Of Risk and Pork: Urban Security and the Politics of Objectivity

Of risk and pork: urban security and the politicsof objectivity

Andrew Lakoff & Eric Klinenberg

# Springer Science+Business Media B.V. 2010

Abstract This article focuses on the Urban Area Security Initiative (UASI)controversy as a case study in the politics of risk assessment. It examines strugglesamong diverse actors–think tank experts, journalists, politicians, and governmentofficials–engaged in the contentious process of establishing a legitimate definition ofrisk. In the field of homeland security, the means of conducting rational riskassessment have not yet been settled, and entrepreneurial officials from urban andregional governments use different techniques to identify local risks and vulner-abilities. In this contentious process, federal bureaucrats are responsible fordetermining how to allocate resources fairly and rationally to different cities andmetropolitan regions, given that local officials have clear incentives to request fundsand little cause to refrain. Although “rationality” is supposed to replace “politics” inmaking bureaucratic decisions over the allocation of resources, what we find insteadis a political struggle over how to define, measure, and manage risk. For politicalactors, victory in debates over urban security comes from codifying one’s interestswithin the technical practice of risk assessment.

Keywords Homeland security . Objectivity . Rationality . Risk assessment . Risk

“Pork 1, Antiterrorism 0,” declared the New York Times in its editorial of June 2,2006 (New York Times 2006a). With the Department of Homeland Security’s (DHS)announcement of its Urban Area Security Initiative (UASI) grants, a long-simmeringdispute over how to apportion funds for urban security had exploded into publiccontroversy. New York City officials were up in arms: the city’s share of UASI fundshad been cut from $206 million the previous year to $124 million, while smaller cities

Theor SocDOI 10.1007/s11186-010-9123-3

A. Lakoff (*)Department of Sociology, University of Southern California,Los Angeles, CA, USAe-mail: [email protected]

E. KlinenbergDepartment of Sociology, New York University, New York, NY, USA

like Atlanta and Louisville received an increase in funding for security measures. TheTimes’ headline suggested that DHS’s decision was a result of “pork-barrel politics”:the implication was that Washington politicians were using the grants program to sendmoney to loyal constituencies without regard to the nation’s actual security needs.New York City Mayor Michael Bloomberg plainly stated this view: “I think the factsare clear. What they’ve really done is taken what was supposed to be threat-based andjust started to distribute it as normal pork” (Esposito 2006). Peter King, aCongressional Representative from New York, was harsher in his judgment, declaringto the Washington Post: “As far as I’m concerned, DHS and the Administration havedeclared war on New York City” (Eggen and Sheridan 2006).

What commentators found both galling and surprising was that the allocationdecisions, which in previous years had been criticized by officials and securityexperts as politically motivated, had ostensibly been based on a new, rational processof risk-assessment. As a DHS document describing the UASI program explained:“The FY 2006 DHS risk methodology represents a major step forward in theanalysis of the risk of terrorism faced by our Nation’s communities. Tremendousgains have been made in both the quality and specificity of information and analysisincorporated within the model, yielding the most accurate estimation possible of therelative risk of prospective grant candidates” (Department of Homeland Security2006b). To support this claim, the document offered a detailed illustration of theempirical foundation and sophisticated mathematical techniques used to determinefunding allocations:

The levels of complexity of the formula and data calculations have likewiseincreased markedly from FY 2003 to FY 2006. For example, in FY 2003, threeprimary equations were used in the risk analysis; in FY 2006, over 4,100equations were used. For FY 2003, approximately 1,500 calculations weremade, in contrast to more than 3 billion calculations in FY 2006. FY 2005UASI formulations were represented within a spreadsheet of about 72,000cells; if the FY 2006 UASI calculations could be included in a spreadsheet, itwould contain more than 20 million cells (p. 3).

This display of numbers was meant to indicate the rationality and objectivity ofDHS’s budget calculations. How, critics then wondered, could the agency have soradically cut funding to the cities considered to be most under threat? In the weeksfollowing the announcement of its 2006 UASI grants, DHS provided a number ofjustifications for its decision: there had been budget cuts to the UASI program; NewYork City’s application had been poorly prepared; the grants were intended forcapacity-building rather than personnel; panels of respected homeland securityexperts had rated applications, and New York City’s had fallen short. Thesearguments only fueled the political firestorm. In the end, the official in charge of thegrant-making process resigned, and DHS sought to appease its critics by introducinga new set of evaluative criteria for the next funding cycle.

There were three key issues at stake in the public controversy over how toallocate federal resources for urban security. The first is a familiar problem infederalist politics in the United States: how to distribute resources fairly andrationally to different cities and metropolitan regions, given that officials represent-ing all parts of the nation haves clear incentives to request funds, and little cause to

Theor Soc

refrain (Kettl 1987; Wildavsky 1998; Roberts 2005). The second concernsbureaucratic rationality, in particular the question of how a fledgling governmentalagency can construct methods of evaluation and calculation that a diverse set ofpowerful political actors will accept as fair and legitimate (Porter 1995; Strathern2000). The third is the emerging political question of how a threat that defiesprobabilistic calculation is brought into the realm of governmental rationality (Beck1999; Giddens 1990).

This article explores these three themes through a close analysis of the UASIcontroversy, in which political officials representing large and powerful cities andprofessional security experts publicly challenged the allocation of federal resourcesin a program designed to supplement municipal security budgets. Specifically, wefocus on the question of how experts and political officials worked to transform theadministrative definition of risk used to allocate security resources.1

Critics of the Department of Homeland Security’s allocation decisions cast thecontroversy as a struggle between “risk” and “pork.” They accused officials ofdistributing federal resources to localities as a means of consolidating political powerrather than making decisions through objective and rational calculation. Thisaccusation assumed that there existed a stable, agreed-upon means for calculatinghomeland security risks. As we will see, however, such a calculative mechanismdid not yet exist. In fact, the key struggle at the heart of the controversy was overhow to define a method of risk calculation in a new and unsettled area ofbureaucratic action. While the debate was framed as a contest between rationalityand politics, at stake was not a choice between the two, but rather a contest overwhat bureaucratic practices constitute legitimate rationality, and a struggle to havethe interests of particular cities and regions codified into the formal mechanism ofrisk assessment.

