, in each case the extension of meaning may be challenged as weakening the importance of external scrutiny.

That 'accountability' is a complex and chameleon-like term is now a commonplace of the public administration literature. A word which a few decades or so ago was used only rarely and with relatively restricted meaning (and which, interestingly, has no obvious equivalent in other European languages (Dubnick 1998, 69-70)) now crops up everywhere performing all manner of analytical and rhetorical tasks and carrying most of the major burdens of democratic 'governance' (itself another conceptual newcomer). In the process, the concept of 'accountability' has lost some of its former straightforwardness and has come to require constant clarification and increasingly complex categorisation (eg Day and Klein 1987; Sinclair 1995).

One sense of 'accountability', on which all are agreed, is that associated with the process of being called 'to account' to some authority for one's actions (eg Jones 1992, 73). Indeed, this sense may fairly be designated the original or core sense of 'accountability' because it is the sense with the longest pedigree in the relevant literature and in the understanding of practitioners (Finer 1941, 338; Thynne and Goldring 1987, 8; Caiden 1988, 25). Such accountability has a number of features : it is external, in that the account is given to some other person or body outside the person or body being held accountable; it involves social interaction and exchange, in that one side, that calling for the account, seeks answers and rectification while the other side, that being held accountable, responds and accepts sanctions; it implies rights of authority, in that those calling for an account are asserting rights of superior authority over those who are accountable, including the rights to demand answers and to impose sanctions. (The inclusion of sanctions in the core of accountability is contestable on the ground that it may appear to go beyond the notion of 'giving an account'. On the other hand, 'calling to account', as commonly understood, appears incomplete without a process of rectification.)

In the context of a democratic state, the key accountability relationships in this core sense are those between the citizens and the holders of public office and, within the ranks of office holders, between elected politicians and bureaucrats. Core accountability has thus commonly covered issues such as how voters can make elected representatives answer for their policies and accept electoral retribution, how legislators can scrutinise the actions of public servants and make them answerable for their mistakes, and how members of the public can seek redress from government agencies and officials. It leads to questions about different channels of accountability and their relative merits, about the balance between accountability and efficiency, and about distinctions between political and managerial accountability.

But more recently, in academic usage at least, 'accountability' has increasingly been extended beyond these central concerns and into areas where the various features of core 'accountability' no longer apply. For instance, 'accountability' now commonly refers to the sense of individual responsibility and concern for the public interest expected from public servants ('professional' and 'personal' accountability), an 'internal' sense which goes beyond the core external focus of the term. Secondly, 'accountability' is also said to be a feature of the various institutional checks and balances by which democracies seek to control the actions of the governments (accountability as 'control') even when there is no interaction or exchange between governments and the institutions that control them. Thirdly, 'accountability' is linked with the extent to which governments pursue the wishes or needs of their citizens (accountability as 'responsiveness') regardless of whether they are induced to do so through processes of authoritative exchange and control. Fourthly, 'accountability' is applied to the public discussion between citizens on which democracies depend (accountability as 'dialogue'), even when there is no suggestion of any authority or subordination between the parties involved in the accountability relationship.

Linguistic development is not necessarily unhealthy and unbending resistance to new meanings is generally futile. The main purpose of this article is to analyse the relentless ramification of 'accountability', rather than to deplore it. Each of the extensions in meaning will be discussed in turn, with an indication of how it has grown out of the core meaning and can still be distinguished from that meaning. However, no analysis of accountability can pretend to be wholly without ulterior purpose. While many of the new analyses of accountability have added significantly to the understanding of public institutions (eg Day and Klein 1987; Romzek and Dubnick 1987), others, it will be suggested, by extending the concept of 'accountability' beyond its accustomed contexts, may involve a degree of unnecessary academic complication. They may also imply a questionable shift of focus away from the central importance of external scrutiny. The original core of accountability, signifying external scrutiny, justification, sanctions and control, is sufficiently distinct and important to warrant separate identification.

'Accountability' and 'responsibility'

One indication of how far the concept of 'accountability' has come to dominate its academic territory is provided by the changing terms used to discuss two seminal debates in public administration. In the United States, academic analysis of the relationships between bureaucrats and the public is grounded in the classic exchanges between Carl Friedrich and Herman Finer over how far public servants should rely on their professionalism and sense of personal morality and how far they should simply be following instructions from their political masters (Friedrich 1940; Finer 1941). This debate was originally couched in terms of different senses of 'responsibility', with Friedrich emphasising the inward responsibility of public servants to their professional standards and values and Finer reasserting the primacy of responsibility to external political direction. Finer, the advocate of external control, used 'accountability' to define his preferred sense of 'responsibility':

First, responsibility may mean that X is accountable for Y to Z. Second, responsibility may mean an inward sense of moral obligation (Finer 1941, 338).
But, significantly, the argument is expressed as a preference for a particular type of 'responsibility' rather than a type of 'accountability'. Half a century later, however, this same debate is now naturally described as an argument over the relative merits of different types of 'accountability', external or internal (eg Harmon and Mayer 1986, 47-49; Romzek and Dubnick 1987, 229; Dubnick 1998, 73; March and Olsen 1995, 165-7; Peters 1995, 318). Thus, 'accountability' has been extended beyond its core meaning of external scrutiny, as used by Finer, and now also includes Friedrich's inner responsibility of the individual to his or her conscience or moral values.

