Safety and Accident Prevention

Homeland security and citizen response to emergency situations: a perspective on the need for a policy approach to information access

Charles Herrick

Published online: 8 March 2009 � Springer Science+Business Media, LLC. 2009

Abstract Open access to public information is a hallmark of American political culture; however, the terrorist attacks on and before September 11, 2001 have prompted a

reevaluation of how ‘‘freedom of information’’ should be balanced against the need for

enhanced homeland security. This essay begins with a summary of legislative and exec-

utive actions that have led to restriction of environmental and health-related information

formerly available to the public. Drawing on studies of disaster behavior, it is argued that

citizen responders may be significantly hampered by restriction of environmental and

public health-related information formerly available by means of public access web sites.

The Lasswellian policy decision process is examined to explore the basis for a balancing test for agencies contemplating restriction of information related to environmental and

public health-related threats. It is suggested that the Office of Management and Budget

(OMB) implement guidance for federal agencies in weighing decisions concerning the

public status of information.

Keywords Security � Public access � Policy � Decision process

Public access to environmental and public health-related information: background and context

Open access to public information is a hallmark of American political thought and culture.

As articulated by James Madison, ‘‘A popular government, without popular information, or

the means of acquiring it, is but a prologue to a farce or a tragedy or perhaps both.

Knowledge will forever govern ignorance, and a people who mean to be their own

governors, must arm themselves with the power knowledge gives’’ (Padover 1953, p. 346).

Over the past quarter century, the principle of informed consent or public right-to-know

has been codified through numerous articles of law and regulation, especially in arenas

such as public health, consumer protection, worker safety, and the environment.

C. Herrick (&) Stratus Consulting Inc., 1920 L Street, NW, Suite 420, Washington, DC 20016, USA e-mail: [email protected]

123

Policy Sci (2009) 42:195–210 DOI 10.1007/s11077-009-9081-7

Provision of public access to information about environmental and health-related risks

is a key component of the U.S. federal environmental protection policy regime. Industrial

manufacturing facilities, water and electric utilities, and other operations that use and store

hazardous chemicals such as chlorine, ammonia, and cyanide pose a serious threat to local

communities. The Environmental Protection Agency (EPA) has estimated that at least 123

plants store toxic chemicals that, if released through explosion, accident, or terrorist attack,

could result in a vapor plume that would put as many as 1 million people at risk. Drawing

on EPA data, the Department of Homeland Security estimates that there are approximately

3,400 facilities in which a ‘‘worst-case’’ release could potentially affect over 1,000 people

(National Journal Group 2001). In addition, thousands of smaller facilities utilize sub-

stances known to harm humans and the environment. For example, EPA regulates about

15,000 facilities under the 1990 amendments to the Clean Air Act because they produce,

use, or store specified chemicals that would pose substantial risk to human health if

released into the air. Understandably, people who live or work near such facilities are

concerned about chronic exposure, as well as risks posed by acute exposure associated with

a major discharge event. 1

Many U.S. environmental laws include provisions intended to enhance or enable public

access to information about environmental and health-related conditions and threats. This

information is useful to citizens in understanding chemical hazards in their communities

and in stimulating communication between industry, stakeholders, and the public to

improve emergency response strategies and tactics at the local level. Environmental and

public health-related statutes that incorporate information access and dissemination pro-

visions include the Comprehensive Environmental Response, Compensation, and Liability

Act (CERCLA), Pollution Prevention Act, Emergency Planning and Community Right-to-

Know Act (EPCRA), Safe Drinking Water Act, Clean Air Act, and the National Envi-

ronmental Policy Act. Hazardous chemical reporting under EPCRA is intended to provide

information to local governments and citizens who live in areas surrounding subject

facilities. The EPCRA requires that an emergency plan, Material Safety Data Sheets, 2

and

a hazardous chemical list be made available to the general public. Similarly, Section 112(r)

of the Clean Air Act requires companies of all sizes that use certain flammable and toxic

substances to develop a Risk Management Program, which includes a hazard assessment

that details the potential effects of an accidental release, an accident history of the past

5 years, an evaluation of worst-case and alternative accidental releases, and an emergency

response program that spells out procedures for informing the public should an accident

occur.

In the arenas of environment and public health, the principle of public right-to-know is

founded upon two compelling imperatives: (1) a need to inform local-level and citizen

1 In late 1985, more than 150 people required medical attention when a toxic chemical—aldicarb—was

released from a Union Carbide facility in Institute, West Virginia. Union Carbide’s officials failed to notify local authorities about the release event because they reportedly believed the gas would settle and remain within the plant’s perimeter. Local authorities were confused about what was happening, what substance was involved, and how to protect and treat citizens. More recently, an explosion at a hazardous waste management and disposal facility in Apex, North Carolina resulted in a massive fire and subsequent evacuation of over 17,000 residents. First responders were hampered because they did not know what types of chemicals were stored at the facility and were confused regarding how best to fight the fire and manage site access until the company provided an inventory of stored materials nearly 24 h after the onset of the fire. 2

Originally established under OSHA to help assure employee safety, Material Safety Data Sheets (MSDS) are printed forms that characterize a hazardous chemical or substance, including physical properties, hazards to personnel, fire and explosion potential, safe handling recommendations, health effects, fire fighting techniques, reactivity, and proper disposal (U.S. EPA 2006).

196 Policy Sci (2009) 42:195–210

123

deliberations concerning the appropriateness of near-by industrial activities; and (2) a need

to help local communities recognize, assess, and effectively respond to toxic release

events.

Post 9/11 information restriction

Data and information that can help citizens recognize and respond to environmental and

health-related risks can also be used by terrorists and other malefactors to plan and execute

attacks on communities and units of critical infrastructure. Because of this, terrorist attacks

on U.S. soil have prompted reevaluation of how ‘‘freedom of information’’ should be

balanced against the need for enhanced security. In recent years, this balance has swung in

favor of information control.