Objectivity and bureaucratic authority

Let us begin by asking why DHS initially sought to justify its UASI allocationdecisions through a public demonstration of the complex calculations of its “riskmethodology.” The display of numbers was meant to indicate that the agency’sallocation decisions had not been shaped by the influence of illegitimate interests. AsMax Weber argued, the authority of a bureaucratic agency is grounded in itsperceived objectivity—its adherence to impersonal standards of judgment: “Bureau-cracy develops the more perfectly, the more it is ‘dehumanized,’ the more

1 This effort to reshape the practice of risk assessment took place through public discussion and debate.Therefore we have scrutinized the public materials produced over the course of the debate, tracking howsignificant actors put in question the technical methods used by DHS to calculate allocations in order toincrease their share of resources. We examined prepared statements as well as discussion transcripts fromhearings held by the Senate Committee on Homeland Security and Governmental Affairs, the HouseCommittee on Homeland Security, and the House Committee on Government Reform. However, much ofthe debate over UASI funding took place outside of Congressional hearing rooms. From the outset, electedpolitical representatives and officials from the White House used the major media to make and defendtheir positions. For this reason, press releases, interviews, editorials news articles, and executivesummaries of policy reports were also important empirical resources for following debates over how todefine risk, vulnerability, and need, as well as how to allocate scare security resources.

Theor Soc

completely it succeeds in eliminating from official business love, hatred, and allpurely personal, irrational, and emotional elements which escape calculation”(Weber 2003). To be objective and rational in public administration means tooperate “according to calculable rules and ‘without regard for persons’” (ibid). In asimilar vein, Ted Porter notes that in US political culture, bureaucratic agencies areexpected to display “mechanical objectivity”: to avoid being seen as biased,“expertise should be mechanized and objectified. It should be grounded in specifictechniques sanctioned by a body of specialists. Then mere judgment, with all its gapsand idiosyncrasies, seems almost to disappear” (Porter 1995).

As Porter emphasizes, such rule-based calculation as the basis for bureaucraticdecision is an ideal rather than a description of typical practice. Only under certaincircumstances are bureaucratic agencies actually required to publicly demonstratethat they are following calculable rules. It is especially weak agencies incontroversial situations that must demonstrate the objectivity of their decisioncriteria: “objectivity lends authority to officials who have very little of their own”(Porter 1995). As Nikolas Rose puts it, “where decisions are contested anddiscretion is criticized, the allure of numbers increases” (1999). Quantitativedecision mechanisms thrive in settings characterized by political controversy andpublic scrutiny. Indeed, it is typically the accusation of bias that forcesadministrators to demonstrate their objectivity publicly–as, for example, whenDHS was accused by its critics of being engaged in “politics” at the expense ofactual security.

This pressure to defend against external critics through the display of mechanicalobjectivity helps explain the impressive number of calculations that were ostensiblyinvolved in DHS’s “risk methodology.” As a new and controversial governmentagency, DHS was highly susceptible to outside criticism. Founded in the wake ofthe attacks of 9/11, the department was a complex combination of twenty-twoalready-existing agencies with diverse organizational cultures and administrativenorms. Moreover, its mandate–to provide “homeland security”–was a newlyinvented sphere of bureaucratic action. On the one hand, the field of homelandsecurity was characterized by a sense of urgency and a demand for new initiatives;on the other hand, homeland security was an elastic and ill-defined concept (Lakoff2007). In this context, an emerging community of homeland security expertsprovided a potential instrument for achieving mechanical objectivity–risk assessment–one that was readily adopted by DHS administrators. But the process of quantifyingrisks and vulnerabilities so that they could be subjected to formal risk analysis wasfraught with problems. DHS’s attempt to demonstrate impartiality publicly was furtherhampered by the requirement that key data remain classified on the grounds ofsecurity concerns.

In this context, as we will see, an explosive debate around the politics of expertisearose. And while it is tempting to see the UASI controversy as simply the story of aninept or corrupt bureaucracy caught by scrupulous whistleblowers, what actuallyunfolded was more interesting: the specific details of technocratic calculationbecame the object of political dispute. In other words, the UASI controversy wasbased not on a conflict between “rationality” and “politics”–as many actors sought toportray it–but rather serves to demonstrate the operations of a contemporary politicsof rational calculation.

Theor Soc

The political administration of risk

Scholars of political rationality have become increasingly attuned to the centrality offormal decision tools, such as risk assessment, in contemporary administrativesettings. Through their apparent scientific rigor and universal applicability, thesedecision tools potentially provide a form of objectivity that will be resilient againstexternal critique. Prominent examples include the use of cost-benefit analysis in landuse policy (Porter 1995), evidence-based treatment protocols in health servicesdelivery (Timmermans and Berg 2003), and risk assessment in environmentalregulation (Jasanoff 1986). These decision instruments share the characteristic ofquantifying diverse entities in order to make them comparable and thus to makedecisions “calculable” (Espeland and Stevens 1998). The premise of such instru-ments is that they eliminate arbitrary, biased, or mistaken judgment from the policyprocess.

These formal tools have been criticized from some quarters for shielding highlypolitical judgments behind an image of objective technical decision. For example, inher analysis of “rational choice liberalism,” S.M. Amadae argues that the use ofquantitative systems analysis in urban planning is anti-democratic insofar as itrestricts input into policy-making to a technocratic elite. She traces the rise ofsystems analysis in the Cold War US to an alliance of private philanthropists, thebusiness community, and scientific policy analysts who sought control overgovernment budgetary decisions. Amadae writes that the agenda of “eliminatingpolitics from decision making,” as one advocate of the systems approach put it(Amadae 2003), was precisely aimed to narrow the scope of decision to policy elites.For such critics, formal decision tools displace both democratic processes and thereasoned judgment of experienced practitioners.