On the other side of the Atlantic and in Westminster jurisdictions generally, the traditional starting point for discussing such issues has been the role of ministers in answering to Parliament for the actions of their departments. Here the classic formulation has been in terms of 'ministerial responsibility'. The problems that arise from attempting to hold ministers solely responsible for all departmental behaviour have traditionally been explored by analysing the different senses of responsibility (eg Hart 1968; Thynne and Goldring 1987, ch 1 ). 'Accountability' might be brought in, as it was by Finer, to identify one of the senses or aspects of responsibility (eg Marshall and Moodie 1959, 68) but 'accountability' was certainly not expected to cover the whole range of activities and processes covered by 'responsibility'. Today, however, the same issues are most naturally canvassed in terms of different approaches to the problem of government 'accountability' (eg Marshall 1991; O'Toole and Chapman 1995; Pyper 1996; Rhodes 1997, 101-3). 'Ministerial responsibility' still remains in currency but more as a technical, constitutional term to be explained in terms of the history of Westminster conventions rather than as a readily understandable concept.

The expansion of 'accountability' has thus been accompanied by a corresponding contraction in 'responsibility'. Sometimes the two terms are used interchangeably, but 'responsibility' is now increasingly confined to its more accustomed ethical territory of personal liability, freedom of action and discretion, that is to the more internal aspects of official activity (eg Harmon and Mayer 1986, 48 -50; Uhr 1993). Whereas formerly 'accountability' was usually seen as a part of 'responsibility' (the external aspect) , the position is now often reversed with 'responsibility' taken to be a part of 'accountability' (the internal aspect).

A more promising development in the territorial jousting between 'accountability' and 'responsibility' is the emergence of a half-way position in which the field is divided between the two. 'Accountability' can then denote one set of responsibility/accountability issues, those concerned with the 'external' functions of scrutiny, such as calling to account, requiring justifications and imposing sanctions (ie the original, core senses of accountability), while 'responsibility' is left to cover the 'internal' functions of personal culpability, morality and professional ethics (eg Uhr 1993; Bovens 1998). Such a distinction allows 'accountability' to stand on its own, no longer under the wing of 'responsibility', and thus recognises its growing salience, while still confining it to its original, and still most widely accepted, sense.

'Internal' accountability

The redescription of the Friedrich/Finer debate, whereby 'accountability' now covers the area of administrative discretion championed by Friedrich against an undue concentration on external scrutiny, indicates the most significant extension of 'accountability' . This area of supposed 'internal' accountability is variously described, in whole or part, as 'professional' (Romzek and Dubnick 1987, 228; Sinclair 1995, 223) 'personal' (Sinclair 1995, 223, 230-31), 'inward' (Corbett 1996, 201-2) or 'subjective' (Kernaghan and Siegel 1987, 298; Gagne 1996).

We may begin with 'professional accountability' which is itself a term of some ambiguity and straddles the line between external and internal aspects of accountability, thus helping to facilitate the extension of 'accountability' to a more internal focus. In the first place, the accountability of professionals can refer straightforwardly to a species of external scrutiny specifically designed to match the complexities of professional knowledge. Romzek and Dubnick, for instance, in their influential typology of accountability systems (Romzek and Dubnick 1987), describe 'professional accountability' as a system marked by deference to expertise where reliance must be placed on the technical knowledge of experts and where close control from outside the organisation is inappropriate. However, though broad scope must be granted to the discretion of professionals, the accountability of the professionals still lies in their ultimate answerability to their administrative and political superiors.

...public officials must rely on skilled and expert employees to provide appropriate solutions. Those employees expect to be held fully accountable for their actions and insist that agency leaders trust them to do the best job possible. If they fail to met job performance expectations, it is assumed that they can be reprimanded or fired (Romzek and Dubnick 1987, 229)

The authors identify accountability with the 'management of expectations' by public servants and note that these expectations can be generated both internally and externally (id 228). But even where the sources of accountability expectations are internal, they are internal to the organisation not to the individual. From the point of view of the particular officials, all accountability involves control from someone else and in that sense is external. This deference to superiors is a familiar feature of accountability within bureaucracies and other hierarchical organisations (sometimes referred to as 'managerial' accountability (Day and Klein 1987, 26)) and clearly falls within the purview of core accountability (Stewart 1986, 126-31). Moreover, in spite of their emphasis on the point of view of public servants, Romzek and Dubnick leave no doubt that the driving force behind all systems of accountability, including professional accountability, is the democratic imperative for government organisations to respond to demands from politicians and the wider public (eg Romzek and Dubnick 1994, 263, 269).

'Professional' accountability is also evident in systems of professional peer review. Because professionals, virtually by definition, lay claim to expert knowledge beyond the full understanding of non-professionals, their actions can be fully assessed only by fellow-professionals (Deleon 1998, 548-51). External scrutiny, if it is to be reliable and effective, requires the establishment of review boards or disciplinary committees containing members of the profession. Standard examples of such peer-review mechanisms are the disciplinary boards operated by the medical and legal professions.

The accountability of professionals to their peers raises potential difficulties in the context of the broader democratic accountability of publicly employed officials to the public. Particularly in relation to highly specialised and independent professions, such as doctors, there may be conflict between the practitioners' accountability to professional bodies and their accountability to the 'lay' representatives of the public they are supposed to serve (Day and Klein 1987, ch 3). But such difficulties stem from the need to satisfy multiple channels of external accountability, a familiar problem for many public officials. For instance, similar conflicts occur for bureaucrats dealing with legislative committees where there may be a clash between the officials' duty to their superiors in the executive and their accountability to the legislature. This type of professional accountability, however, need not involve any extension of the core meaning of accountability. It still requires external accountability to superiors.