Under current operational guidelines, rules governing exemptions to the 1974 Freedom

of Information Act (FOIA) have been relaxed, enabling federal agencies to withhold

information that would previously have been released. An October 2001 memorandum

from Attorney General John Ashcroft directed agency heads to rescind the ‘‘presumption of

disclosure’’ that had been operative under previous administrations. Agencies are no longer

compelled to articulate a plausible scenario of harm, merely to assure that information

withholding decisions have a ‘‘sound basis’’ in legal reasoning (U.S. DOJ 2001). Under the

2002 Homeland Security Act, Congress added a broad exemption concerning information

on ‘‘critical infrastructure’’ to the range of data and information that federal agencies need

not disclose under the Freedom of Information Act (Tombs 2005; Sidel 2004). Similarly,

the Public Health Security and Bioterrorism Preparedness Act of 2002 stipulates that

certain types of public infrastructure information are exempt from disclosure requirements

under the federal FOIA (U.S. Congress 2002). In addition, the DOJ Information Security

Oversight Office (ISOO) instructed agencies to take appropriate steps to assure the security

of ‘‘sensitive but unclassified information’’ related to America’s homeland security (Pozen

2005). Federal efforts to restrict information releases are being mirrored at the state level.

In a 2006 study, the National Conference of State Legislators (NCSL) determined that 50

U.S. states and territories have enacted statutory provisions exempting various types of

critical infrastructure information from disclosure under FOIA.

In response to the 9/11 attacks and executive directives, federal agencies removed

thousands of documents from their web sites and classified many more. For example, an

article in the Bulletin of the Atomic Scientist cited a host of information sources ‘‘tempo- rarily removed’’ from agency web sites and/or public information reading rooms (Costner

2002). Examples of federal government information resources removed or restricted from

public access are described in Exhibit 1. As illustrated in Exhibit 1, many of the information

resources removed from public access were developed for use by citizens and community-

level stakeholders to better understand and deal with health-related environmental risks,

including catastrophic releases of hazardous substances due to natural disasters, techno-

logical failures, human error, and terrorist attacks. While agency actions appear to have

been informed by assessments of the potential sensitivity of publicly available information

resources, there is no documentation of efforts to weigh the restriction of information

against the foregone values and benefits associated with its release.

At the U.S. Department of Energy, information removals were formalized through

distribution of a department-wide directive to ‘‘review the operational information

accessible to members of the public and remove or restrict access, as appropriate, to

information that may be used to target the Department of Energy.’’ The directive went on

Policy Sci (2009) 42:195–210 197

123

to identify a number of potentially sensitive items, including emergency planning hazard

assessments, safety analysis reports, environmental impact statements, detailed site and

facility maps, photographs of facilities, and personal data on employees’’ (Costner 2002,

p. 1). However, the directive did not provide guidance and/or criteria for how managers

should frame this decision process or assess potential security risks against the social value

of the information disseminated.

Agency documentation and Congressional testimony suggest that the situation was

similar at the U.S. EPA. Shortly after 9/11, the chief of information systems in the U.S.

Environmental Protection Agency’s Region II office—which includes New York—wrote

in an email ‘‘If there are any sites that look like they would offer site specific information

about facilities, especially hazardous chemicals, water supplies, etc., we should break the

links.’’ To bolster this effort, EPA headquarters delivered an Agency-wide request in late

September asking those in the agency responsible for disseminating information to identify

potentially ‘‘sensitive’’ information, and particularly ‘‘resources which provide information

on chemicals, and/or location, and/or amounts, and/or impacts on the environment or

human health’’ (OMB Watch 2002, p. 1). The criteria outlined in EPA’s agency-wide

assessment are reproduced in Exhibit 2. As was the case for DOE, EPA did not perform a

formalized assessment of public values or mission objectives affected through the decision

to restrict information.

As mentioned earlier, many of the removed or restricted information items were orig-

inally complied and intended for use by citizens dealing with toxic or hazardous release

events. To illustrate better how citizens use environmental and health-related information,

we next review recent sociological research on how individuals and groups react under

emergency and disaster conditions.

The realities of disaster response behavior

Much discussion of how the general public might respond after an emergency or disaster

event highlights the possibility of mass panic and social disorder. However, research on

population responses to a wide range of natural and technological disasters suggests a

Exhibit 1 Information resources removed from public access

The Environmental Protection Agency removed Clean Air Act, Risk Management Program information from its public access web site, including response plans to protect human health and the environment in the event of a release.

The Department of Transportation took down most of the national pipeline mapping system that enables communities to identify hazardous pipeline routes.

The Environmental Protection Agency removed a database known as ARCHIE, the Automated Resource for Chemical Hazard Incident Evaluation, which enabled users to evaluate how different chemicals behave under alternative release circumstances.

The U.S. Geological Survey instructed government libraries to destroy copies of a CD-ROM describing public water supplies, used by communities to protect source water.

The Department of Energy removed environmental impact statements that alerted local communities to potential dangers from nearby nuclear energy plants, as well as information on the transportation of hazardous materials.

The Department of Homeland Security released rules for implementing the National Environmental Policy Act (NEPA) that include categorical exemptions for the release of information dealing with disaster cleanup efforts.

198 Policy Sci (2009) 42:195–210

123

tendency toward reasoned adaptability and cooperation (Glass and Schoch-Spana 2002,

p. 218). A recent review by Lee Clarke confirms that people don’t tend to panic during

disaster situations (Clarke 2003). People may well be terrified, but ‘‘social bonds remain

intact and the sense of responsibility to others remains…strong’’ (Tierney 2003). Further, ‘‘an emergent norm process occurs [wherein] survivors share their tools, food, equipment,

and, especially, their time. Groups of survivors tend to emerge to respond to each others’

needs’’ (Fischer 2006).

The literature on disaster behavior also shows that disaster ‘‘victims’’ invariably con-

stitute the real first responders in most emergency situations. Studies indicate that

individuals and groups close to disaster sites tend to converge, seek information, engage in

collective decision making, perform rescues, and help transport injured victims to emer-

gency care providers. It is an under-appreciated fact that the vast majority of live rescues

are carried out by community residents—so-called ‘‘occupant responders’’—rather than by

official response agencies or search and rescue teams (Tierney 2003, p. 36).