Other analysts, however, have seen the adoption of such instruments in publicadministration as a more dynamic and contingent process, one whose politicalimplications are not predetermined. This work has sought to characterize the specificcontexts in which formal decision tools come to function as universal standards ofrationality. Thus, according to Porter, the widespread adoption of cost-benefitanalysis in the US government over the course of the twentieth century is bestunderstood as the product of “bureaucratic conflict in a context of overwhelmingpublic distrust” (Porter 1995). Through the example of the U.S. Army Corps ofEngineers’ justification of its massive water projects, he shows that when powerfulconstituencies challenged such projects, the Corps was forced to develop highlytechnical methods of calculation. Given a widespread “trust in numbers” rather thanin experienced judgment within American political culture, quantification provided adefense for costly projects that might otherwise have been scuttled.

Similarly, the use of a formal decision tool may be a political strategy within afragmented organization, as Wendy Espeland has shown in her study of conflictswithin the Bureau of Reclamation over the Orme Dam Project in Arizona. In thiscase, cost-benefit analysis was initially adopted to restore the public credibility of theBureau, but then became critical to the surprising eventual rejection of the DamProject. Such cases indicate that the political valence of a given decision tool isunderdetermined. As Espeland writes, “rationality is a highly contingent accom-plishment, a potential that is enacted and propelled by people in contexts with

Theor Soc

disparate motives, constraints, and levels of self-consciousness” (Espeland 1998). Inthe case of UASI, we might then ask: how did “risk assessment” become theaccepted form in which rational and objective decisions about homeland securityresources would be undertaken? And how was the actual decision mechanismdesigned?

Of risk and pork

DHS faced not only the problem of calculation, but also that of organization. Theagency had been formed rapidly in 2002, in the wake of the attacks of 9/11, underthe pressure of an agreement that it was necessary to prevent and prepare for furtherattacks. Its establishment was the largest reorganization of the federal governmentsince the 1947 National Security Act. Twenty-two agencies, comprising 184,000employees and with an annual budget of over $40 billion per year, were broughttogether in an organization whose goals–and ways of achieving these goals–were notentirely clear (Falkenrath 2001; Klinenberg and Frank 2005; Perrow 2007). DHSwas internally divided from its inception, with multiple visions of its mission,including border control, law enforcement, intelligence gathering, surveillance, andemergency preparedness. Five years later the goals of the agency remained subject tointense political contestation: What exactly did “homeland security” refer to? Bywhat means was it to be provided? And what agencies–at the local and federal level–were responsible for its provision? All of these issues–as well as more mundanequestions around the control of resources–were at stake in the UASI controversy.

The backdrop to the controversy was an ongoing dispute around the allocation ofhomeland security funds in a federal system (Collier and Lakoff 2007). Followingthe establishment of DHS, officials from big cities–especially New York andWashington, DC–complained that a disproportionate amount of the federal budgetthat had been earmarked for homeland security was going to small states. Indeed, theten states with the highest per capita homeland security grant funding for fiscal year2003 were (in order) Wyoming (at $61 per person), Vermont, North Dakota, Alaska,South Dakota, Delaware, Montana, Rhode Island, Hawaii, and New Hampshire,while the ten states with the lowest per capita allocations were (beginning withthe bottom): California (at $14 per person), Texas, New York, Florida, Illinois,Pennsylvania, Ohio, Michigan, Georgia, and New Jersey (Roberts 2005; Perrow2007). Given the apparently greater threat of terrorism to the nation’s financial andgovernmental centers, these allocation levels were a source of political challenge.Christopher Cox, the former Republican congressman who later became Chairmanof the Securities and Exchange Commission, complained that DHS was allocatingfunds to states and cities “without the prerequisite analysis of risk. Theseauthorities, then, occasionally find themselves looking for ways to spend the money”(Chertoff 2005).

Stories about wasteful spending of DHS funds in small states made the agencyvulnerable to attacks from big city politicians. In Iowa, it was reported, localauthorities had used DHS funds to purchase traffic cones and wall clocks with built-in hidden cameras. And in Tiptonville, Tennessee (population 7,900), the county hadapparently used DHS funds to acquire an all-terrain vehicle and a few defibrillators,

Theor Soc

one of which it placed in the high school gymnasium. Some critics charged that suchirrational allocations could be explained by the political influence of particularCongressional officials or by voting patterns of the respective regions. Otherscomplained that the new pools of federal money available through DHS encouragedentrepreneurial local officials to exaggerate or even invent security needs, sincedoing so would help them secure resources that they could use for a range ofprojects.

Journalistic exposés and complaints from political officials in urban areas sparkedan emerging conflict between what were seen as rational approaches to securitythreats (“risk”) on the one hand, and the illegitimate use of security resources forpolitical purposes (“pork”) on the other. “Pork” was shorthand for violation of thenorm of impersonal bureaucratic rationality. A particular target of criticism was theformula in homeland security appropriation bills that automatically allocated acertain percentage to each state, rather than distributing it according to the degree ofthreat a given area faced. Critics called on DHS to develop an objective, “risk-based”method of determining allocations–a reform resisted by congressional officials fromstates that benefited from existing arrangements.

The dispute over funding allocation mechanisms for urban security pointed to along-standing problem for federalism: how to equitably, efficiently, and effectivelydistribute funds for the provision of public goods from the federal government tolocalities? Overlapping with this question was a tension between politicalrepresentatives from large urban areas, who had longstanding concerns about federalneglect of cities, and those from suburban and rural districts. The tension wasnotably severe during the early 2000s, as politicians from big cities, primarilyDemocrats, charged that the Bush administration and the Republican Congress wereignoring their core needs. Although no one disputed that densely populatedmetropolitan areas faced heightened vulnerability, high-ranking members of theSenate supported a DHS funding formula mandating that 40% of the homelandsecurity budget be equally divided among the fifty states, leading critics such asCharles Perrow to argue that “the funding was almost exactly in reverse order of thethreat” (Perrow 2007)–though it was not clear what criteria such critics were using todetermine the actual order.

In response to controversy over the funding formula, in April 2003 Congressintroduced the Urban Area Security Initiative (UASI), which was intended tosupplement the security budgets of the most vulnerable cities. Formally, the goal ofthe program was to address the specific needs of “large high threat urban areas” inbuilding the capacity to “prevent, respond to, and recover from threats or acts ofterrorism” (Ridge 2003).2 The UASI program announcement spelled out fivecategories of authorized program expenditures: planning, equipment, training,exercises, and management and administration. In its initial iteration, UASI allocated$96 million in direct assistance to seven urban areas: New York, Washington, DC,Los Angeles, Seattle, Chicago, San Francisco, and Houston. But officials from otherlarge cities immediately complained that their constituents had been neglected

2 The DHS Office of Grants and Training provides a breakdown of the iterations of the various homelandsecurity grant programs on its website. http://www.ojp.usdoj.gov/odp/grants_programs.htm; accessed May15, 2007.

