Nadia Hilliard
THE FREEDOM OF INFORMATION ACT (FOIA) appeared as part of a wave of transparency and accountability reforms in the latter half of the twentieth century that included inspectors general, auditing bodies, and independent commissions. John Keane’s concept of monitory democracy captures the democratic practices made possible by such reforms: monitoring tools permit citizens to hold public officials to account not only in elections but also through permanent, ongoing scrutiny of government activity.1 Beyond their scrutinizing function, these tools are unified by their focus on transparency and the dissemination of information to the public. Together they comprise what has become a web of accountability, a set of instruments and a corresponding mode of democratic practice, with novel forms of citizen engagement, centered on information extraction and control. Political philosophers have long studied the process of “refining and enlarging” popular sentiment through the institutions of representative government, but the flow of information from state to citizen had gone undertheorized until recent decades. The process by which state-generated information is produced and transformed and the questions of who will mediate and translate it are the theoretical questions that accompany the demand for transparency.
The emergence of these monitory tools has instantiated new political roles for citizens and governing officials, redistributing authority in the public sphere. On the citizens’ side, such monitory practices expand the range of democratic activities for the general public far beyond periodic voting and stimulate demand for new technologies to map and render complex government information intelligible. In government, the need for bureaucrats to manage, interpret, and narrate the voluminous output of government data has been part and parcel of the expanding postwar administrative state, leading to a burgeoning, but not yet fully developed, role as a proactive informant. These developments, however, have the potential to set in motion paradoxical dynamics that might undermine the very democratic values they are said to promote. Three paradoxes—a bureaupathology paradox, an expertise paradox, and an expectations gap paradox—have the potential to compromise the ability of transparency-directed initiatives to enhance democracy. Although my concerns about the effects of these paradoxes are primarily theoretical, they suggest caution in the design of transparency tools and the need for further empirical research into the by-products of the transparency movement.
To illustrate these paradoxes, I trace two complementary, monitory instruments: FOIA and the inspector general (IG) system. IGs are hybrid inspectors-auditors-investigators tasked with rooting out fraud, waste, and abuse in the federal bureaucracy, and they perform routine audits and special investigations at the behest of Congress and in response to internal agency requests. Their most basic duty is to send regular reports to Congress on the state of their host agency; agency heads cannot interfere with these investigations or reports except through a set of primarily national security-related exemptions. Structurally, the loyalty of IGs is divided between Congress and their agency heads in order to preserve their independence, a state of affairs that one IG described as being akin to “straddling a barbed wire fence.”2 Although they have certain law enforcement authority and can issue subpoenas and initiate proceedings in the Justice Department, IG recommendations are not legally binding and can easily go unheeded.
Before assessing the three paradoxes, it is necessary to articulate the relationship between FOIA and the IGs, and the place of each of them in the wider system of public accountability. The values of accountability and transparency have inspired a host of spatial metaphors to describe the dynamics of these relationships, such as “architecture of accountability,” “ecology of transparency,” and “inverse panopticon.”3 Each of these metaphors emphasizes the interrelatedness of accountability and transparency-promoting tools or arrangements, which include structural features (separation of powers, for instance), political factors (for example, partisan rivalries), and institutional or administrative tools (such as FOIA or the IGs). The spatial dimension of these metaphors matters. Those who seek information, those who disseminate it, and those who try to prevent its dissemination all come from distinct nodes in a web of actors, and their perspectives and interests have consequences for the meaning of the information that is released. This spatial dimension is precisely what pairs FOIA and the IGs as complementary tools operating from different positions vis-à-vis the executive: FOIA serves as an external prompt for the public release of governments records, whereas IGs ferret out information from the inside, as full members of their host agencies responding, in many cases, to internal requests and complaints for investigation. Both tools are directed at a very specific body of information. They both target executive branch agencies but omit the president, Congress, and the courts; FOIA also omits some components of the Executive Office of the President. Both have the capacity to influence the way government-generated information is presented to the public and which audiences will have the most ready access to it.
The relationship between FOIA and the IGs goes beyond functional similarities; they also play mutually reinforcing, or mutually corrective, roles. IGs have both a direct and an indirect effect on the way FOIA works. They can encourage agencies to make FOIA more of a priority and check that an agency is following FOIA regulations. They also can influence how documents are classified. In the 2009 Reducing Over-Classification Act, the 111th Congress instructed IGs to review the classification process and “to identify policies, procedures, rules, regulations, or management practices that may be contributing to persistent misclassification.”4 (FOIA staff themselves name overclassification as a significant barrier to timely fulfillment of requests.5) IG involvement in the FOIA process can thus affect not only the range of what can be effectively requested through FOIA but the very standards by which classification decisions are made in the future. Requests for full IG reports, moreover, have been made regularly through FOIA when these documents have been partially redacted. Because FOIA and the IGs operate in this shared accountability space, they are both subject to, and in turn influence, the common practices affecting all government transparency initiatives.