Professional accountability, however, does sometimes extend beyond purely external scrutiny to cover adherence to professional norms or 'standards'. A fully qualified professional is someone who has acquired the techniques and values of the particular profession and is required to exercise professional judgment, typically in an unsupervised context (Friedrich 1940; Romzek and Dubnick 1987, 229). Such independent action in accordance with professional standards is now being described in the academic literature as involving the exercise of accountability (Sinclair 1995, 223). In one sense, such a usage does not go beyond the core sense of external scrutiny. The concept of 'account-ability' includes an implication of potentiality, literally an 'ability' to be called to 'account'. It may thus refer to the potential for external scrutiny under which most expert professionals work, however independent they may be in their day-to-day decisions. Every medical doctor, for instance, knows that any action he or she takes (or does not take) could potentially become the object of disciplinary investigation or a legal action. In this respect, professionals are literally accountable in their professional actions because they are able to be called to account later for any of their actions. But, in this case, the accountability they potentially face in all their actions is still conceived of as an external accountability, in the core sense.

However, professional accountability for independent discretionary acts is also being used to refer to the purely personal exercise of judgment and adherence to internalised standards, regardless of any external scrutiny or sanction, actual or potential. In this sense, accountability is being clearly identified with the sense of personal responsibility which covers the conscientious performance of duties, sometimes referred to as 'role-responsibility' (Hart 1968, 212-4; Lucas 1993,193-4; Harmon and Mayer 1986, 49). Here, the sole source of obligation is internal, in the professional's personally endorsed values. The sanction, if any, is also internal, in the individual's own sense of professional guilt or malfeasance. In this respect, internalised professional accountability shades into what is knows as 'personal' (Sinclair 1995, 223 ) or 'inward' accountability (eg Corbett 1996, 201-2) where the basis for action is not so much the individual's professionalism as his or her own personal sense of morality involving general moral values such as honesty, integrity and fairness. Indeed, personal and professional values are hard to separate, as is clearly evident from the way in which codes of professional ethics typically refer to general values such as honesty, fairness and justice, as well as to standards specific to the profession in question. If there is any 'other' to whom one is accountable for maintaining such personal and professional standards it is a hypothesised inner self or personal conscience (Corbett 1996, 201).

Day and Klein note a similar internalisation of accountability in their study of lay members of the public placed in supervisory roles over expert providers of public services in health, police, water, education and social work. Their main focus was on how the non-expert board members could hold expert professionals to account. However, as well as being asked about the accountability of professionals under their supervision, board members were also asked about their own accountability, to whom, if anyone, they themselves were accountable (Day and Klein 1987, 100-01; 128-30; 152-3; 183-5; 217-19). Because most of those interviewed were appointed rather than elected in their own right to represent the public, this question posed a potential problem which some respondents obviously had difficulty in answering. Many opted for accountability to the general public while others mentioned accountability to the authority that had appointed them (and a few who had been appointed by partisan local authorities preferred accountability to a political party (id 184)). A substantial number accepted that there was no clear accountability at all. In the face of this impasse, some followed the suggestion that they were ultimately accountable to themselves or their consciences (id 101, 218). As the authors summarise their findings:

Lastly, many members tended to define accountability in terms of their responsibility, either to the community being served or to their own sense of what was sensible or proper: they internalised accountability, as it were, as a general duty to pursue the public good according to their own criteria of what was right (id 229).

Such extension of 'accountability' from external reckoning to internal agonising involves an easy transition. External accountability seeks to investigate and assess actions taken (or not taken) by agents or subordinates and to impose sanctions. The extent to which individual agents or subordinates can fairly be held accountable for particular actions, particularly when it comes to the matter of sanctions, depends on whether they can be said to have been genuinely involved in deciding those actions. Thus, external accountability and blame inevitably raise issues of individual choice and personal responsibility (Bovens 1998). Individual choice and responsibility, in turn, lead to questions of professional morality and personal values. After all, the term 'responsibility' itself has travelled a similar route from the external to the internal, from the capacity to 'respond' or answer to someone else, to the capacity to act freely and 'responsibly'. Indeed, in the case of 'responsibility', the transition is so firmly established that, as already noted, the secondary, internal sense has now come to predominate over the original, external sense. If 'responsibility' can make this change, why should not 'accountability' follow suit, at least part of the way?

On the other hand, the internalisation of accountability, though understandable, is not logically inevitable. Granted the complexities involved in separating the internal from the external aspects of human action, a sufficiently robust distinction can still be maintained between having to account to someone else for one's actions and not having to do so. It make sense to say that particular public servants are accountable to certain other people and bodies through certain mechanisms for the performance of certain tasks. These processes of external accountability can be said to have a particular impact on the decisions and behaviour of public servants. However, a full explanation of why public servants decide and behave as they do will also require reference to other 'internal' factors such as their personal and professional values which, as Friedrich recognised, can be identified separately from the requirements of external accountability. The growing use of 'accountability' and 'responsibility' to refer to the external and internal aspects of behaviour respectively helps to mark this important distinction.

The internalisation of 'accountability', though finding currency among academics, does not appear to be widely accepted among actual practitioners themselves. For instance, in her research into the views on accountability of senior public servants in the Australian state of Victoria, Sinclair endorses a wide-ranging view of accountability. She includes 'professional' and 'personal' accountability as forms of accountability (along with 'political', 'public' and 'managerial') and quotes Day and Klein as evidence for an internalised sense of accountability. However, it is notable that her discussion of these inner forms of accountability in open-ended conversations with her respondents (Sinclair 1995, 229-31) is conducted in terms of concepts other than accountability. 'Professional' accountability is largely a matter of probity and being business-like while 'personal' accountability raises issues of honesty and the public interest. Whereas the external, forms of accountability ('political', public' and 'managerial') were volunteered by the respondents themselves in answer to questions about their accountability, the two internal forms are the researcher's own labels. In this respect, her supposed 'chameleon' of accountability is, at least in part, an invented animal.