Theor Soc

without justification. Later in 2003, Congress introduced UASI II, which provided$506 million to thirty other eligible cities (ibid.). According to DHS, the cities inTiers I and II had been chosen and their relative allotments “determined by formulausing a combination of current threat estimates, critical assets within the urban area,and population density” (Department of Homeland Security 2003a; Department ofHomeland Security 2003b).

Thus, built into UASI was a decision mechanism that would ostensibly lead to theallocation of funds according to the relative risk faced by each jurisdiction. As earlyas the first round of UASI grants, however, officials complained about the operationof the funding mechanism. The problem was how to know which cities carried themost risk. Without agreed-upon criteria of evaluation, any politician could make acase for the priority of his or her own jurisdiction. Thus, in his introduction of theMayor of Detroit to a 2003 hearing of the Committee on Governmental Affairs on“Investing in Homeland Security,” Senator Carl Levin of Michigan argued, withregard to the list of eligible cities in UASI I:

The first round of these grants doesn’t add up in terms of threats. Detroit is thelargest U.S. border crossing for trade with Canada or Mexico. In fact, Canadais our largest trading partner with over $1 billion worth of goods and servicescrossing the border every day. More than 40% of that trade passes through theMichigan-Ontario border. Detroit has already been the site of several anti-terrorism probes, and it is a microcosm of all the complex issues that require abalancing of civil liberties and security needs (Levin 2003).

Levin added a point that would come up again: the classified nature of theDHS mechanism for determining risk: “Governor Ridge offered to share theclassified threat analysis information used for the grants, and I look forward toreviewing it.” At any rate, Ridge assured Levin that Detroit would receivesupport. And indeed, Detroit was among the cities made eligible for fundingunder UASI II. Since then, the city has been included as an eligible city in eachyear’s UASI allocations.

However, the legitimacy of DHS’s funding formula remained a source of dispute.In November 2003, New York Times columnist Dan Barry wrote that “the agencyused a formula that mixes facts and alchemy” instead of rational calculation. Barryquoted a New York City official: “The aid is not based on threat. It’s based onformulas that allow the government to spread it around the country. That’s politics;that’s not threat assessment” (Barry 2003). In February 2004, the Times reported thatmany critics of DHS, including Gov. Pataki and Mayor Bloomberg, had “chargedWashington with treating the money like pork-barrel spending, instead of doling itout based on actual risk” (Baker and Santora 2004). Pataki criticized the Senate forbeing “supportive of allocations based on so much to each state without regard to athreat-based analysis.”

The unspoken assumption of this critique was that there existed an objective wayof determining “actual risk,” one that DHS was ignoring due to political influence.The critique was based on a more general understanding of the characteristics oflegitimate decision-making about funding allocation in a bureaucratic agency: thatit should not be based on political exigency but on rational calculation. In the caseof UASI, DHS was vulnerable to a denunciation of partiality because it did not yet

Theor Soc

have a stable mechanism of impersonal calculation with which to defend itself(Porter 1995).

The accusation that DHS was engaged in pork barrel politics—as opposed toobjective risk analysis–became a common refrain in public criticisms of DHSbudgeting practices. As of early 2003, DHS had been directed by the White House todevelop a “uniform methodology” to establish priorities for managing risk, but ayear later there was no such methodology.3 In May 2004, Mayor Bloomberg testifiedto the 9/11 Commission that DHS counterterrorism funding was “pork barrel politicsat its worst” (Shenon and Flynn 2004). The 9/11 Commission Report, released2 months later, adopted this position, arguing that “Congress should not use thismoney as a pork barrel,” but rather should consider factors such as populationdensity, vulnerability, and the presence of critical infrastructure in allocating funds:“We recommend that a panel of security experts be convened to develop writtenbenchmarks for evaluating community needs” (9/11 Commission 2004).

The role of such “security experts”–typically located in think tanks and policyinstitutes–was to define the standards of rationality that would guide DHS budgetarydecisions. Thus, for example, in December 2004, think tanks based in WashingtonD.C., The Center for Strategic and International Studies (CSIS), and The HeritageFoundation, released an influential set of recommendations for the reform ofhomeland security policy, entitled DHS 2.0: Rethinking the Department ofHomeland Security. The report affirmed the distinction between risk and politics,arguing that in the current allocation process there were “no federal criteria for theminimum capabilities needed to protect an American community, no fundingformula that is based on risk analysis and divorced from politics, and no fundingsystem that can assure a sustained flow of funds for specified projects that areconsistent with the real security needs of the community and national strategicpriorities” (Carafano and Heyman 2004). Comparing the per capita homelandsecurity funds going to California ($5.03) with those to Wyoming ($37.94), theauthors agreed with the 9/11 Commission that “the current grantmaking process is indanger of becoming pork-barrel legislation.” Critics of DHS argued that suchallocation decisions favored Republican states and deliberately undercut large citiesrun by Democrats.

How, then, to move from politics to impersonal decision–from “pork” to “risk”?The think tank experts recommended the development of a standardized method tolink overall homeland security strategy to specific funding decisions: “Establishingnational performance standards for preparedness is essential to evaluating readiness,determining priorities, and targeting investments.” Indeed, the authors argued, thematter was pressing: DHS had estimated a completion date of 2008 for such amethodology, but “America cannot wait that long.” The agency, they instructed,should “deliver a comprehensive threat/vulnerability assessment to America’s topnational security decision makers by December of 2006. It should be based oncommonly accepted risk-assessment methodology. Spending and missions conducted

3 The 2003 National Strategy for the Physical Protection of Critical Infrastructures and Key Assets stated:“DHS … will develop a uniform methodology for identifying facilities, systems, and functions withnational level criticality to help establish … priorities.” (Office of the White House 2003, p. 23; cited inMoteff 2004, p. 2).