Despite the potential of these tools to improve democratic governance, demands for transparency can come with pitfalls. Indeed, the very tools designed to enhance transparency create dynamics and incentive structures that, in practice, might undercut the advantages of a transparent government apparatus. First, monitory instruments such as FOIA and the IGs require substantial administrative capacity to work as instruments of accountability, and the costly bureaucratic apparatus that they demand can undermine the very streamlined, democratized administration they are intended to promote. Second, despite the formal availability of government records and reports through FOIA and the IG system, accessing many such documents often requires expertise to navigate the process, allowing corporate and other private interests to benefit disproportionately from the laws. Moreover, many government documents require expertise to interpret, which privileges certain actors and skills. A final paradox revolves around expectations: if in theory a transparent, accountable government should inspire trust and confidence on the part of the citizenry, in reality the monitory tools designed to achieve that transparency can raise false hopes of what government (and grassroots activism) can realistically achieve and might ultimately lead to disappointment, dissatisfaction, and distrust. Although in some ways furthering the values of transparency and democracy, the emergence of these interconnected tools must be viewed cautiously in light of the democratic paradoxes they pose.
THE BUREAUPATHOLOGY PARADOX
One of the immediate problems posed by demands for accountability and transparency is the practical issue of how and by whom they will be implemented. The IG system and FOIA require substantial bureaucratic resources—staff, reliable funding, and expertise—to carry out their mandates, and funding has grown steadily for both tools since the 1970s. Yet material growth, which induces administrative complexity, can complicate calls for transparency and accountability, both of which are enabled by organizational simplicity and clarity. The growth of the IGs and FOIA comes, then, as a form of bureaupathology, the phenomenon of organizational dysfunction that results from excessive bureaucracy.6 According to one version of this logic, the government expands in size (girth) and in thickness (layers), multiplying and obfuscating lines of accountability and making transparency more difficult to achieve; further reforms meant to promote accountability contribute to a vicious cycle by making the government still bigger and less intelligible.
Part of the bureaupathology paradox results from sheer bureaucratic bulk. The Inspector General Act of 1978 led to the establishment of an army of IGs at the federal level, with the initial set of twelve IGs expanding to seventy-two by 2017. Each federal department and agency now has an IG and attendant office, often with hundreds of staff; altogether, the federal IG system adds over 13,000 personnel to the ranks of the government.7 Thousands more IG personnel exist at the state and local level. Although these offices have no doubt reduced bureaucratic waste in many instances (according to their own estimates, bringing a potential $14 return on each dollar they spend8), they also add to the size and complexity of a bureaucracy that they were in the first place designed to rationalize and hold to account. In theory, IGs could diminish administrative efficiency and accountability by increasing government outlays and contributing to additional regulations. Indeed, both government and academic reviews of IG work in the early 1990s found evidence of IG-induced red tape.9 Subsequent reforms, including the Clinton-era Reinventing Government initiative, which sought to redefine IGs to have a more collaborative role, and the 2008 Inspector General Act, which strengthened their independence and broadened their investigative powers, have curbed this tendency. The IG community as a whole produces many consequential reports, but expansion of the relevant statutes has bred further strands of bureaupathology as well. The 1988 Inspector General Act reform, which added IGs at more than thirty smaller agencies, or designated federal entities (DFEs), did not provide its new cohort with the same powers of investigation as “establishment IGs” (IGs in large departments, appointed by the president and confirmed by the Senate).10 Because DFE-IGs are selected by the department head and not subject to external scrutiny, their independence is compromised and their work often ineffective, reproducing precisely the bureaucratic redundancy they were asked to overcome.11
FOIA also adds a notable layer to the bureaucratic apparatus, and this apparatus continues to grow. To sketch a quick portrait of its material impact: the U.S. federal government saw the largest number of total FOIA requests then on record in 2014, with 714,231 requests made to the one hundred-odd agencies subject to the legislation; in 2015, the number was nearly as high at 713,168.12 Moreover, 14,274 appeals were processed across the federal government, and 4,121.59 full-time FOIA staff were employed in fiscal year 2015.13 The total estimated cost of administering the act was $480,235,967.62 that year, and less than 1 percent of this cost was recovered in FOIA fees.14 (Beyond their merely nominal contribution to the operating costs of FOIA, processing fees were criticized by FOIA staff as “labor intensive and hard to collect” in one government-wide survey.15) FOIA officers at many agencies deem the current level of resources insufficient to respond to requests in a thorough and timely manner.16 Recent efforts to improve the act’s implementation and efficacy, such as the Obama administration’s creation of a Chief FOIA Officer (CFO) in each agency in 2009, have arguably created new bureaucratic layers that do little to facilitate the processing of requests and that, moreover, have provoked negative reactions by agency staff.17
In short, bureaucratic bulk can be troubling for transparency because the greater the number of layers of government, the more difficult it is to trace decision-making processes and to pinpoint responsibility when things go wrong. But bureaupathology is not merely about bureaucratic bulk. It also can result from unintended consequences of bureaucratic structures that generate effects opposed to the original purpose of the organization. For instance, the dual reporting requirement of the original Inspector General Act, which divides IGs’ loyalty between Congress and their host department in order to ensure their independence, has limited the degree to which they can engage a broader public. These statutory provisions have served to shape the norms of the profession itself, yet in some cases they may limit the ability of IGs to effect oversight. While some IGs do work closely with Congress or the media, these IGs are in the minority, and IGs who share too much information outside of their host department—either through early releases of reports to the press or through individual “tell-all” memoirs—are controversial within the community.18 IGs who are too close to Congress find their host departments less willing and cooperative during investigations, limiting their effectiveness.