Natural disasters tend to be somewhat repetitive and to concentrate in particular areas.

For instance, seismic belts, hurricane prone regions, tornado ‘‘alleys,’’ flood plains, and

areas susceptible to wildfire are well documented (Alexander 2002, p. 5; Putnam 2002,

p. 1). The statistical frequency of many natural hazard events is well established. Similarly,

some types of catastrophic technological failures are also subject to statistical regularity

(Perrow 1984). This general predictability allows for some degree of event-specific

planning and mitigation. For example, development of event-specific mutual assistance

agreements and training exercises, and pre-positioned inventories of equipment and sup-

plies known to be useful in responding to specific types of disasters. However, even when

major disaster events are accurately forecast, they frequently contain elements of surprise

and almost always become occasions for extensive response improvisation (Tierney 2003,

p. 41; Comfort 2002, p. 106; Bruins 2000, p. 70). Disaster response tends to be accom-

plished through ‘‘loosely-coupled collections of individuals, groups, and organizations that

continually change and that have permeable boundaries’’ (Tierney 2003, p. 42). Louise

Comfort has described disaster response networks as ‘‘complex adaptive systems’’

(Comfort 1999, pp. 100–102).

The need for response improvisation is likely to be especially marked in the event of a

terrorist attack. Terrorist attacks, especially those carried out on specific U.S. domestic

resources, do not occur with any type of statistical regularity. This makes them exceedingly

difficult to predict, except possibly in the abstract (Herrick et al. 2006; Alexander 2002;

Comfort 2002). As Richard Little argues, ‘‘we find ourselves in a time where former

contexts of threat, vulnerability, and target have all changed and continue to do so. Threats

Exhibit 2 Criteria utilized by the U.S. EPA to guide restriction or removal of publicly available infor- mation resources

‘‘EPA has developed four criteria for assessing the sensitivity of our information resources: ‘type,’ ‘specificity,’ ‘connectivity,’ and the ‘availability’ of information. Information on a facility’s or a pollutant’s location, chemical identification, volume, acute effects, and plant processes and management falls within the ‘type’ criterion. The ‘specificity’ criterion builds on the type of information and assesses the level of detail available for each type. The ‘connectivity’ criterion looks at the degree to which individual pieces of information can be connected to create realistic scenarios. Finally, the ‘availability’ criterion assesses the level of control that EPA has over releasing the information. This criterion ascertains whether or not EPA is the sole provider of a particular piece of information. If information is widely available through other sources outside of EPA’s control—such as information available from State or local government agencies, public interest groups, in textbooks or from universities—then EPA’s removal may not substantially alter its availability’’ (OMB Watch 2002, p. 1).

Policy Sci (2009) 42:195–210 199

123

are unpredictable and the full range of threats probably unknowable’’ (Little 2004, p. 57).

This virtual inability to predict specific terrorist attacks, coupled with the almost inevitable

reality of response improvisation for all types of disasters, suggests that we will continue to

be faced with situations in which citizens and occupant responders need to react quickly

and intelligently to unfolding disaster events. Thomas Glass and Monica Schoch-Spana

articulate five guidelines for improving disaster response by citizens and the public. One of

the five guidelines focuses specifically on the importance of information provision; and all

five guidelines include a significant public information dissemination component. In the

face of uncertainty, the public needs accurate descriptions of risk and exposure conditions,

instruction in protective measures, and related information. Provision of such information

is valuable not only because it decreases the chances of misinformed response activities

and logistical confusion but also equally important, the flow of information is itself a

source of comfort and reassurance. When disaster strikes there tends to be an urgent need

for reliable information.

Homeland security and environmental and public health-related information: factors that complicate risk characterization

Determination of whether a given information object might or might not be germane to a

terrorist plot is an inherently complicated and speculative matter, depending not only on

the form and content of the information, but also upon intelligence findings concerning

terrorist plans, tendencies, and technical capabilities (Rand 2004). Moreover, even if we

could be reasonably sure that a particular type of information reveals an actionable vul-

nerability, it is not clear whether publicizing vulnerabilities makes them more or less likely

to be exploited. David Pozen suggests that disclosure of security-related information might

actually reduce risk by alerting the public to potential threats. Public attention and vigi-

lance may in turn spur industries and agencies to opt for inherently safer inputs, operations,

and products (Pozen 2005; Echeverria and Kaplan 2002). While lacking an unambiguous

base in empirical documentation, this general line of reasoning has long been the staple of

the U.S. right-to-know advocacy community (Wolf 1996).

It is also logical to presume that suppression of risk in one context or location has the

potential to move it to another. ‘‘Shifting risks may be more dangerous than tolerating

them, both because those who face new risks may be unaccustomed to them and because

those who no longer face old ones may become more vulnerable when conditions change’’

(Douglas and Wildavsky 1983). Removal of information resources such as those outlined

in Exhibit 1 could actually increase public risk by effectively weakening laws designed

specifically to protect human health, environmental quality, and other social values. As an

example, it is well substantiated that public release of data from EPA’s Toxics Release

Inventory has compelled industries to change processes and inputs, and in so doing, reduce

emissions (Cohen 2001; Wolf 1996). Current policies of information restriction could

prove to be examples of the type of risk shifting behavior described by Douglas and

Wildavsky.

Industrial facilities and other components of infrastructure utilize literally thousands of

chemicals and compounds, some of which are synthesized exclusively for specific pro-

cesses at particular plants. Very few specialists, much less the lay public, can be said to

understand the human and environmental risks associated with fugitive or catastrophic

release of these substances, relevant exposure scenarios and pathways, fate and transport

dynamics, and appropriate treatment regimes. Owing to this complexity, the rationale

200 Policy Sci (2009) 42:195–210

123

behind the public’s right-to-know cannot be satisfied by the mere provision of raw data. To

the contrary, meaningful application of the public’s right-to-know requires active facili-

tation on the part of the government. In the U.S., the Environmental Protection Agency and

several state agencies have taken steps to enable and facilitate the public’s awareness and

understanding of environmental and health-related risks, partially through development of

interpretive information products, interactive maps and models, and web-based data

applications. Many of the information resources being withheld or restricted were origi-

nally developed to help citizen and occupant responders make sense of disaster situations,

assess threats, and implement rational and effective response actions (Herrick 2004; Weick

and Sutcliffe 2001).