Theor Soc

by DHS could then be logically prioritized based on what is most vital to protect thenation” (ibid.). As we will see, however, identifying what constituted a “commonlyaccepted risk-assessment methodology” in the field of homeland security was notobvious.

The challenge was not merely improving the decision tool, but also finding onethat would satisfy the different players in an escalating political conflict. In a March2005 Hearing of the Senate Committee on Homeland Security and GovernmentalAffairs, DHS Secretary Michael Chertoff testified about the Department’s efforts tomove toward a risk-based methodology. Senators from small states that benefitedfrom the automatic funding formula were resistant to such a shift. For example, onequestion was whether there should be a population threshold of eligibility for UASIfunding. The Committee’s Ranking Member, Joseph Lieberman of Connecticut,criticized the existing population threshold of 225,000 as without grounds. “Thereare clearly some smaller cities in which there are real targets for terrorism. And Iurge you to review that requirement and basically make a threat assessment free ofan arbitrary threshold” (United States Senate 2005).

Another example of such conflict involved the question of whether minimumstate allotments should be replaced by risk-based methods. Senator Mark Pryor ofArkansas defended the formula of minimum state allotments on the basis of whatcritics might have called pork barrel politics, but which he considered job security:“I have had constituents who have spent their lives and dedicated themselves tohomeland security in Arkansas, and they are starting to hear rumors through DHSthat the State cannot count on that minimum.” But Pryor also defended the minimumallotments on the grounds of uncertainty–the incalculability of risk in a climate ofunpredictable threats:

Senator Pryor: Don’t we need to prepare ourselves for the next risk, not the lastrisk?

Secretary Chertoff: Absolutely. That is absolutely right.

Senator Pryor: And aren’t we assuming that because these happened in urbanareas before, it is going to happen in urban areas again?

Secretary Chertoff: I agree with you, we should not assume that…. I could notbe in more agreement with you that it would be a huge mistake for us to spendall our time fighting the last war and not thinking about the next one (ibid.).

Thus, the “real security needs” that DHS 2.0 had pointed to were not yetstabilized. For defenders of the existing system, experts and big city critics had anoverly narrow conception of the threats to be protected against: from theirperspective, vulnerable targets–such as agriculture, water, and transportationsystems–were not limited to major urban centers such as New York and Washington.There were no impersonal means to adjudicate such a dispute: without agreed-uponstandards of measurement, it was impossible to compare the relative level of threatand vulnerability of various cities and regions, and so there was no definitive way todetermine whether homeland security funds were being allocated objectively, that is,independently of “political” considerations. What it would mean for the agency toact rationally was still in formation.

Theor Soc

Toward reform

The failed government response to Hurricane Katrina in August and September of2005 intensified the debate over urban security, as experts questioned whether DHShad neglected predictable hazards. It also brought in new voices, as journalists,research organizations, think tanks, and scholars began searching for the structuraland conjunctural causes of the breakdown at DHS. As stories about the chaoticresponse from DHS circulated through the media, the agency was publicly brandedas incompetent. Critics charged that, through wasteful spending and a myopic focuson terrorism, DHS had actually made American cities more vulnerable to disasters.The problem of urban security became a leading public issue, and DHS faced strongpressure to reform.

In December 2005, the 9/11 Commission published its “Final Report” on theimplementation of its recommendations for the reform of homeland security policy.Although the report had no legislative authority, it was released to the media as away of trying to influence administration policy through public pressure (Adut2008). The Final Report gave DHS an “F” in its evaluation of progress on therecommendation to “allocate homeland security funds based on risk” (9/11 PublicDiscourse Project 2005). Soon afterwards, DHS introduced its new, “risk-based”methodology for calculating UASI grants. The agency’s January 2006 press releaseannouncing the changes listed 35 eligible urban areas, including 95 cities withpopulations of 100,000 or more. DHS explained how these cities had been chosen:“The fiscal year 2006 UASI list of eligible applicants and recipients is determinedthrough a robust risk formula that considers three primary variables: consequence,vulnerability, and threat” (Department of Homeland Security 2006a).

In an accompanying “Discussion,” DHS highlighted a major change in thegranting process: for the first time, instead of being given an automatic allotment,applicants would have to “submit an Investment Justification requesting resources toundertake specific initiatives that tie to the Urban Area Homeland Security Strategyand the National Preparedness Goal” (Department of Homeland Security 2006b). Inthis way, individual jurisdictions would be encouraged to define and manage threatsaccording to norms of efficiency and efficacy. In the model of “entrepreneurialgovernment,” competition among jurisdictions and evaluation by experts wouldensure the rational distribution of security resources in a federal system.4

The “Discussion” extolled DHS’s risk-based methodology: “Tremendous gains havebeen made in both the quality and specificity of information and analysis andincorporated within the model, yielding the most accurate estimation possible of therelative risk of prospective grant candidates” (ibid). The term “relative risk” wouldbecome central to later discussions: its determination depended on a process of countingthe numbers of critical assets located in different jurisdictions. DHS’s announcementexplained how the model had been improved: “The formula has progressed from asimple count of ‘high’ and ‘low’ criticality and numbers of threat reports in FY 2003 to afully risk-based computation that is attack-scenario based and uses infrastructure-specific vulnerability and consequence estimates” (ibid). Such information about

4 On the rise of “entrepreneurial states,” see Eric Klinenberg, Heat Wave: A Social Autopsy of Disaster inChicago (2002: Chapter 4.).

Theor Soc

vulnerability and consequences was based on an impressive amount of data: “The FY2006 approach evaluates risk to well over 120,000 specific infrastructures.” How theseinfrastructures had been tabulated, and their risk assessed, was not made clear in thedocument, but would shortly become an object of intense scrutiny.