Commitment to the principles of open government, including proactive disclosure, requires administrative capacity. IG reviews, for example, must be made freely available on government websites, and all Offices of Inspectors General (OIGs) are required to maintain a substantial, proactive online presence by posting press releases and videos of congressional testimony, sending email updates, and announcing (on the Internet and by tweet) the release of each report. By making their work public—indeed, by increasingly considering the public to be a primary audience—this set of practices provides the narrative raw material with which third parties can pursue political (rather than administrative) demands for accountability. As they mature, such practices may encourage citizens and the media to participate more actively in the process of accountability and may indeed enhance American democracy. But the technological infrastructure required to make IGs’ work traceable and open to both government officials and the wider public has yet to be fully constructed. In a December 2015 hearing assessing the IGs’ effectiveness and transparency, Senator Heidi Heitkamp (D-ND) observed that “the current set-up right now doesn’t allow Congress or the public to track outstanding recommendations or recommendations over time,” prompting the creation of a database of IG recommendations to make their work (and that of their host agencies) more transparent.19 Despite expressing support for the idea, the Council of Inspectors General on Integrity and Efficiency’s Chairman and Department of Justice IG Michael Horowitz commented that “it takes a fair amount of staff to do that kind of work and information technology infrastructure. We do not think we even have the IT capacity to do that at this point.”20
Like the IG mandate, amendments to FOIA have made the interpretation and proactive dissemination of information a government responsibility as well. Providing access to records makes it incumbent upon the government to maintain the physical and digital infrastructure with which to make this access permanent, managed, and possible. The E-FOIA Amendments of 1996, for instance, require agencies to maintain an online reading room of standard agency documents (such as rules of procedure, statements of policy, and even positions on legal matters), as well as preemptively to make available records for which they anticipate receiving multiple requests. Although Congress has ordered agencies to set up “procedures for identifying records of general interest or use to the public that are appropriate for public disclosure, and for posting such records in a publicly accessible electronic format,” David Pozen notes that these novel responsibilities carry with them the possibility of inattention, lack of compliance, and information “overload” to the public.21
The pursuit of transparency has thus occasioned the emergence of a new, inescapably political role for government officials in both IG and FOIA offices: that of proactive informant. A proactive transparency strategy in the U.S. government is still in its infancy. Cultivation of this role might ultimately provide a corrective to the pitfalls now faced by the current, largely reactive FOIA-based transparency strategy, but it will require careful attention to how release decisions are made by bureaucrats.22 Achieving a balance between public relations impulses and a commitment to democratic integrity will depend in part on the ability of extra-governmental actors—public intermediaries—to discipline the discretion of proactive informants and to parse the significance of the documents that emerge.
THE EXPERTISE PARADOX
FOIA is frequently hailed as a tool of citizen empowerment. Although the IG system was not intentionally designed to promote citizen involvement in bureaucratic oversight, IG reports can also provide the public with the raw information with which to mount grassroots political action in the face of bureaucratic misdeeds. However, the often-specialized information released by IGs demands interpretation and, frequently, technical and contextual knowledge to render it politically meaningful. Despite its facade as a popular tool available to any citizen, FOIA requires different kinds of expertise to be used successfully and, as a result, it has been used disproportionately by different constituencies. Expertise is not antidemocratic; democracies need expertise, both within government and outside of it, as an aid to and check on political authority.23 But ensuring that expertise does not come into conflict with democratic principles requires an awareness of the potential for novel forms of expertise to create unintended distributional inequalities. It also requires protections for the independence of experts in government from bureaucratic or political capture. In short, the questions of how state information is framed, translated, and narrated to the public, and who is performing those tasks, are of crucial importance for understanding contemporary democratic processes.
Expertise is required at two moments in the process of seeking information through FOIA: in the access stage (the moment of filing a request for documents with an agency) and in the interpretation stage (understanding the content and significance of the records). At the access stage, FOIA is formally available to any person. All agencies have FOIA offices and websites with guidelines to facilitate making a request. But making effective FOIA requests often demands money and expertise, including the ability to pay agency fees (or the knowledge of how to have these fees waived) and the ability to appeal (and litigate, if necessary) if and when agencies prove unresponsive.24 Corporate requesters operate in this space with the advantage of administrative acumen and, often, with the legal experience and resources to navigate a long and obstacle-ridden process. Successful FOIA requests also frequently require certain kinds of insider knowledge. Seth Kreimer observes that “for FOIA requests to generate illuminating documents, they must be precisely framed, and framing such requests requires knowledge regarding the activities to be illuminated.”25 This is certainly the case with “deep secrets” (the existence of government programs unknown to the wider public) but also, more broadly, for many government programs associated with national security.26
An immediate, but unintended, consequence of this premium on insider expertise is a skewed distribution of the constituencies that use FOIA. Margaret Kwoka demonstrates that commercial organizations make the bulk of FOIA requests at many agencies.27 While corporate concerns are not necessarily at odds with the public interest, Kwoka’s research suggests that their requests often focus on enhancing investor knowledge or unearthing information about competitors.28 These requests advance the goals neither of government oversight nor of increasing public awareness of government activity.29 In practice, FOIA’s structure creates significant inequalities of access to, and benefits from, government information.