A common response to concerns about terrorist utilization of publicly-available infor-

mation resources has been to remove the information from the web, but to keep it available

in hardcopy format in public reading rooms (OMB Watch 2002; Costner 2002). This

strategy is problematic because the ready availability of these resources by means of

internet and/or wireless technology is essential to enable its utilization in a context that

actually matters, which is to say, in situ and in real-time by first and/or occupant responders

(Tombs 2005). There are at least four reasons why resources such as these cannot be

replaced by the mere availability of hard-copy documents and records in government

reading rooms: (1) document rooms are not open 24/7/365, and therefore may be

unavailable when disaster strikes; (2) documents and records do not provide interpretive

tools for use by responders in emergency situations; 3

(3) reading rooms are only operated

in limited locations; and (4) the reading room itself may be inaccessible due to the disaster

event. In other words, these information applications were developed to address a prima

facie need-to-know on the part of citizens and the emergency response community.

Removal of these information resources almost certainly reduces emergency response

effectiveness and subjects localities to the real and well-documented threats associated

with acute exposure to hazardous substances.

It is also argued that access to potentially sensitive information resources should be

granted only to authorities such as fire and police departments. Yet as has been docu-

mented in the case of Hurricane Katrina and other disasters, a common aspect of major

disasters is the break-down in authority systems (Wise 2006). For a given disaster situation,

it is frequently difficult to anticipate who or what organization will need to assume authority (Weick and Sutcliffe 2001). While aspects of this deficiency are being addressed

through improved Incident Command Systems (ICS) and expanded training programs, and

there are other aspects of the problem likely to remain the province of self-organizing,

emergent groups (Comfort 1999, pp. 100–102).

While it would be reckless to suggest that all information should be provided to the

public through open access media, it is not clear that domestic vulnerabilities have been

reduced through restriction of the type of infrastructure and health-related information

being considered in this essay. Indeed, it seems plausible to argue that some domestic risk

profiles may have actually increased due to the restriction of environmental and public

health-related data and information. Based on these considerations, the advisability and

3 Many of the information resources removed shortly after 9/11 include simple, interactive maps, models,

and decision aids to enable users to customize information to address local characteristics and other unique circumstances. For example, EPA’s Automated Resource for Chemical Hazard Incident Evaluation (ARCHIE), which enabled users to evaluate how different chemicals behave under alternative release circumstances.

Policy Sci (2009) 42:195–210 201

123

appropriateness of information restriction and/or withholding decisions needs to be judged

in terms of associated trade-offs in social value (Dahl 2004; Gozdor et al. 2003; Pozen

2005; Tombs 2005; Smock 2003). It would thus seem reasonable that federal agencies

formulate, publish, and adopt balancing criteria to guide the management of potentially

sensitive public information. How might such a balancing exercise be approached?

Public access and the need to approach information security by means of a policy decision process

In the western political tradition, the related concepts of public right-to-know and informed

consent are rooted in the moral primacy of the autonomous individual. As Faden et al.

(1986, p. 8) explained, the term autonomy is associated with political values such as

freedom to choose, self-mastery, and responsibility for individual choices. The concept of

autonomy entails both passive and active components, and can be understood in terms of

‘‘external non-constraint and the presence of critical…capacities integral to self-gover- nance’’. Open access to information that can affect an individual’s well-being or life

prospects is therefore critical to the meaningful exercise of individual freedom and choice,

both for political life and personal conduct and development.

But what does it mean to say that an individual has a right to access specific types of information? Rights are powerful assertions of claims that demand respect and socio-

political status (Shue 1980). If an individual appeals to rights, a response is warranted. As

Ronald Dworkin puts it, rights are ‘‘trumps’’ (Dworkin 1977, p. 92). However, it is not the

case that rights are nonderogable. ‘‘The assertion that rights are powerful normative

considerations does not imply that their weight is absolute or that exceptions cannot be

built into their scope’’ (Nickel 1987, p. 17). Rights can conflict with one another, some

rights are more important than others, and there are cases where the exercise of one right

may necessitate the temporary suspension of another right. In other words, there are

occasions in which it is appropriate to recognize and act upon trade-offs among different

rights.

As we have tried to illustrate, the specter of domestic terrorism has spawned a clash of

competing values: on the one hand, there are legitimate reasons to bolster U.S. homeland

security, but on the other hand, citizen deliberation is fundamental to the liberal political

process and the public has a right-to-know about environmental and health-related risks

(Gozdor et al. 2003). The current tendency toward information withholding and restriction

is based on the presupposition that certain types of data, records, and information can be

accessed and used by terrorists to identify and exploit vulnerabilities in U.S. domestic

infrastructure. This is not an unreasonable position; there is evidence that terrorist groups

have and continue to guide their activities through research and utilization of publicly

available information, with perhaps the most notorious example being the commercial

airline timetables utilized by the 9/11 perpetrators. 4

We cannot argue that the right-to-

know clearly trumps security-related concerns, nor can we support a claim that security-

related concerns—by their very recognition—trump the public’s right- and/or need-to-

know about environmental and health-related emissions and exposure risks. This is not an

4 As documented in the 9/11 Commission Report, Khalid Sheikh Mohammed prepared for the hijacking

operation by collecting ‘‘Western aviation magazines; telephone directories for American cities;…brochures for schools; and airline timetables, and he conducted internet searches on U.S. flight schools’’ (National Commission on Terrorist Attacks Upon the United States 2004, p. 157).