Though Day and Klein themselves identify an internalised sense of accountability, their reported research shows that their respondents embraced such a concept with reluctance. In their interviews, internalised accountability was proffered as a solution to the dilemma facing board members who felt they ought to be publicly accountable but could not say to whom. But not all respondents took this route. Others were prepared to face the consequence that they were actually unaccountable (eg Day and Klein 1987, 183). Moreover, even those who did opt for inner accountability appear to have done so only under probing questioning. It was not a species of accountability which came immediately to mind like external accountability. That is, the notion of internalised accountability may sometimes be seized on as a last resort by those who feel that they ought to be accountable to someone but cannot identify to whom. Though the notion has sufficient logical plausibility to be intelligible and acceptable, it should not be represented as common usage.
'Accountability' and 'control'

Another extension of accountability is its application to various methods of imposing control over public organisations. The core sense of accountability is clearly grounded in the general purpose of making agents or subordinates act in accordance with the wishes of their superiors. Subordinates are called to account and, if necessary, penalised as means of bringing them under control. In a democracy, it is because the people wish to control the actions of public officials that they (or their representatives) make these officials answer, explain and accept sanctions. Indeed, the auditing of government agencies was identified with an earlier, more specific sense of 'control' (as in government 'controllers' or 'comptrollers' (Gregory 1990). Understanding 'control' in the broadest sense of making public agencies do what the public and their representatives want, accountability and control are intimately linked because accountability is a vital mechanism of control (Uhr 1993, 6).

However, accountability is sometimes taken to be more than just a mechanism of control; it becomes identified with control itself. The problem of accountability thus becomes: 'How do political leaders and the public persuade, cajole and force administrative agencies to do their bidding' (Peters 289). Constructing an appropriate structure of accountability amounts to constructing institutions which will guarantee that public officials are appropriately constrained. Indeed, if the central issue of democracy is to control the government so that it complies with the people's preferences, then the entire complex edifice of a modern democratic political system becomes in effect a system for securing government accountability (Day and Klein 1987, ch 1).

From this perspective, institutions of accountability include all institutions that are aimed at controlling or constraining government power, for instance legislatures, statutory authorities, and courts. Devices of accountability then include the separation of powers, federalism, constitutionalism, judicial review, the rule of law, public service codes of conduct and so on, all of which have an effect on the control of public power. Also to be added are the key extra-governmental institutions of a democratically effective civil society which help to constrain governments, for instance competitive markets, interest groups and the mass media (eg Peters 1995, 300-301). In this way, accountability threatens to extend its reach over the entire field of institutional design.

On the other hand, such an extension, though understandable, is not logically inevitable. If 'accountability' is kept to its central sense of external scrutiny, it refers to only one type of institutional mechanism for controlling governments and government officials, where governments and officials are actually called to account, made to answer for their actions and to accept sanctions. In this case, there are other important types of control mechanism, besides accountability mechanisms. For instance, constitutional constraints or legal regulations control governments by restricting their freedom of action, requiring them to pursue certain purposes or to follow certain procedures. Public officials usually have full knowledge of these legal constraints and frame their policies and decisions so that they stay within the legal limits imposed upon them. For the most part, their compliance is unquestioning and unquestioned and issues of formal accountability do not arise.

Certainly, if agencies or officials step outside stated constitutional or legal limits they face the prospect of sanctions and will thus be held accountable, in the core sense, for their improper actions. But being accountable for alleged breaches of the law does not mean that compliance with the law is also an act of accountability or that the law itself is an accountability mechanism, again in the core sense. In this core sense, the legal accountability mechanism is confined to that part of the law which lays down enforcement procedures. The main body of the law, which most public servants follow as a matter of normal practice, is an instrument for controlling their behaviour but not for holding them accountable.

Within the bureaucracy itself, accountability can also be seen as merely one means whereby external agencies and the public control government policy. Government departments and other agencies are subject to different types and degrees of policy control, whether from executive politicians, legal mandates or their own internal leadership (Romzek and Dubnick 1987). Bureaucrats will normally be held accountable for their performance of given policies, both internally within the organisation and often also to outside agencies. Such accountability, however, would usually be considered as only one mechanism by which the policy process is shaped to satisfy the demands of superiors. In the 'policy-cycle' literature, for instance (eg Parsons 1995, 78-9), accountability occurs as part of later stages of 'evaluation' and 'correction'. During the prior stages of the cycle, such as 'issue definition' or 'options analysis', control is more likely to be exerted through policy- makers' attending to explicit instructions or being sensitive to political pressures. Admittedly, such analyses of the policy process are notoriously artificial and over-schematic and, no doubt, issues of external evaluation and accountability impinge at all stages of actual policy-making through the 'law of anticipated reactions'. The need eventually to answer to supervisors casts a long and influential shadow over the behaviour of public servants. Again, we should remember the element of potentiality in accountability, the implication that an official may be called to account for anything at any time. None the less, the point remains that accountability, in the core sense of answering to external scrutiny, is only one aspect of how government agencies are subjected to public control. From this perspective, bureaucratic agencies are not primarily structures of accountability. They involve accountability but are equally structures of policy development and policy implementation.