The next day, the Times editorial page celebrated DHS’s announcement,proclaiming: “Risk Wins a Round Over Politics” (New York Times 2006b). However,the editorial added that: “Congress, thanks to the influence of small-state senators,still lets politics play a disturbingly large role in the distribution of otherantiterrorism funds,” citing Senator Susan Collins of Maine, Chair of the SenateHomeland Security Committee, as a prime example of those defending the minimumallotment formula. Washington D.C.-based think tank experts also voiced support ofDHS’s new direction. For instance, American Enterprise Institute’s Veronica deRugy wrote that DHS had finally addressed the problems of the current system:“DHS will use more and better data in its assessments and has refined its criteria forassessing risk and deciding which cities will be eligible for funding” (de Rugy2006b). Moreover, de Rugy thought that small cities should be completely excludedfrom funding: “DHS should go a step further and focus solely on the nine citiesidentified by industry experts as the most likely terrorist targets.”

But along with these endorsements of the new approach, there was a hint of futureturmoil. Again, powerful senators resisted the move toward risk-based calculation. OnJanuary 30, Senator Lieberman wrote to Secretary Chertoff to express his “deepdisappointment” about the exclusion of Connecticut cities such as New Haven(population 124,000) from eligibility, as well as his “concerns about the credibility ofthe still evolving ‘risk’ analysis methodology developed by DHS.” Liebermanquestioned the scientific claims of the new risk assessment methodology: “Risk analysisis an art, not a science; and it is an inexact art at best” (Lieberman 2006). In other words,according to Lieberman, there were no objectively rational grounds for shiftinghomeland security resources from Connecticut to New York City and Washington, D.C.

Legitimacy crisis I: relative risk

While advocates of risk-based allocation were initially pleased with DHS’s newmethodology, they were taken by surprise by the eventual results of the calculativeprocess. At the end of May 2006, DHS released the results of the year’s UASI grantscompetition at a press conference in Washington D.C. George Foresman, the newlyappointed Undersecretary for Preparedness, opened the event with a reference to theongoing debate over the agency’s funding mechanism: “This year really representsthe first year that we are holistically as a department getting away from simple,straight forward, formula-driven grants processes to true risk-based processes”(Foresman and Henke 2006).5 But the announcement led to pointed questions about

5 Post-Katrina, Foresman also noted that the grants would serve the goal of “all-hazards” preparedness:“The dollars that we continue to provide down to states and down to communities represent dollars ofinvestment toward capabilities to deal with the Katrina-scale events, whether it’s caused by a breach in alevee by Mother Nature, or whether it’s caused by a breach in a levee as a result of a terrorist event. Andso, frankly, at the end of the day, we just simply see these as wise investments towards being able tomanage our national risk” (Foresman and Henke 2006).

Theor Soc

what exactly “true risk-based processes” meant. At one point, a reporter noted thatNew York City’s UASI grant had been cut from $207 million in 2005 to $124million in 2006, and asked the official in charge of the program, Tracy Henke, for anexplanation. Henke referred haltingly to DHS’s new methodology, and to budgetaryconstraints:

Question: Why is there such a dramatic decrease in the money that New Yorkis getting if this is based purely on risk?

Assistant Secretary Henke: Understand it’s based on risk and effectiveness.Also understand that we have less money this year … approximately $125million less for the Urban Area Security Initiative Program overall (ibid.).

The reporters were not satisfied with Henke’s answer, given that many other citieshad seen their funding increase, and they raised the issue again:

Question: How you can justify what appears to be a substantial drop in NewYork City’s funding if a threat-based matrix is being applied? …The cut toNew York seems to far surpass the cut in the overall program. And I’mwondering why.

Assistant Secretary Henke: New York…is just one area. We have to look at thenation as a whole and understand that we are—we’re only as strong as ourweakest link (ibid.).

Foresman then stepped in to address the question in terms of the idea of “relativerisk”: it was not that New York’s overall risk was less, but that DHS had come tounderstand better the relative risk of other cities. On the one hand, he explained,New York was now less vulnerable because the city had already done such a goodjob of addressing its risk; on the other, more data about other cities had led to moreawareness of their needs. “Other metropolitan areas are getting better at managingand measuring their own risk and being able to quantify it,” Foresman explained.“We’re getting better at the federal level…. We have hoards of data on New YorkCity. We didn’t have that same level of data on other metropolitan areas.… And sothe data set, if you will, that informs the risk ranking for the non-New York areas isbetter. Therefore, in some cases, you’re seeing that they’re getting additional dollars”(ibid.).

This explanation did not satisfy the assembled reporters either. Given recentscandals at DHS, they were highly attuned to any evidence of “pork,” particularlywhen it came at the expense of the major metropolitan areas where the mostinfluential newspapers and television stations are based. One reporter pointed outthat Louisville’s appropriation had increased to $8 million, and that the Chairman ofthe House Homeland Security Subcommittee on Appropriations was from Kentucky.This looked suspicious, and he asked if “political considerations” were part of theallocation process. “Political considerations play no part in the allocation process—none whatsoever,” Foresman insisted. “We understand much better today than wedid twenty-four months ago what critical assets are in Louisville. And right off thetop of my head … I don’t happen to know what they are” (ibid.).

Foresman’s admitted lack of detailed information did not help his credibility. Asthe press conference continued, journalists remained either skeptical of or confused

Theor Soc

by the particular rationale Foresman was trying to articulate: that the disproportion-ate cuts to New York City and Washington DC were due not to “politics” but to anincreased knowledge about the “critical assets” of the rest of the country’s cities.Here, another problem emerged: DHS insisted that the sensitive nature of thenation’s security risks prevented it from releasing a complete list of critical assets.Since the DHS decision-making process was classified, it remained opaque—to thepublic as well as to social analysts–whether their budget decisions had in fact beenmade using “objective” methods. Without details on how the calculations had beenmade, the public would have to take DHS’s word that the process was objective anddisinterested, rather than “political.” But after Katrina, the department lacked thecredibility to keep further inquiries at bay simply by citing “classified information.”

Media reaction from New York and Washington DC over the next few days washighly critical. The June 2nd New York Times editorial argued that DHS’ssubstantive criteria for evaluating the proposals were flawed. New York, it seemed,had emphasized routine personnel costs–such as police overtime–in its application,whereas the grant was intended for “capacity building” expenses such as equipmentpurchase, training, and exercises. “The panels appear to have been too focused onpolitics, and not enough on safety,” wrote the Times editors, suggesting that theRepublican White House was penalizing big cities for partisan reasons. “Theresistance to financing operating costs, like police overtime, ignores the fact thatday-to-day groundwork does the most to combat terrorism.”