202 Policy Sci (2009) 42:195–210

123

issue that can be settled in the abstract or by appeals to political values, but requires

formalized consideration of all substantial interests and values that bear upon outcomes

and careful consideration of alternative policy approaches. The canonical approaches of the

policy sciences have been assembled to address issues such as this.

The basic conceptual framework of the policy sciences was laid out by Harold Lasswell

and later elaborated by Brewer (1974), Brunner (1982), and others. Fundamentally, policy

‘‘is a social process of authoritative decision making by which the members of a com-

munity clarify and secure their common interests’’ (Lasswell and McDougal 1992). The

policy sciences offer methods and intellectual orientations that offer ways to improve the

formulation and implementation of policy. (Clark 1992) According to Tribe, the policy

sciences provide ‘‘devices for organizing thought in rational ways—methods for sorting

out issues and objectively clarifying the empirical relationships among alternative actions

and their likely consequences (1973).

The conceptual framework of the policy sciences involves a series of activities and

considerations that are essential to responsible and effective community-level decision

making. (Lasswell 1971) Although constituting distinct functions, these activities can be

carried out simultaneously, and are often combined. (Clark 1992) As Clark writes, the

decision process is a means of reconciling (or at least managing) conflict through politics in

order to find a working specification of a community’s common interests (Clark 2002). The

outcomes of a decision process are the rules or norms it generates. The basic activities of

the classical policy decision process are outlined below:

Intelligence: Decision making depends upon the collection, compilation, and analysis of data and information on past trends, current status, relevant processes, and alternative

future projections for factors and conditions that bear upon the issue under consider-

ation. Depending on the issue, this process may need to be specialized, intensive, and

protracted. Decisions based upon inadequate characterization of a problematical

situation have potential to result in irrelevant, inefficient, or counterproductive

outcomes.

Promotion: Promotion involves recommendation and the marshalling of support for policy alternatives. Ideally, the promotion process should entail active debate about

different courses of action. The promotion process helps citizens and political

stakeholders to refine and focus their values and expectations, while dismissing

alternative courses of action.

Prescription: Prescription is the process through which a community formulates and enacts policies and guidelines for action. The prescription process implies that specific

community members posses recognized and legitimate authority to execute and enforce

rules. To be politically appropriate and effective, a prescription should include three

elements: (1) topically appropriate and technically adequate content, (2) a designation

and justification of authority, and (3) a clear stipulation of what will be controlled by the

policy and how that control will be exercised.

Invocation: Invocation involves actions taken and resources utilized to invoke a rule or prescription. As Clark writes, ‘‘it is the initial or provisional characterization of the kinds

of behavior that will be permitted in terms of a prescription’’ (Clark 2002). In many

cases, governmental bodies employ individuals specialized in this function, such as law

enforcement officials, who typically exercise provisional judgments about compliance

with prescriptions. It is essential that invocation activities be clear, transparent, and

widely communicated; and that individuals acting in an official capacity are

appropriately credentialed, skilled, and trained. Lacking these conditions, invocation

Policy Sci (2009) 42:195–210 203

123

of a prescription runs the risk of appearing arbitrary in its application of governmental

authority.

Application: Application is the final, authoritative demonstration and characterization that specific behaviors or actions either do or do not comply with a prescription. Such

determinations may be founded upon legal, administrative, or technical factors, and are

thus likely to involve specially designated authorities or officers. The application

function often involves formalized interpretation of rules with respect to particular cases

or instances. While application and invocation clearly overlap, application pertains to

the later stages of an event lifecycle.

Appraisal: Appraisal involves the critical evaluation of a decision process and its particular outcomes. Through a formal and systematic process of appraisal, a community

can characterize and assess the degree to which its policy goals have been achieved.

Data and criteria relevant to the appraisal of a policy should be consistent with the

results of assessments conducted under the intelligence element of the decision process.

Termination: Termination involves the cessation, repeal, or fundamental alteration of a prescription. Often overlooked during policy formulation, termination is an action

invoked by determination or recognition that a policy has accomplished its objective(s)

and is no longer necessary, or alternatively, has failed to accomplish its goal and must be

substantially changed or abandoned in order to attempt a different solution.

As outlined in this essay, the U.S. information restriction regime is comprised of a

mixed bag of loosely coordinated statutes, executive memoranda, and legal interpretations.

It includes a variety of departmental or agency-level actions, of differing degrees of

formality, authorized at various levels of authority. As illustrated in Table 1, there appear

to be many ways in which decisions to remove or restrict environmental and public health-

related information fail to address basic aspects of the policy sciences decision process.

Most importantly, the decision process appears to have been one-sided in that more

consideration was given to security risks associated with contemplated terrorist actions

than was focused on environmental and public health-related risks due to pollutant and

toxic exposures.

Faithful and systematic application of the Lasswellian decision process does not

guarantee perfect policy. However, formal consideration and application of explicit criteria

does help to expose ‘‘mistakes, inconsistencies, contradictions, and deceptions in the

decision process.’’ It can also make clear how [a] decision factor might be improved,

according to specific standards, in practice’’ (Clark 2002).

A mechanism for balancing information access with information security

The general question of how to balance information release with national security is by no

means new. Since the end of WWII, policy makers, policy practitioners, and security

experts have engaged in a number of efforts to critique and improve the management of

sensitive information. In 1997, a commission comprised of Daniel Patrick Moynihan,

Samuel Huntington, John Deutch and others looked for ways to improve U.S. govern-

mental processes for declassification of sensitive information. According to the

Commission, persistent problems associated with sensitive information management

include huge volumes of material subject to review, limited resources within agencies, staff

resistance to mandated declassification efforts, a legacy of poor records management, and a

lack of leadership insistence that recordkeeping and declassification should be priority

204 Policy Sci (2009) 42:195–210

123

agency missions (RCPRGS 1997). Despite the institutional difficulties that plague

declassification efforts, the basic process rests upon the simple and non-controversial

rudiments of risk management. As noted by the Commission, declassification involves

identification and use of ‘‘sensible, cost-effective, and routine ways to separate the

Table 1 Narrative summary of decision process steps as they relate to removals and restrictions of envi- ronmental and public health-related information (Costner 2002; OMB Watch 2002; Dahl 2004; Gozdor et al. 2003; Pincus 2006)

Decision process element

Summary

Intelligence Removals appear to have been guided by simple templates or ‘‘rubrics’’ to help identify data or information that might tend to be sensitive if available for use by individuals or groups contemplating malevolent action. The degree to which alternatives to removal were considered is unclear. Nor is it clear whether officials attempted to balance risks associated with removal against those associated with open access and potential utilization by terrorists. Finally, there is no evidence that the process was conducted in response to specific, ‘‘actionable’’ threats to U.S. infrastructure.