On such a more restrictive view of accountability, only a few institutions, such as audit offices, ombudsmen and administrative tribunals, are properly described as 'institutions of accountability' because their primary function is to call public officials to account. Other institutions may adopt an accountability role, though it is far from their primary purpose. For instance, legislatures have a variety of functions - legislative, financial, investigative and so on - among which holding the executive publicly accountable is merely one, albeit extremely important, function. Within the overall operations of a modern legislature, accountability is particularly associated with certain procedures, such as committee inquiries or (in systems of parliamentary government) the formal questioning of ministers. But it may occur at any point, in legislative debate or financial authorisation, where members of the government are required to explain their actions and take the consequences. Thus legislatures are vital institutions for securing public accountability but accountability is only one among a number of their purposes.

Similarly, the legal system, whose overall function is to enforce the community's laws, is primarily concerned with regulating the behaviour of private organisations and individual citizens. In most political systems, the courts also play an important part in upholding the constitution and restraining government power. In so far as the courts offer members of the public the opportunity of holding public government officials to account for their actions they can be said to act as institutions of public accountability. Again, accountability, in its core sense, can be seen as an important, but not necessarily defining, role for the legal system and the courts.

Similar distinctions can be applied within the private sector and civil society. Some non-governmental institutions may be said to have a primarily accountability function, for instance watchdog groups set up with the prime purpose of monitoring and scrutinising particular areas of government activity. Other organisations, such as sectional interest groups or the media, have a more partial or incidental (though no less important) accountability focus. The dominant function of most interest groups, for example, is to lobby for their members' interests, a goal which sometimes involves active investigation of government policies and the interrogation of government officials. As recognised stakeholders, they may lay claim to a right to question and be informed and thus enforce accountability on governments. The media, too, though in the business of entertaining and informing (usually for profit), also assist the accountability of governments by their capacity to make politicians and officials face public scrutiny.

Thus, a broader, institutional and organisational approach to accountability may encourage a tendency to equate accountability with the general democratic issue of how to design public institutions so that they are amenable to public control. The reason for this expansion of its meaning is clear. The need to hold the government to account is itself critical to all effective attempts to control public power. Without mechanisms for demanding explanation, applying judgment and imposing sanctions, institutions that are designed to control will fail to achieve their purpose. So central is accountability as a means of achieving control that it can easily be taken to stand for control itself. Thus mechanisms designed to control, such as legal regulations or political instructions, can be taken for mechanisms of accountability even when they do not directly involve any actual accounting or scrutiny.

On the other hand, such an extension of meaning is not irresistible. A reasonably clear distinction may be still be maintained between accountability and control by which accountability remains merely one means, or set of means, for enforcing control, through the demand for explanation and the imposition of sanctions. An institutional emphasis in the study of accountability, as in the typology of accountability systems constructed by Romzek and Dubnick (1987), is very much to be welcomed. No one type of institutional structure can be guaranteed to deliver effective accountability for all types of public activity. Designing public institutions in order to maximise the accountability of their officials requires a careful matching of appropriate institutional structures to the differing types of issues and skills involved (Deleon 1998). However, such an emphasis can still make sense within a more restrictive understanding of 'accountability' itself.

'Accountability' and 'responsiveness'

A third extension of 'accountability' is to equate it with the responsiveness of public agencies and officials to their political masters and the public. 'Responsiveness', like 'control', refers to the aim of making governments accord with the preferences of the people. However, whereas 'control' stresses the coercive role of external pressure, 'responsiveness' points more widely to the public servants' general compliance with popular demands, from whatever motive. Just as accountability is such a powerful means of imposing control that is has sometimes been identified with control, it has also been seen as so important in encouraging responsiveness that the two concepts have on occasion been merged.

'Responsiveness' has been applied to two distinct relationships between officials and the wider public. In the first place, public agencies are expected to be responsive to other actors within the political system, particularly to elected politicians aiming to control their activities. They need to follow and anticipate the directions of their political masters (Saltzstein 1992; Romzek and Dubnick 1994, 266). Secondly, agencies providing services to members the public are called on to be responsive to the needs of their clients in a way analogous to private sector firms being sensitive to consumer demands (Hughes 1998, 236-7). In this case, the responsiveness of government officials is directly to the public rather than indirectly via the public's representatives.

The former avenue of responsiveness, that through the people's representatives, is a longstanding concern of democratic public administration. How to make the appointed bureaucracy comply with the wishes of the elected branches of government has been one of the recurrent themes of the responsibility/accountability literature. It is at the heart of the Friedrich/Finer debate and is also a key assumption of ministerial responsibility. If officials can be made compliant with their political superiors then the main objective of accountability will have been achieved. Indeed, according to some analysts, accountability is to be identified in terms of responsiveness. Thus O' Loughlin, in response to Romzek and Dubnik and their emphasis on managing expectations within agencies, seeks to reaffirm 'responsiveness to outside actors as the central spirit of accountability' (O' Loughlin 1990, 283). He develops measures of accountability which include the extent to which officials anticipate the wishes of their superiors and the quality of communications systems which exist between agencies and their superiors. The effectiveness of the accountability system is then assessed in terms of the extent to which the actions of officials are aligned with their political masters.