In response to mounting criticism, on June 7th Secretary Chertoff published anop-ed piece in the Times entitled “New York, You’re Still No. 1.” He pointed out thatNew York still received the most funds in the program, and defended UASI’semphasis on capacity building projects rather than ongoing personnel needs: “we arelooking to pay for new equipment and projects that increase the nation’s overallpreparedness,” he said, rather than pay for routine and recurring operating expensessuch as police overtime (Chertoff 2006). At stake in this discussion of whetherfederal block grants could be used to pay for personnel was a broader issue ofgovernmental responsibility for the provision of urban security: was the kind ofsecurity provided by police the obligation of cities or of the federal government?Think tank experts weighed in on both sides. De Rugy of the American EnterpriseInstitute took the anti-big government position: “Grants are not entitlements. Localgovernments should bear the primary responsibility for public safety” (de Rugy2006a). In contrast, Michael O’Hanlon of Brookings defended federal funding forlocal personnel, echoing the Times’ emphasis on “groundwork”: “Homelandsecurity is an activity that, like infantry combat, requires a lot of boots on theground.” (O’Hanlon 2006).

In any case, Chertoff’s commentary failed to defuse the controversy. Criticismover the cuts to New York and Washington DC escalated, as leading national mediaoutlets railed against corruption in DHS and influential Congressional officialsorganized hearings before the House Committee on Homeland Security. On June 21,Mayors Bloomberg of New York and Williams of Washington DC testified in theCapitol, and Bloomberg used the opportunity to lambaste DHS: “I have said then, Isay now: this was a stab in the back to the city of New York. It was indefensible. Itwas disgraceful. And to me it raises very, very real questions about the competencyof this department in determining how it’s going to protect America.” Williams was

Theor Soc

more circumspect, focusing on the technical details of the DHS process. He voicedsupport in principle for the agency’s effort to “move to a more objective, transparentand risk-based approach to the allocation of scarce resources,” but asked pointedlyabout the rationality of the criteria that had led to cuts to DC’s funding. Theoutcomes were “astounding,” Williams said, and “the viability of the analysis isquestionable” (U.S. House of Representatives 2006).

Secretary Foresman testified after the mayors. Once again, he tried to explain the“relative risk” rationale underlying DHS’s analysis, and why it had led counter-intuitively to cuts in the allocation to New York and Washington, DC. In hisexplanation of the agency’s funding decision, Foresman relied heavily on one aspectof the DHS risk assessment methodology–the inventory of critical assets as part of adetermination of infrastructural vulnerability. He emphasized the distinction betweenthreat and risk. To calculate the latter, one needed data not only on the likelihood ofan attack–what he called the “threat”–but also on the number of vulnerable criticalassets in the country:

Threat is [just] one element of risk analysis. For instance, last year 11,300critical facilities nationwide were factored into our risk analysis. This year,there were more than 260,000. These are facilities that, if attacked, could causegrave impacts on those who live and work inside or nearby or could cause anational level impact, similar to what Mayor Bloomberg described an attack onNew York City doing (ibid.).

Foresman here invoked an avalanche of newly acquired data to explain DHS’sfunding decision: 260,000 critical facilities instead of 11,300.6 This aspect of theDHS rationale was not much broached in public discussions–even though it wasperhaps the department’s main defense of its allocation decision. Again, Foresmanwas describing not threat, but a different aspect of the risk calculus: vulnerability,which was, according to the DHS risk assessment formula, independent from thequestion of the likelihood of attack. It was from this vantage that an increase in thenumber of critical assets counted in Louisville, even in the absence of anyknowledge about new threats to the city, could lead to an increase in its quantity ofrelative risk, and therefore in its share of UASI funds.

Where did this definition of risk come from? It was DHS’s answer to thebureaucratic demand, described above, that the agency develop a “commonlyaccepted risk-assessment methodology” to guide its allocation decisions. DHS’smethodology, adapted from technical fields such as civil engineering and operationsresearch, calculated risk as “a function of consequence, vulnerability, and threat.”7

Given that the likelihood–the “threat”–of a terrorist attack in a place was impossible

6 Note that the number 260,000 is confusing, given the earlier number of 120,000 and the later number of77,000.7 The methodology was explained in more detail in the 2006 National Infrastructure Protection Plan(NIPP). NIPP defined consequence as “the range of loss or damage that can be expected from anyincident”–including terrorist attacks and natural or manmade disasters. Vulnerability was defined as “thecharacteristics of an asset, system, or network’s design, location, security posture, process, or operationthat render it susceptible to destruction, incapacitation, or exploitation by mechanical failures, naturalhazards, terrorist attacks, or other malicious acts.” Finally, threat indicated “an estimated value thatapproximates the likelihood that a specific asset, system, network, sector, or region will suffer an attack oran incident.”

Theor Soc

to calculate quantitatively, risk assessment had to focus on other aspects, such as thevulnerability of potential targets, to come up with quantitative measures of risk.8 Inprinciple, then, these three factors were brought together to assess the risk to a givenlocation or asset. According to this framework, a necessary precursor to thecalculation of risk was the construction of an inventory of the nation’s criticalinfrastructure. This inventory was called the “National Asset Database” [NADB]. Itwas this massive database of critical infrastructure and key assets that, Foresmanclaimed, grounded DHS’s calculations of “relative risk,” and therefore the agency’s2006 UASI allocation decisions. What critics called “pork,” DHS insisted, was infact the outcome of a complex, but rational and objective system of calculation.