Promotion The removal of environmental and public health-related information from federal agency web sites has been reviewed ex post in the context of Congressional hearings, GAO audits, and commentary on the part of the media and non- governmental watch-dog organizations. However, public commentary was not sought before-the-fact, and stakeholders were caught off-guard by removal of valued data and information applications formerly available through public access web sites.

Prescription Actions to redact or remove information from public circulation have been authorized in terms of legal precedents and statutory provisions, and initiated under the signatory authority of senior political and career officials. However, stakeholders express confusion regarding the substantive rationale behind removal decisions, and do not understand why some information items have been removed, while others remain available. In others words, the prescription function seems inconsistent and lacking in logical basis.

Invocation It is not clear that all individuals charged with review and removal of sensitive materials have been trained in security assessment. Information assessment activities and subsequent restrictions were not conducted in the public venue; nor did agencies publish their intention to conduct security screening exercises and remove information from public venues. In many ways, this series of agency actions lacked formal, meaningful invocation.

Application The removal of environmental and public health-related information seems to lack precedent, transparency, and the stipulation of reasonably foreseeable outcomes. It is not clear whether and how specific removals are subject to appeal or other forms of adjudication.

Appraisal The implications of information removal and access restrictions have not been evaluated by the agencies mentioned in this essay, at least not in a manner that has been subject to public review and critique. Nor have they provided broadly available mechanisms through which to elicit public and stakeholder feedback on the impacts associated with removal of specific information items. Further, it is not clear whether agencies have taken steps to obtain data and assess whether risks have migrated due to information restrictions, whether removals have resulted in unintended consequences, or, most fundamentally, whether and how removals have impeded non-security-related agency mission objectives.

Termination The information removal and restriction activities outlined in this essay (see Exhibit 1) have no clear sunset or termination date. Some materials removed shortly after the 9/11 attacks have been replaced, others made available through restricted conditions, and others remain unavailable.

Policy Sci (2009) 42:195–210 205

123

categories of materials no longer warranting protection from those needing to stay secret

(RCPRGS 1997; Quist 1993). Such efforts attempt to balance plausible harms associated

with information release against benefits expected to ensue from enhanced access, in

essence asking the same questions as would be asked in the context of decisions regarding

environmental and health-related information withholding or disclosure.

How should the government undertake the challenge of balancing information access

against the legitimate need for information restriction? The U.S. government’s policy for

management of federal information systems and resources is framed by a relatively

obscure document published by the Office of Management and Budget (OMB), known as

OMB Circular A-130. Using prose that sound almost Jeffersonian, Circular A-130

acknowledges that ‘‘[g]overnment information is a valuable national resource…’’ and that ‘‘[t]he free flow of information between the government and the public is essential to a

democratic society.’’ The document further states that ‘‘management of federal information

resources should protect the public’s access to government information’’ (US OMB 2000,

p. 4).

Information provision is clearly part of the organic mission of U.S. government agen-

cies. Moreover, as already noted, many U.S. laws and statutes require the collection and

public dissemination of specific types of data and information. Indeed, in some cases,

Congress has required that the government take steps to assure that stakeholders and

citizens are able to utilize disseminated data and information in an appropriate and

effective manner. 5

This means that federal agency information management activities are

subject to the Administrative Procedures Act (APA). The APA is intended to make agency

activity as open as possible to participation from outside the agency and to reduce the

scope for ‘‘arbitrary and capricious’’ decisions by requiring that agency actions be justified

and susceptible to administrative and/or court challenge. As Steven Kelman writes, even

[t]he simplest, most bare-boned procedures require that agencies publish all proposed

actions in the Federal Register and allow at least 30 days after publication for submission of written comments’’ (Kelman 1987, p. 93). Under Section 552, agencies are required to

make public ‘‘statements of the general course and method by which [their] functions are

channeled and determined, including the nature and requirements of all formal and

informal procedures.’’

Based upon the stipulations of A-130 and widely-established standards of APA practice,

it would seem reasonable to expect that U.S. federal agencies should (a) publish notifi-

cation of the planned or anticipated removal of publicly-available data and information;

and (b) articulate a decision process and associated assessment criteria through which to

determine whether specified records or information resources ought to be restricted or

removed from public access.

Congress and the Executive Branch commonly use devices such as guidelines, technical

manuals, bulletins, circulars, and interpretive memoranda to stipulate or refine policy

(Kerwin 1994, p. 258). Consistent with its policy coordination and oversight role, the

Office of Management and Budget could devise and provide guidance for departments and

executive agencies to utilize in formulating approaches to weigh competing values with

respect to decisions regarding the disposition and management of sensitive information. In

recent years, OMB guidance has been issued and used to address a wide range of sub-

stantive issues, including risk assessment protocols, state and local grants management,

5 For example, CERCLA includes provisions that not only require information sharing with the public, but

also require citizen participation and information exchange with technical experts (Gozdor et al. 2003, p. 10980).

206 Policy Sci (2009) 42:195–210

123

privacy, information technology procurement, data and information quality, economic cost

and benefit characterization methodologies, and statistical definitions.