Grassroots and public interest groups have taken measures to try to reduce some of the barriers to information access that ordinary citizens experience. The National Security Archive’s guide for making effective FOIA requests, for example, includes detailed descriptions of how to read government documents, how to develop a strategy for document acquisition, and how to interpret records once they are produced.30 Activists have developed technological innovations such as FOIA Mapper, a tool that assists citizens in crafting FOIA requests. Its inspiration arose when “years of FOIA requests taught [the designer Max] Galka that there was no easy way to determine which organization had the information he wanted,” and, as a result, he tried to plot the way that government agencies store data.31 Both the National Security Archive and the FOIA Mapper facilitate new forms of democratic practice and public mediation that target a second-order dimension of information access. Rather than making records themselves accessible, they create and disseminate explanations and “maps” to guide access. The degree to which these efforts can ultimately mitigate the structural inequalities generated by FOIA is yet to be seen.
In the second moment of information dissemination—interpretation—two types of bureaucratic peculiarities surface that affect the success of monitory tools in delivering and translating state information to citizens. First, the ability to make sense of a government document requires knowledge of how to read certain kinds of reports, as well as contextual knowledge of the agency (or event) in question. Even among people with legal and political competence, a gap exists between those who can access government data and those who can understand what the data mean. Illuminating here is Adam Candeub’s insight that information is only transparent—in the sense that it helps citizens uncover influence in the public decision-making process—if it can be “computed” and translated into a set of meaningful observations.32 If the most incriminating piece of information about a government program is embedded in an audit or technical report, citizen action will depend on having access to the relevant technical expertise as much as access to the document itself.
The advent of “big data” deepens this challenge. With the mounting reliance on advanced computational methods by government agencies for policy design, the capacity of third-party actors to trace the decision-making process increasingly demands an understanding of how to “read” and analyze large data sets. This skill can also indirectly affect the ultimate impact of IG reports. The Sunlight Foundation, a transparency advocacy organization, initiated the construction of an IG report database on the ground that “many uses of government data require first obtaining all the data.”33 IG reports are posted on the respective websites of individual OIGs, but there is no central repository for all such reports. Systemic analysis of government activity can thus demand more than an ability to read and parse a report; increasingly, it requires both the technological infrastructure and database-related expertise with which to conduct wide-ranging searches for information.
Big data poses challenges for IGs as well. Some OIGs have successfully used data analytics to oversee their host departments’ data, but most have struggled to attract staff with the necessary skills and to build technological infrastructure. In many cases, inert bureaucratic culture and high investment costs have prevented individual OIGs from adopting these methods; in others, departmental reluctance to grant IGs access to data sets has stymied their efforts.34 The Inspector General Act as it currently stands does not explicitly permit access to data systems and thus provides an opportunity for agencies to subvert the oversight process.
Big data itself is not ipso facto less transparent than ordinary data; on the contrary, Candeub argues that “big data and decentralized data processing can in fact render the government decipherable in key ways, and indeed, seems necessary given the increased government electronic surveillance of its citizens.”35 Just as big data facilitates the government’s surveillance capacities, it can facilitate citizen awareness and scrutiny. Nonetheless, the use of big data in government policy making poses new challenges for the interpretation and accessibility of data, and it will lead to further inequalities between citizens with different interpretive capacities, privileging those with statistical, auditing, and computing expertise. Joshua Tauberer, a prominent transparency advocate, offers a useful illustration of the need for particular kinds of expertise in the translation and interpretation of big data. Describing the winner of the 2002 Pulitzer Prize for investigative reporting, a fourteen-part series detailing the child mortalities that resulted from inattentive social services in Washington, D.C., he writes:
The series could not have been told without access to government records at scale…. The value of those government records came from reporters’ skills in turning the records, and of course interviews, into something pointed, understandable, and actionable for their readers. Put another way, the knowledge that the [Washington] Post’s readers gained from the 14-part series could not have been FOIAs from the government directly. The knowledge came from skilled synthesis by mediators who took a large quantity of raw data material and produced a completely different information product for consumers.36
Tauberer’s narrative demonstrates both the challenges and promise of big data in furthering a transparent and accountable government. Analysis of big data enabled the force of the story, but that very analysis required a novel expertise in how to marshal and interpret information. While the press has always played a role in interpreting and narrating the significance of government documents and actions, the new skills required for this mediating role now privilege not only “traditional journalists, issue advocates, and organizers, but also app builders [including] programmers, statisticians, designers, and entrepreneurs.”37
IGs are responsible for the production and dissemination of many internal audits and investigations, but the fruits of their labor have long been decried as being at best boring and at worst incomprehensible.38 Often, citizens find IG reports difficult to parse, as the conflicting interpretations of the State Department IG’s June 2016 review of Hillary Clinton’s emails suggested.39 Understanding these reports requires knowledge of the scope of IG investigative authority (why their investigations may go only so far), how the severity of the offense is framed and articulated in IG reviews (often as rule or policy infractions, without judgment as to their broader significance), and the kinds of evaluative statements IGs can and cannot make (they are constrained, for instance, by strong norms against speculation). IG reports need not be quite so inscrutable. Indeed, periodic innovations in how IGs frame their reports have strengthened their influence and impact. Justice Department IG Michael Bromwich, for instance, introduced the “special review” during his tenure in the 1990s to complement the statutorily mandated semi-annual reviews; these reviews permitted an extended treatment of problems that did not conform to existing IG categories of investigation. More recently, the Special Inspector General for Afghanistan Reconstruction has introduced “interactive,” mobile-friendly, online reports, and managed to attract a larger readership in both Congress and the public.40
In sum, the disclosures made by FOIA officers and IGs are hailed as moments of publicity, but members of the public are often unable to do much of anything with these disclosures without the help of public intermediaries. Reading an IG report, or any government audit, requires knowledge and understanding of the categories developed by auditors, as well as a familiarity with the tone and scope of such reports. Mining the vast data sets that increasingly serve as the basis for policy formation requires statistical know-how and awareness of what data have been collected by government. In its current incarnation, FOIA facilitates the dissemination of event-specific information; recognizing long-term trends and patterns requires the collection and analysis of data that transcend single events or points in time. Narratives and tools developed by public intermediaries serve as crucial bridges to render actionable the fruits of these monitory instruments. Although journalists have traditionally been best poised to play this role, the necessary skills with which to make such information intelligible to citizens have evolved to include a host of specialized technical competencies in addition to basic political judgment and narrative proficiency.