The identification of accountability with this type of responsiveness is, once again, readily understandable. It draws attention to the importance of anticipated reactions by officials as part of the effect of scrutiny mechanisms. That is, the effectiveness of accountability mechanisms is to be observed not simply in the occasions when officials are actually brought to account. Much more important in securing compliance is the ever-present threat of being called to account, the potential implicit in accountability. In addition, the emphasis on communications systems points to an important ingredient in successful accountability structures. Accountability depends on the free flow of appropriate information and on effective forums for discussion and cross-examination. Unless those calling subordinates to account have full access to the relevant people and the relevant information their investigations and assessments will be frustrated.

Once again, however, the identification is not inevitable. Instead of equating 'accountability' and 'responsiveness', one can restrict 'accountability' to its core sense, in which case it becomes just one among a number of different motives that induce officials to follow and anticipate the wishes of their superiors. The fear of being called to account, of facing scrutiny and possible penalty, is undoubtedly a pervasive motive among public officials. On the other hand, other motives are not insignificant. For instance, professional public servants may accept the superior legitimacy of elected representatives and agree to defer to their political masters out of democratic conviction. More cynically, ambitious public servants may wish to ingratiate themselves with their superiors not though fear of accountability but in the hope of personal advancement. In such cases, it is not the fear of being called to account but rather personal values or career advantage that motivate responsiveness.

The other avenue of responsiveness, that of officials directly to members of the public, has figured prominently in the managerial literature on public-sector reform (eg OECD 1987; Osborne and Gaebler 1992, ch 6). One of the main aims of the managerial-reform movement has been to make public agencies as responsive to their clients as private sector companies are to their customers. Public officials have been encouraged to be more directly approachable and accommodating to members of the public and less concerned with following set procedures or deferring to the instructions of their bureaucratic superiors. 'Client focus' is the major catch phrase and citizens' charters, agencification and competitive provision are some of the key mechanisms (Stewart 1992). This change in emphasis is sometimes described as an extension of 'accountability' (Hughes 1998, 236-7). As well as being accountable 'upwards' through the hierarchical chain of managerial command, public servants, particularly those engaged in service delivery, are now also seen as accountable 'outwards', immediately to the public, through the requirement that they respond directly to their clients' expressed needs (eg Corbett 198-200). Managerialist manifestos (eg OECD 1987, 29; 126; World Bank 1998, 111) use 'responsiveness' and 'accountability' to citizens as interchangeable synonyms. Market-style relationships are said to provide an alternative channel of accountability because they provide incentives for providers to take the wishes of consumers 'into account' (Stone 1995, 521).

On the other hand, the identification can be challenged. If 'accountability' implies rights of scrutiny and possible sanction, then much of the client focus urged on the public sector does not imply accountability but rather reflects a general political imperative to provide better service to the public. Some of the client focus mechanisms certainly involve accountability procedures in the core sense, for instance where there is scope to complain about unsatisfactory service and seek redress. Charters with complaints procedures, ombudsmen and administrative tribunals are all institutions of accountability designed to make public servants more responsive to the public's needs. But where public providers have been induced to be more 'customer friendly' through other means, such as management incentives or changes in corporate culture, they have certainly been made more 'responsive' but not necessarily more 'accountable'.

Similarly, the private sector's focus on service to consumers, which provides much of the impetus for increased client responsiveness on the part of public agencies, need not be seen as due to any superior accountability in the private sector. While a customer may hold a private-sector provider accountable in the case of a faulty individual purchase or contract, he or she has no general right to demand that the private provider offer services that meet his or her perceived needs. In a competitive market, the main mechanism of responsiveness is consumer choice, the capacity of the consumer to 'exit' to an alternative provider. Accountability, on the other hand, as usually understood, is a 'voice' not an 'exit' option in which subordinates are required to account and to accept direction. In the private sector, accountability applies more to owners and shareholders, who have rights to call the company's managers to account for the company's performance, than to customers whose main right is to refuse to purchase.

'Accountability' and 'dialogue'

A final extension of 'accountability' is where the term is used to stand for the public dialogue which is seen as an essential part of democracy. Here it is the language-based nature of accountability that is stressed, rather than its institutional or motivational aspects. Accountability is seen to be a dialectical activity, requiring officials to answer, explain and justify, while those holding them to account engage in questioning, assessing and criticising. It thus involves open discussion and debate about matters of public interest and so becomes equated with the principles of deliberative democracy. Day and Klein, for instance, ground their analysis of accountability in the assumption that it is a social activity requiring a 'shared set of expectations and a common currency of justifications' including 'agreement about the language of justification' (Day and Klein 1987, 5). In their conclusion they are drawn to the importance of dialogue between the various actors involved and assert that 'political deliberation .. is at the heart of accountability' (id 244).

March and Olson begin their treatment of accountability (March and Olsen 1995, ch 5) with the notion of an 'account' itself which they understand as providing an interpretation or explanation. They thus place explanation and justification at the core of accountability. Calling people to account means inviting them to explain and justify their actions within two competing logics, that of consequences and that of appropriateness (id 154). The accountability required of democratic governments produces contestable political 'accounts' within a context of shared beliefs and values (id 167-8) and thus helps to build the autonomous public sphere and civil society sought by Habermas and others (id 180-1). Harmon, too, in his critique of rationalism in public administration, sees the accountability of public servants as lying, at least in part, in a form of continuous, open-ended dialogue between themselves and their publics (Harmon 1995, 191-97). Again, in effect, accountability is being linked to explanation and justification which implies a dialectical exchange between officials and those to whom they are accountable.