Based on OMB guidance, agencies would develop and publish assessment criteria and a

weighing approach to formalize and guide decision making regarding whether to dis-

seminate and how best to manage potentially sensitive environmental and public health-

related information. Based on the fundamental values at play in this debate, it would seem

reasonable that OMB and agency criteria would need to recognize and address at least five

factors (Rand 2004; Herrick et al. 2008):

1. Will the document or information directly reveal a potential vulnerability or weakness in security? Agencies should consider whether the information would be useful for

selecting one or more specific potential targets, or executing an attack on a potential

target.

2. If the data or information requested were combined with other information, could it

reveal a vulnerability or weakness in security? If so, is the other information readily

available in the public domain?

3. To what degree the public or particular stakeholders benefit through the ready availability and/or release of environmental or health-related data and information?

The ability to articulate a need for the requested information should weigh heavily

toward a decision to maintain and/or enhance public access.

4. If the Agency does not release the information, will it forego a defined benefit? Agencies operate legislated programs and execute policies in order to achieve specific

mission objectives. If information redaction or restriction impedes or retards mission

functions, it should be avoided or otherwise mitigated. The ability to articulate a

benefit through information sharing should weigh heavily toward appropriately

managed dissemination and/or disclosure.

5. Is the document or information already widely available or easily retrievable to the public? Agency records and documents may have been released already and

disseminated to external organizations or be available through web sites or other

media within open sources or the public domain. If the requested information is

already readily available, withholding may not be a viable option.

Consistent with the intent and authority of the APA, a balancing framework should also

define and include administrative mechanisms to allow affected parties to seek appeal or

functional redress for the denial of information resources. It should also require a reporting

function under which agencies communicate annually to OMB, the President, and Con-

gress regarding types and instances of information withheld or restricted from public

access, information previously withheld but restored to public access, and a detailed

narrative rationale behind any such decisions. Such a framework, combined with the

appeal and reporting process, provides an appropriately thorough and transparent mech-

anism through which to consider the trade-offs entailed in agency decisions to withhold or

release certain types of data and information. Importantly, the reporting stipulation would

provide a source of data to support program evaluation and future empirical analyses of the

efficacy of alternative withholding and release policies.

Conclusion

There is little argument that the U.S. government has expanded, intensified, and further

institutionalized its use of force and coercion in the aftermath of 9/11. However, the

Policy Sci (2009) 42:195–210 207

123

constitutional legitimacy and cultural appropriateness of these security-related enhance-

ments remains open to question. As we have discussed, this tension is especially acute in

the related arenas of informed consent and public right-to-know. While the public’s right-

to-know is not absolute, it is instrumental to the exercise of fundamental rights and tra-

ditional political values. Focusing on the environment and public health, this essay has

explored sociological and rights-based reflections that point toward the need for a policy

decision process addressing dissemination or restriction of potentially sensitive

information.

In the early 1940s, Harold Lasswell introduced the construct of the garrison state

(Lasswell 1941). Animated by totalitarian claims to world order, the garrison hypothesis

asserts that civil freedoms will diminish as preparation for war and maintenance of security

become political society’s dominant thrust. The ‘‘main components of the garrison state

include compulsion, full labour, abolition of political parties, abolition of legislatures, and

abolition of elections’’ (Morgan 2004). While these extreme outcomes did not come to

pass, other aspects of the garrison concept may have begun to materialize. Commentators

such as Goldsmith and Mayer argue that factors such as warrantless wiretapping, indefinite

detention without due process, extraordinary renditions, coercive interrogation, and a

pervasive disregard of Congressional prerogative constitute a material erosion of the

American constitutional system (Goldsmith 2007; Mayer 2008) Indeed, there is a striking

similarity between Lasswell’s basic fear and the practical problems associated with

maintenance of civil liberties while simultaneously addressing a perpetual threat from

terrorists.

Acknowledgments Some of the research underlying this essay was sponsored by the American Water Works Association Research Foundation. The project was managed by Mr. Frank Blaha, who also provided helpful comments on an early draft of this paper.

References

Alexander, D. (2002). Principles of emergency planning and management. Oxford: Oxford University Press. Brewer, G. (1974). The policy sciences emerge: To nurture and structure a discipline. Policy Sciences, 5,

239–244. doi:10.1007/BF00144283. Bruins, H. (2000). Proactive contingency planning vis-à-vis declining water security in the 21st century.

Journal of Contingencies and Crisis Management, 8(2), 63–72. doi:10.1111/1468-5973.00125. Brunner, R. (1982). The policy sciences as science. Policy Sciences, 15, 115–135. doi:10.1007/BF00143074. Clark, T. (1992). Practicing natural resource management with a policy orientation. Environmental Man-

agement, 16(4), 423–433. doi:10.1007/BF02394119. Clark, T. (2002). The policy process: A practical guide for natural resource managers. New Haven: Yale

University Press. Clarke, L. (2003). Conceptualizing responses to extreme events: The problem of panic and failing grace-

fully. In L. Clarke (Ed.), Terrorism and disaster: New threats, new ideas. Oxford: Elsevier. Cohen, M. (2001). Information as a policy instrument in protecting the environment: What have we learned?

Environmental Law Reporter, 31, 10425–10431. Comfort, L. (1999). Shared risk: Complex systems in seismic response. Oxford: Pergamon, Elsevier Science

Ltd. Comfort, L. (2002). Rethinking security: Organizational fragility in extreme events. Public Administration

Review, 62(Special Issue), 98–107. doi:10.1111/1540-6210.62.s1.18. Costner, B. (2002). Access denied. The Bulletin of the Atomic Scientists, 58(2), 58–62. http://www.

thebulletin.org/print.php?art_ofn=ma02costner. Dahl, R. (2004). Does secrecy equal security? Limiting access to environmental information. Environmental

Health Perspectives, 112(2), 104–107. Douglas, M., & Wildavsky, A. (1983). Risk and culture. Berkeley: University of California Press. Dworkin, R. (1977). Taking rights seriously. Cambridge: Harvard University Press.

208 Policy Sci (2009) 42:195–210

123

Echeverria, J., & Kaplan, J. (2002). Poisonous procedural ‘‘Reform’’: In defense of environmental right to know. Washington, D.C.: Georgetown Environmental Law and Policy Institute, Georgetown University.