THE EXPECTATIONS GAP PARADOX
New tools generate excitement. As innovations of the postwar era, FOIA and the IG system have been promoted under the banner of transparency as correctives to the pathologies of the administrative state, and each carries with it implicit promises of better, more democratic, and more efficient government. The IG system was born alongside the Government in Ethics Act and the Civil Service Reform Act, and it came with the promise of restoring integrity to a government tarred by Watergate and a spate of bureaucratic fraud and embezzlement scandals in the preceding decade. FOIA, signed into law on Independence Day (July 4), 1966, inspired tremendous hope on the part of transparency and accountability advocates. Yet abstract promises can lead to unrealistic expectations that undermine the successes that monitory instruments do achieve. In one government-wide survey, FOIA staff at multiple agencies overwhelmingly identified the “unrealistic expectations of requesters [as] a significant problem,” suggesting that some FOIA resources be directed toward hosting meetings or panels to educate requesters.41 Managing the public’s expectations of what kinds of information exist and what can be reliably collected and organized in response to a particular request for records presupposes common expectations on the part of the government and the public.
Among other expectations, more transparency should, in principle, lead to more public trust because we (as citizens) can see which actions have been taken, by whom, and why. Yet even as FOI laws have multiplied and efforts toward greater transparency have intensified worldwide, a marked decline of societal trust has been identified, in various guises, by Robert Putnam, Onora O’Neill, and a spate of polls.42 This decline in trust has been attributed in part to the rise of demands for transparency and accountability. The results of transparency measures often show governments to be less than fully efficient or honest, giving citizens further reason to be suspicious of public officials. In her Reith lectures on trust, O’Neill observes:
Increasing transparency can produce a flood of unsorted information and misinformation that provides little but confusion unless it can be sorted and assessed…. And unless the individuals and institutions who sort, process and assess information are themselves already trusted, there is little reason to think that transparency and openness are going to increase trust.43
Precisely because many documents released by FOIA are “unsorted” (that is, their meaning unexplained and not contextualized), their significance will depend in part on how they are released and narrated within the public sphere. O’Neill is not opposed to transparency itself. Rather, she is critical of the cultural practices that have grown in its name but that fail to bring about truly intelligible information. She condemns the rise of “audit culture,” in which “increased demands for control and performance, scrutiny and audit have been imposed, and in which the performance of professionals and institutions has been more and more controlled.”44 She laments the decline of thoughtful, critical evaluation and worries that the unrealistic expectations established by the “accountability revolution” will fuel the same sense of mistrust it was designed to address.
Although O’Neill does not write of either directly, she identifies a crucial contradiction at the heart of monitory instruments such as FOIA and the IGs. Whereas both tools ostensibly bolster the government’s accountability to the public, she argues that accountability reforms in practice reinforce institutions’ accountability to “regulators, to [other] departments of government, to funders, and to legal standards [which] … impose forms of central control—quite often indeed a range of different and mutually inconsistent forms of central control,” prompting the question of whose interests are primarily served by such reforms as the introduction of the CFO in FOIA offices or the centralized control of IG performance indicators by the Council of Inspectors General for Integrity and Efficiency.45 To be sure, these alternative loci of accountability are hardly antidemocratic and, if coupled with well-developed standards of professionalization, could be true sources of public accountability. O’Neill’s analysis, however, underscores the possible value conflict that arises if performance indicators are chosen that reflect, say, a bias toward efficiency in an agency in which civil rights violations occur that would not turn up on efficiency measures.