Certainly, the requirement that rulers and officials publicly account for their actions, whether to legislatures, the courts, in the media or on the hustings, does force them to engage in a form of dialogue with their public. Such dialogue may be seen as a crucial element in modern deliberative democracy. Moreover, the activities of questioning and answering implicit in such accountability undoubtedly entail a key role for explanation and justification. Being called to account for one's actions often requires one to explain and justify what was done. Even where apparently 'bare' information is sought, such as in financial accounting, the information will only make sense within an explanatory and justificatory framework assumed by the questioner and accepted, or contested, by the respondent. The various discourses of accountability, including assumptions of institutional and personal responsibility, are an important aspect of accountability and worth careful academic investigation (eg Sinclair 1995, Bovens 1998).

However, it is one thing to recognise that accountability involves public explanation and justification and another to identify it with the dialogue between citizens found (or at least hoped for) in a deliberative democracy. The key difference is that the accountability of public officials, at least in its core sense, implies an unequal relationship of superior and subordinate in which the latter is required to take directions from the former and to accept sanctions, if necessary, for unsatisfactory performance. The dialogue of accountability occurs between parties in an authority relationship and can only be understood in the context of that relationship. This relationship is crucially different from that presupposed by democratic debate which takes place in a public space between citizens conceived of as equals. To be fair, the various theorists who have advanced the concept of accountability as dialogue have not been blind to the need for sanctions to enforce accountability (Day and Klein 1987, 247-8; March and Olsen 1995, 165-7; Harmon 1995, 194-5) and have thereby recognised the tension that such enforcement may create with a more idealised form of democratic dialogue. None the less, to merge accountability with the general dialogue of democratic citizens is to extend it beyond its normal context of authority and control.

Conclusion

The scope and meaning of 'accountability' has been extended in a number of directions well beyond its core sense of being called to account for one's actions. It has been applied to internal aspects of official behaviour, beyond the external focus implied by being called to account; to institutions that control official behaviour other than through calling officials to account; to means of making officials responsive to public wishes other than through calling them to account; and to democratic dialogue between citizens where no one is being called to account. In each case the extension is readily intelligible because it is into an area of activity closely relevant to the analysis and assessment of accountability. The question of whom to hold to account for what raises immediate issues of personal responsibility and 'internal' values. The effectiveness of accountability mechanisms must be examined in the context of how bureaucracies are to be controlled and how bureaucrats can be made responsive to the wishes of elected politicians and the public. The public explanation and justification involved in accountability make it an important component of deliberative democracy. On the other hand, as already demonstrated, these connections may be recognised without necessarily extending the meaning of 'accountability' to include the areas and activities with which it is so closely related.

In some respects, the issue of definition is not just about terminology but also about institutional and administrative policy, about the relative emphasis to be placed on external scrutiny and sanctions compared with other means of securing the compliance of officials in a complex democracy. Given the current potency of 'accountability' as a political value, restricting it to its original meaning of external scrutiny is a means of siding with Finer and others in their resolute warning about the dangers of elevating administrative discretion beyond the reach of political direction. Conversely, embracing the extension of accountability to cover the pursuit of personal and professional values is sometimes linked with a recommendation to reduce the ever-increasing demands of political and managerial accountability and to place more trust in the independent judgment of public servants (Sinclair 1995, 233).

Similarly, in the case of responsiveness to members of the public, those accepting the equation of 'accountability' with a generalised 'client focus' are more likely to place faith in private sector, market-style management mechanisms as a way of reducing the insensitivity of bureaucrats and to be relatively unconcerned about any possible erosion of traditional political accountability (Hughes 1998). On the other hand, those who insist on a clear distinction between 'accountability' and 'responsiveness' may do so because they wish to reassert the value of citizens' rights to call public servants to account and to enforce sanctions through political processes (Rhodes 1997, 101-3).

The present author, it will be apparent, is more sympathetic to the linguistically conservative end of this spectrum which restricts 'accountability' to its assumed core and places particular emphasis on holding the powerful to account through political and legal channels of external scrutiny and sanctions. Some additional support for this stance has been found in the usage of actual practitioners, reported above, who did not volunteer the internalisation of 'accountability' until pressed by researchers. The other extensions in meaning also appear to be more the creations of academics pursing their own intellectual agendas rather than the results of shifts in everyday usage. In the absence of further research, the suspicion must remain that the extension of accountability beyond its traditional, external focus, has little general support outside the academic community.

Refusal to accept the extensions of meaning proposed for 'accountability' does not, of course, solve the many complex issues that have been associated with accountability. The problem of the congruence between external scrutiny and internal discretion still remains, whether it is described as a clash between 'accountability' and 'responsibility' or between two aspects of 'accountability'. So too does the issue of the relative merits of citizens' rights of redress compared with competitive market provision as alternative means of ensuring satisfactory delivery of public services, whether it is a comparison of 'accountability' and market 'responsiveness' or of two types of provider 'accountability'. Within the purview of accountability in its restricted sense, there are still important distinctions to be made, for instance between political and managerial accountability (Day and Klein 1987, ch 1; Hughes 1998, ch 11), between ministerial and other avenues of public accountability (Uhr 1993, Finn 1993, Stone 1995, Pyper 1996, Thomas 1998), between the different accountability systems suited to different types of issue (Romzek and Dubnick 1987, 1994; Deleon 1998), and between the various processes of accountability (Mulgan 1997). However, in these continuing discussions, the interests of both analytical clarity and citizens' rights may be better served by keeping the concept of 'accountability' itself within limits.