Faden, R., Beauchamp, T., & King, N. (1986). A history and theory of informed consent. Oxford: Oxford University Press.

Fischer, H. (2006). Disaster myths and their implications for disaster planning and response. Natural Hazards Observer, 31(1), 6–7.

Glass, T., & Schoch-Spana, M. (2002). Bioterrorism and the people: How to vaccinate a city against panic. Clinical Infectious Diseases, 34(15), 217–223. doi:10.1086/338711.

Goldsmith, J. (2007). The terror presidency: Law and judgment inside the Bush administration. New York: W.W. Norton.

Gozdor, C., Campbell Jones, S., Klick, K., & Steinhilber, M. (2003). Where the streets have no name: The collision of environmental law and information policy in the age of terrorism. Environmental Law Reporter, 33(12), 10979–10995.

Herrick, C. (2004). Objectivity versus narrative coherence: Science, environmental policy and the U.S. data quality act. Science & Public Policy, 7, 419–433.

Herrick, C., Pratt, J., Raucher, R., Kalas-Adams, N., Cotruvo, J., Darr-Bornstein, K., et al. (2006). Emer- gency response and recovery planning for water systems: A kit of tools. Denver: American Water Works Association Research Foundation.

Herrick, C., Scherer, E., & Walter, G. (2008). Critical information policies for water utilities. Denver: American Water Works Association Research Foundation.

Kelman, S. (1987). Making public policy. New York: Basic Books. Kerwin, C. (1994). Rulemaking: How government agencies write law and make policy. Washington, DC:

Congressional Quarterly Press. Lasswell, H. (1941). The garrison state. American Journal of Sociology, 46, 455–468. doi:10.1086/218693. Lasswell, H. (1971). A pre-view of policy sciences. New York: American Elsevier. Lasswell, H., & McDougal, M. (1992). Jurisprudence for a free society: Studies in law, science, and policy

(Vol. 2). New Haven: New Haven Press. Little, R. (2004). Holistic strategy for urban security. Journal of Infrastructure Systems (June), 52–59. doi:

10.1061/(ASCE)1076-0342(2004)10:2(52). Mayer, J. (2008). The dark side: The inside story of how the war on terror turned into a war on American

ideals. New York: Doubleday Books. Morgan, M. (2004). The garrison state revisited: Civil-military implications of terrorism and security.

Contemporary Politics, 10(1), 5–19. doi:10.1080/13569770410001701224. National Commission on Terrorist Attacks Upon the United States. (2004). The 9/11 Commission Report.

New York: W.W. Norton and Company. National Journal Group. (2001). Threat assessment: Chemical plants could be targets. Global Security

Newswire. http://www.nti.org/d_newswire/issues/2001/12/17/11s.html. Nickel, J. (1987). Making sense of human rights. Berkeley: University of California Press. OMB Watch. (2002). EPA turns over documents on information removal, yet questions remain. OMB

Watch. http://www.ombwatch.org/article/articleprint/738/-1/39. Padover, S. (1953). The complete Madison. New York: Harper and Brothers. Perrow, C. (1984). Normal accidents: Living with high-risk technologies. New York: Basic Books. Pincus, W. (2006). GAO finds Pentagon erratic in wielding secrecy stamp. The Washington Post. July, 14,

A19. Pozen, D. (2005). The mosaic theory, national security, and the freedom of information act. The Yale Law

Journal, 115, 628–679. Putnam, L. (2002). By choice or by chance: How the internet is used to prepare for, manage, and share

information about emergencies. First Monday, 7(11), 1–11. Quist, A. (1993). Security classification on information. Oak Ridge, TN: Oak Ridge National Laboratory. Rand. (2004). Mapping the risks: Assessing the homeland security implications of publicly available geo-

spatial information. Santa Monica: Rand Corporation. RCPRGS. (1997). Report of the commission on protecting and reducing government secrecy. (Senate

Document 105–2) Washington, DC: United States Government Printing Office. Shue, H. (1980). Basic rights. Princeton, NJ: Princeton University Press. Sidel, M. (2004). More secure less free?. Ann Arbor: University of Michigan Press. Smock, R. (2003). Prologue to a farce or a tragedy? Accessing government records in the wake of 9/11/

2001. The Public Historian, 25(2), 123–127. doi:10.1525/tph.2003.25.2.123. Tierney, K. (2003). Disaster beliefs and institutional interests: Recycling disaster myths in the aftermath of

9–11. In L. Clarke (Ed.), Terrorism and disaster: New threats, new ideas. Oxford: Elsevier.

Policy Sci (2009) 42:195–210 209

123

Tombs, R. B. (2005). Policy review: Blocking public geospatial data access is not only a homeland security risk. URISA Journal, 16(2), 49–51.

Tribe, L. (1973). Policy sciences: Analysis or ideology? Philosophy & Public Affairs, 2, 66–110. U.S. Congress. (2002). Public health security and bioterrorism preparedness act of 2002. Public Law, 107–188. U.S. Department of Justice. (2001). FOIA Post. Washington, DC: United States Department of Justice Office

of Information and Privacy. http://www.usdoj.gov/oip/foiapost/2001foiapost19.htm. U.S. Environmental Protection Agency. (2006). Terminology reference system. http://www.epa.gov/trs/. U.S. Office of Management and Budget. (2000). Circular A-130: Management of federal information

resources. Washington, DC: United States Office of Management and Budget. Weick, K., & Sutcliffe, K. (2001). Managing the unexpected. San Francisco: Jossey-Bass. Wise, C. (2006). Organizing for Homeland security after Katrina: Is adaptive management what’s missing?

Public Administration Review (May/June), 302–318. doi:10.1111/j.1540-6210.2006.00587.x. Wolf, S. (1996). Fear and loathing about the public right to know: The surprising success of the emergency

planning and community right-to-know act. Journal of Land Use and Environmental Law, 11(2), 217–320.

210 Policy Sci (2009) 42:195–210

123