Yet even O’Neill struggles to overcome one paradox of the logic of transparency. Although critical of audit culture in general, she suggests that citizens “energetically” rely on proxy evidence of trustworthiness—such as the analysis of “auditors, examiners, regulators, evaluators, peer reviewers,” and other experts—when assessing the more complex activity of public officials.46 Paradoxically, this might imply a relocation of trust from what O’Neill calls “informed and independent evaluation” to the “performance indicators” that are symptomatic of the audit culture pathologies of which she is suspicious.47 For instance, evaluations of IGs that focus on simple quantitative outcome measures, such as the number of criminal indictments, to the exclusion of broader political processes in which the IGs are embedded can miss the significance of IG reviews that contribute to long-lasting solutions engineered by Congress. There is a profound difference between the multidimensional standards reflective of democratic integrity relied upon by informed evaluation (which can take the form of an audit or peer review) and simple managerial benchmarks. But the difference can easily be blurred in practice when both methods are used by the experts to whom O’Neill points. Despite her enthusiasm for authoritative proxies in which to place our trust, she concedes that even their analyses can be too complex for the average citizen to comprehend and to use for meaningful political participation.
There is a further danger in delegating public judgment to the fruits of audit culture. The evaluative categories and language of auditors and regulators can be at odds with the primary democratic values that transparency is, in theory, supposed to promote. For instance, audits and investigations that evaluate the efficiency of a particular public program or institution do not always take into account the unquantifiable social and political goods associated with it. Schools serving disadvantaged communities might score low on performance assessments, but might provide community support or engender social capital in ways that national assessment and audit systems fail to measure. Similarly, the No Child Left Behind Act, in an effort to boost educational performance and hold educators to account, created incentive structures that led teachers to teach to tests and even to cheat to adhere to inflexible performance standards.48 Unrealistic expectations about what teachers (and students) can achieve, coupled with enforcement of those expectations through accountability mechanisms such as the demands to achieve quantifiable “Adequate Yearly Progress” and to make performance measures publicly available, have arguably placed struggling schools in cycles of failure from which they cannot emerge.
Unrealistic expectations about transparency on the part of the public might hamper the democratic potential of FOIA and the IG systems, but unrealistic expectations on the part of other public officials can also limit their political effects. Both because of a lack of visibility (even within government) and a vague mandate with multiple principals, IGs operate within an ambiguous framework. They report both to Congress and to their agency head; they serve a range of different audiences, including their host agency, Congress, the president, and the public; and they are tasked with both auditing and investigation, which demand different bureaucratic cultures and expertise. This ambiguity has some benefits for IG independence. It has permitted some IGs to push the boundaries of their position, sometimes privileging certain audiences over others, or concentrating on certain types of wrongdoing at the expense of others. However, the lack of clarity about the IGs’ role and their varied performance has led elected officials to expect more from IG scrutiny than is necessarily realistic given budget and time constraints. In congressional hearings, members of Congress frequently demonstrate ignorance of the scope of IG authority when questioning IGs on their investigations (for instance, about what they can and cannot investigate, and whether or not they are part of the law enforcement community). Illustrative of these false expectations is an exchange between Senator Arlen Specter and Justice Department IG Glenn Fine regarding the IG’s scope of authority during a 2003 hearing on the detention of terrorist suspects. After Specter pressed Fine on why he had not pursued certain lines of investigation, Fine responded with an emphatic insistence that “litigation decisions by department attorneys are actually not subject to the Inspector General’s authority.”49 Unlike other IGs, the Justice Department IG is excluded from overseeing legal matters in its host department, which are deemed the responsibility of the Office of Professional Responsibility. Although the Justice IG is the only establishment IG to have this exclusion under the Inspector General Act, many OIGs command a unique constellation of powers and responsibilities. Incorrect understanding of what each IG can do to hold the executive branch to account can serve to fuel partisan spats rather than inspire genuine administrative or political reform.
Both FOIA and the Inspector General Act feed into and suffer from the dynamics of the expectations gap paradox. The appearance of transparency is fueled by the existence of FOIA and the IGs, and in many ways these frameworks do serve the twin goals of transparency and accountability. But excessively high (and vague) expectations on the part of both the public and public officials about the kind of government these institutions might enable leads to inevitable disappointment, just as the release of incriminating records and reports further sows the seeds of citizen mistrust.
CONCLUSION: TRANSPARENCY AND PUBLIC INTERMEDIARIES
Discussing the feminist movement, Hannah Arendt once asked, “What will we lose if we win?”50 Her question was meant to underscore the potential unintended consequences of pursuing what seemed like a primary political goal. The question can be transposed into the domain of transparency: what do we lose if we succeed in building a transparent society? Do the unintended social and political dynamics of these methods outweigh, or compromise, the benefits of transparency? The pursuit of transparency through the use of monitory mechanisms such as FOIA and the IGs may advance one vision of transparency but simultaneously undercut other democratic values.
Although analytically distinct, the bureaupathology, expertise, and expectations gap paradoxes are not separate empirically. The cost and bureaucratic bulk required to maintain a transparency regime contribute to the regime’s visibility and thus to its promise of efficacy, yet they also create opportunities for privileged economic groups to dominate the process of accessing information and, to some degree, to control how that information is used. Administrative expertise in requesting records, and audit and statistical expertise in interpreting them, play a role both in creating differential access and in contributing to a transformation in the dynamics of trust. The volume of information and emerging forms of technical expertise that accompany these transparency strategies can render government practices unintelligible and further undermine basic public comprehension of the workings of government.51 Both FOIA and the IGs fall prey to these paradoxes.