REFERENCES


Bovens, M. (1998), The Quest for Responsibility - Accountability and Citizenship in Complex Organisations (Cambridge: Cambridge University Press)

Caiden, G. E. (1988), 'The problem of ensuring the public accountability of public officials' in J. G. Jabbra and O. P. Dwivedi (eds), Public Service Accountability (West Hartford: Kumarian), 17-38

Corbett, D. (1996), Australian Public Sector Management Second edition (Sydney: Allen and Unwin).

Day, P. and Klein, R. (1987), Accountabilities: Five Public Services (London: Tavistock).

Deleon, L. (1998), 'Accountability in a 'reinvented' government', Public Administration 76, 539-58

Dubnick, M (1998), 'Clarifying accountability: an ethical theory framework' in C. Sampford and N Preston (eds), Public Sector Ethics (London: Routledge), 68-81

Finer, H. (1941), 'Administrative responsibility and democratic government', Public Administration Review 1, 335-50

Finn, P. (1993), 'Public trust and public accountability', Australian Quarterly 65 (Winter), 50-59.

Friedrich, C. J. (1940), 'Public policy and the nature of administrative responsibility' in C. J. Friedrich and E. S. Mason (eds) Public Policy (Cambridge: Harvard University Press), 3-24

Gagne, R. L. (1996), 'Accountability and public administration', Canadian Public Administration 39, 213-25

Gregory, R. (1990), 'Parliamentary control and the use of English', Parliamentary Affairs 43, 59-76

Harmon, M. M. (1995), Responsibility as Paradox (Thousand Oaks: Sage)

Harmon, M. M. and Mayer, R. T. (1986), Organization Theory for Public Administration (Boston: Little, Brown)

Hart, H. L. A. (1968), Punishment and Responsibility (Oxford: Oxford University Press), 186-209

Hughes, O. E. (1998), Public Management and Administration Second edition. (London: Macmillan)

Jabbra, J. G. and Dwivedi, O. P. (eds) (1988), Public Service Accountability: A Comparative Study (West Hartford, Connecticut: Kumarian Press).

Jones, G. W. (1992), 'The search for local accountability' in S. Leach (ed), Strengthening Local Government in the 1990s (London: Longman), 49-78

Kernaghan, K. and Siegel, D. (1987), Public Administration in Canada (Toronto: Methuen)

Kinley, D. (1995), 'Governmental accountability in Australia and the United Kingdom', University of New South Wales Law Journal 18, 409-27

Lucas, J. R. (1993), Responsibility (Oxford: Clarendon Press).

March, J. G. and Olsen, J. P. (1995), Democratic Governance (New York: Free Press)

Marshall, G. (1991), 'The evolving practice of parliamentary accountability; writing down the rules', Parliamentary Affairs 44, 460-9

Marshall, G. and Moodie, G. C. (1959), Some Problems of the Constitution (London: Hutchinson)

Mulgan, R. (1997), 'Processes of accountability', Australian Journal of Public Administration 56 (1), 25-36

OECD (1987), Administration as Service: the Public as Client (Paris: Organisation for Economic Co-operation and Development)

O' Loughlin, M. G. (1990), 'What is bureaucratic accountability and how can we measure it?', Administration and Society 22, 275-302

Osborne, D. and Gaebler, T. (1992), Reinventing Government (Reading, MA: Addison-Wesley)

O' Toole, B. J. and Chapman, R. A. (1995), 'Parliamentary accountability' in B. J. O' Toole and G. Jordan (eds), Next Steps. Improving Management in Government? (Aldershot: Dartmouth), 118-41

Parsons, W. (1995), Public Policy (Cheltenham: Edward Elgar)

Peters, B. G. (1995), The Politics of Bureaucracy Fourth Edition(White Plains: Longman)

Pyper, R. (ed) (1996), Aspects of Accountability in the British System of Government (Eastham: Tudor Business)

Rhodes, R. A. W. (1997), Understanding Governance. Public Networks, Governance, Reflexivity and Accountability (Buckingham: Open University Press)

Romzek, B. S. and Dubnick, M. J. (1987), 'Accountability in the public sector: lessons from the Challenger tragedy', Public Administration Review 47, 227-38

Romzek, B. S. and Dubnick, M. J. (1994), 'Issues of accountability in flexible personnel systems' in P. W. Ingraham and B. S. Romzek (eds), New Paradigms for Government (San Francisco: Jossey-Bass)

Saltzstein, G. H. (1992), 'Explorations in bureaucratic responsiveness' in L. B. Hill (ed), The State of Public Bureaucracy (M. E. Sharpe: Armonk), 171-89

Sinclair, A. (1995), 'The chameleon of accountability', Accounting Organizations and Society 20, 219-37

Stewart, J. D. (1986), The New Management of Local Government (London: Allen and Unwin)

Stewart, J. (1992), 'The rebuilding of public accountability' in N. Flynn (ed), Change in the Civil Service (London: Chartered Institute of Public Finance and Accountancy)

Stone, B. (1995), 'Administrative accountability in the "Westminster" democracies: towards a new conceptual framework', Governance 8, 505-26

Thomas, P. G. (1998), 'The changing nature of accountability' in B. G. Peters and D. J. Savoie (eds), Taking stock : assessing public sector reforms (Montreal: McGill-Queen's University Press)., 348-93

Thynne, I. and Goldring, J. (1987), Accountability and Control: Government Officials and the Exercise of Power (Sydney: Law Book Company).

Uhr, J. (1993), 'Redesigning accountability', Australian Quarterly 65 (Winter), 1-16.

World Bank (1998), World Development Report 1997 (Oxford: Oxford University Press)

(c) 1998 The Australian National University
feedback & queries: PPP Program Coordinator
date last modified: Wednesday 11 October 2000