But if these tools of transparency and accountability have the capacity to undermine public trust and democratic integrity (which includes values potentially at odds with demands for administrative efficiency, such as respect for civil rights), they also may contain the seeds of rectification. O’Neill offers narrative as a corrective for cruder, or more quantitative, measures of accountability, and this is where the promise of both IGs and the media lies in making the relationship between FOI laws and trust a positive one. The need for legitimate narratives with which to make sense of government data will only continue to grow. IG and FOI mechanisms can, if properly structured and supported by the news media, help to overcome each other’s limitations in this regard. IGs can make agency adherence to FOIA regulations a priority in their oversight and recommendations and can, moreover, bring to light the existence of documents that might be publicly unknown. We might also see a seed of promise in the ambiguity of the IG mandate: because of the gap between what Congress wants the IGs to do and what they can do, there is much scope for reform and new deployment. When IGs’ work is effectively translated and narrated to the public, their professional status as impartial auditors and investigators lends legitimacy to their narratives.
The question of which democratic values are more or less served by transparency is not merely an abstract question; it is a choice that societies must continually make. Although the question itself deserves public deliberation, only a concerted effort to encourage different kinds of public intermediaries will mitigate the unintended and paradoxical consequences of transparency-enhancing practices. In discussions of transparency tools, the focus should be not only on access to information but also on the processes by which that information is translated to the public and made politically meaningful. All three paradoxes discussed in this chapter point to the need for public intermediaries—both governmental and nongovernmental—to make these tools truly democratic. Such intermediaries include journalists skilled in working with big data and contributors to the technical infrastructure and conceptual mapping that permit analysis of government data at scale.
The success of representative democracy depends crucially on ordinary citizens’ understanding of the actions of public officials. Monitory instruments, broadly speaking and taken as a whole, have emerged as a dominant contemporary strategy to produce this knowledge. Responding to the legitimation and trust crises experienced by governments worldwide demands a web of skilled public intermediaries who can exploit these powerful new instruments for the good of the many rather than the few.
NOTES
This chapter has benefited from feedback received at the City University of London Department of International Politics’ faculty research seminar and at the Rothermere American Institute’s (University of Oxford) Graduate Politics seminar. In particular, detailed and thoughtful comments from Mark Erbel, Iosif Kovras, Neil Armstrong, David Pozen, and Michael Schudson were invaluable.
1. See generally John Keane, The Life and Death of Democracy (London: Simon & Schuster, 2009).
2. Pub. L. No. 95−452, § 2, 92 Stat. 1101 (1978). Regarding IGs’ independence and their relationship with agency management, see Sherman Funk, “Dual Reporting: ‘Straddling the Barbed Wire Fence,’” Journal of Public Inquiry (Fall 1996): 13; and Council of the Inspectors General on Integrity and Efficiency, “IG Authorities,” July 14, 2011, 3.
3. See Kathleen Clark, “The Architecture of Accountability: A Case Study of the Warrantless Surveillance Program,” BYU Law Review (2010): 357–419, at 377–81; Seth F. Kreimer, “The Freedom of Information Act and the Ecology of Transparency,” University of Pennsylvania Journal of Constitutional Law 10 (2008): 1011–80; David Brin, The Transparent Society (New York: Basic Books, 1999).
4. Pub. L. No. 111–258, § 6(b); 124 Stat. 2651, quoted in Wendy Ginsberg and Michael Greene, “Federal Inspectors General: History, Characteristics, and Recent Congressional Actions,” Congressional Research Service Report R43814, June 2, 2016, 11, https://fas.org/sgp/crs/misc/R43814.pdf.
6. See Edward T. Giblin, “Bureaupathology and the Denigration of Competence,” Human Resource Management 20 (1981): 22–25; “Bureaupathology,” in Oxford Dictionary of Business and Management, 5th ed., ed. Jonathan Law (Oxford: Oxford University Press, 2009).
7. Council of the Inspectors General on Integrity and Efficiency, Progress Report to the President, Fiscal Year 2015 (Washington, D.C.: CIGIE, 2016), 2.
8. Council of the Inspectors General on Integrity and Efficiency, Progress Report to the President, Fiscal Year 2015.
9. See Albert Gore and National Performance Review, From Red Tape to Results: Creating a Government That Works Better and Costs Less (Washington, D.C.: Government Printing Office, 1993); Paul Light, Monitoring Government: Inspectors General and the Search for Accountability (Washington, D.C.: Brookings Institution, 1993), 223.
10. Pub. L. No. 100–504, title I, §§ 102(a)–(d), (f), (g), 104(a), 105–107, 109, 110, 102 Stat. 2515–2529 (1988).
11. Personal interview with David C. Williams, Inspector General of the U.S. Postal Service, May 23, 2017.
12. U.S. Department of Justice, Summary of Fiscal Year 2015 Annual FOIA Reports (Washington, D.C.: Department of Justice, 2016), 1–2.
13. U.S. Department of Justice, Summary of Fiscal Year 2015 Annual FOIA Reports, 19–20.
14. U.S. Department of Justice, Summary of Fiscal Year 2015 Annual FOIA Reports, 20.
15. Citizens for Responsibility and Ethics in Washington, “FOIA at the Mid-Term,” 6.
16. Citizens for Responsibility and Ethics in Washington, “FOIA at the Mid-Term,” 3.
17. Citizens for Responsibility and Ethics in Washington, “FOIA at the Mid-Term,” 3–4.
18. Personal interviews with David C. Williams, May 26, 2017, and Michael Bromwich, Department of Justice Inspector General, 1994–1999, January 3, 2013.
19. U.S. Congress, Senate Committee on Homeland Security and Governmental Affairs, Subcommittee on Regulatory Affairs and Federal Management, Implementing Solutions: The Importance of Following Through on GAO and OIG Recommendations, S. Rep. 114-265, 1st Sess. (2016), 3.
20. U.S. Congress, Senate Committee on Homeland Security and Governmental Affairs, Subcommittee on Regulatory Affairs and Federal Management, Implementing Solutions, 16.
21. FOIA Improvement Act of 2016, Pub. L. No. 114–185, § 4, 130 Stat. 538, 544 (to be codified at 44 U.S.C. § 3102(2)); David E. Pozen, “Freedom of Information Beyond the Freedom of Information Act,” University of Pennsylvania Law Review 165 (2017): 1097–1158, at 1151–52.
22. On the fundamentally “reactionary” nature of FOIA and on alternative strategies to transparency, including proactive disclosure, see Pozen, “Freedom of Information Beyond the Freedom of Information Act.”
23. See Michael Schudson, Why Democracies Need an Unlovable Press (Cambridge: Polity, 2013), chap. 10.
24. Kreimer, “The Freedom of Information Act and the Ecology of Transparency,” 1020.
25. Kreimer, “The Freedom of Information Act and the Ecology of Transparency,” 1025.
26. See David E. Pozen, “Deep Secrecy,” Stanford Law Review 62 (2010): 257–339.
27. See Margaret Kwoka, chapter 4 in this volume; and Margaret B. Kwoka, “FOIA, Inc.,” Duke Law Journal 65 (2016): 1361–1437.
28. Kwoka, “FOIA, Inc.,” 1414.
29. Other studies over time bear out Kwoka’s empirical claim. See Government Accounting Office, Government Field Offices Should Better Implement the Freedom of Information Act, LCD-78-120 (1978), 37; John E. Bonine, “Public-Interest Fee Waivers Under the Freedom of Information Act,” Duke Law Journal (1981): 213–78, at 216–17; Mark Tapscott and Nicole Taylor, Few Journalists Use the Federal Freedom of Information Act: A Study by the Center for Media and Public Policy (Washington, D.C.: Heritage Foundation, 2001); and Coalition of Journalists for Open Government, “Frequent Filers: Businesses Make FOIA Their Business,” July 3, 2006, www.spj.org/rrr.asp?ref=31&t=FOIA.
32. Adam Candeub, “Transparency in the Administrative State,” Houston Law Review 51 (2013): 385–416, at 403.
34. Personal interview with David C. Williams, May 23, 2017.
35. Candeub, “Transparency in the Administrative State,” 412–13.
37. Tauberer, “Open Government, Big Data, Mediators.”
38. Daniel L. Feldman and David R. Eichenthal, The Art of the Watchdog: Fighting Fraud, Waste, Abuse, and Corruption in Government (Albany, N.Y.: SUNY Albany Press, 2013), 115.
39. For two contrasting interpretations in the mainstream media, see Rosalind S. Helderman and Tom Hamburger, “State Dept. Inspector General Report Sharply Criticizes Clinton’s Email Practices,” Washington Post, May 25, 2016; and Jeffrey Marburg-Goodman, “Correcting the Record (Again): Hillary Clinton’s Handling of State Department Emails,” Huffington Post, June 7, 2016, www.huffingtonpost.com/jeffrey-marburggoodman/correcting-the-record-again_b_10323830.html.
41. Citizens for Responsibility and Ethics in Washington, “FOIA at the Mid-Term,” 5.
42. On the decline of public trust, see, among others, Robert D. Putnam, Bowling Alone: The Collapse and Revival of American Community (New York: Simon & Schuster, 2000); and Onora O’Neill, A Question of Trust: The BBC Reith Lectures 2002 (Cambridge: Cambridge University Press, 2002).
43. O’Neill, A Question of Trust, 73.
44. O’Neill, A Question of Trust, 73.
45. O’Neill, A Question of Trust, 53.
47. O’Neill, A Question of Trust, viii.
48. Letter of Linda M. Calborn, Government Accountability Office, to Secretary of Education Arne Duncan, May 16, 2013, “K-12 Education: States’ Test Security Policies and Procedures Varied,” www.gao.gov/assets/660/654721.pdf; Anya Kamenetz, “When Teachers, Not Students, Do the Cheating,” NPR All Things Considered, September 29, 2014.
49. Statement of Glenn Fine, Lessons Learned—The Inspector General’s Report on the 9/11 Detainees, Committee on the Judiciary Hearing, S. Hrg. 108-257, June 25, 2003.
50. Joanne Cutting-Gray, “Hannah Arendt, Feminism, and the Politics of Alterity: ‘What Will We Lose If We Win?’” Hypatia 8 (Winter 1993): 35–54.
51. I thank David Pozen and Michael Schudson for making this point.