Legitimacy crisis II: “critical assets”

At this point, the controversy over UASI might have died down, as journalists lostinterest in the details of the agency’s calculative methods in the face of its publicdisplay of numbers. It turned out, however, that the asset database was far from areliable source of data on the nation’s critical assets. Soon after the hearings, adevastating critique of DHS’s methodological claims came from a surprising place:within DHS itself. In June 2006, the DHS Office of the Inspector General released areport on the actual contents of the NADB. According to the report, “the presence oflarge numbers of out-of-place assets taints the credibility of the data” (DHS Office ofthe Inspector General 2006; cited in Wang 2006). Examples of such out-of-placeassets included the Bourbon Festival, High Stakes Bingo, Frontier Fun Park, andMuzzle Shoot Enterprises. It turned out that Indiana had significantly more “criticalassets” listed than New York or California. The Inspector General concluded that theNADB was not yet ready for use in grant-making decisions: “In light of the variationin reporting between various sectors and states as well as the lack of detailedinformation on sites, we are not confident that the NADB can yet support effectivegrant decision-making” (ibid). And, according to the report, there was no immediatehope that the NADB would improve enough to make rational risk assessmentpossible.9

8 According to the NIPP, the DHS risk management framework combined multiple elements to produce a“rational assessment” of risks that would then drive the prioritization of infrastructure protection efforts.9 “We cannot predict when IP [infrastructure protection] will have both the NADB and risk assessmenttools to provide comprehensive risk assessment of the country’s critical infrastructure and key resources.”(p. 1) The Inspector General’s analysis of the process through which the NADB had been constructedhelps to explain the presence of so many odd, “out-of-place” assets in the database. In the summer 2003,DHS had identified 160 “nationally critical assets” as part of Operation Liberty Shield, a defensivecounterpart to Operation Iraqi Freedom. This list was then expanded to 1,849 assets as part of the“Protected Measures Target List” (PTML). Also in 2003, in a separate program, the Office of DomesticPreparedness (ODP) asked states to provide data on critical assets as part of a state self-assessmentprogram. The criteria in the data call were quite general, as states were told: “you should consider anysystem or asset that if attacked would result in catastrophic loss of life and/or catastrophic economic loss.”(p. 8) In July 2004, the Office of Infrastructure Protection (IP) initiated a second data call—this time withsomewhat more specific guidelines on the identification of critical infrastructure, but again encouragedstates to submit any asset they thought was important (p. 12). Over the following year, states identified48,701 assets. By 2006, the database included 77,069 items.

Theor Soc

The timing of the release of the Inspector General’s Report, just after the UASIcontroversy had erupted, and after Foresman had defended DHS’s decisions on thegrounds of “relative risk” (which required the counting of “critical assets”), couldnot have been worse for DHS. A New York Times front-page article of July 12brought the issue to national attention. The article began with a list of some of the“out-of-place” assets the report had described: “It reads like a tally of terrorist targetsthat a child might have written: Old MacDonald’s Petting Zoo, the Amish CountryPopcorn factory, the Mule Day Parade, the Sweetwater Flea Market and anunspecified ‘Beach at End of a Street’” (Lipton 2006). Such articles multiplied. AsMSNBC reported, “the department’s database of vulnerable critical infrastructureand key resources included the Old MacDonald’s Petting Zoo near Huntsville, Ala.,a bourbon festival, a bean festival and the Kangaroo Conservation Center inDawsonville, Ga.” (MSNBC 2006).

Significantly, the Times article linked the database to the UASI controversy–eventhough such a link had not been made in the Inspector General’s report: “The databaseis used by the Homeland Security Department to help divvy up the hundreds ofmillions of dollars in antiterrorism grants each year, including the program announcedin May that cut money to New York City and Washington by 40%, while significantlyincreasing spending for cities including Louisville, Ky., and Omaha.” New Yorkpoliticians responded with predictable outrage. Senator Chuck Schumer said: “Nowwe know why the Homeland Security grant formula came out as wacky as it was,” hesaid. “This report is the smoking gun that thoroughly indicts the system.”

The Times also quoted Foresman, who argued that the asset database was notcentral to allocation decisions, but was “just one of many sources consulted indeciding antiterrorism grants.” But the article added to existing public perceptions ofwidespread incompetence at DHS, exemplified by its color-coded alert system, ducttape advice, and its disastrous response to Hurricane Katrina. Elements of theInspector General’s report soon reached the world of late night comedy. ThusStephen Colbert quipped: “I, for one, don’t know how democracy could surviveonce somebody takes out the Sweetwater Flea Market” (Wang 2006). Once again,DHS appeared to be rife with incompetence, and its reputation made it alaughingstock.

Small states sought to defend themselves from the outcry. The MissouriHomeland Security Coordinator said that the state had already updated its list ofcritical infrastructure to ensure that “our most vulnerable resources would beprotected. …There is nothing comical about any of the sites we have submitted toour national counterparts” (State of Missouri Department of Public Safety 2006). Aneditorial in Indiana defended the grant program: “many people are having a lot offun and venting a lot of outrage over the fact that U.S. Department of HomelandSecurity’s National Asset Database lists Indiana No. 1 for the number of sitesvulnerable to terrorist attack,” but perhaps the fault lay with states like New York forfailing to include their relevant assets and therefore losing out in the federalistsystem of competition for resources: “If the New York list fails to include suchthings as the Empire State Building or the Statue of Liberty, and Indiana’s includessuch things as Lehman’s Amish Country Popcorn in Berne and lots of farms andnursing homes, it hardly seems fair to blame the federal government” (Fort WayneNews-Sentinel Editors 2006).

Theor Soc

DHS remained under siege. In a July 21 op-ed published in USA Today, Col.Robert Stephan of the DHS Infrastructure Protection Office defended the assetdatabase and the risk assessment process. “We use a variety of tools to identify andassess the risk to our nation’s infrastructure,” he said. “One is the National AssetDatabase. It is not, as has been reported, a list of the most critical assets in our nationnor does it drive the department’s funding decisions. The database is merely thefirst step in creating a ‘phonebook’ of more than 77,000 facilities, assets andsystems across the nation, so we can conduct more detailed risk analysis”(Stephan 2006). According to Stephan, the list of 77,000 assets was then filteredthrough a careful process of assessment: “Only assets that are the most vulnerableand represent significant public health, economic and national defense consequen-ces make the final cut. This carefully screened set of target listings, roughly 600nationwide, drives our decisions on the focus of operations, resource investmentsand grants.”

Yet this latter list was classified, and outside analysts were unable to scrutinize thedecision tool that was, DHS maintained, at the heart of its rational evaluation system.Stephan insisted that: “These target lists are classified so they won’t become a roadmap for our enemies,” and accused the DHS Office of the Inspector General ofwillfully misrepresenting the use of the database in order to make light of thework of the Infrastructure Protection office. “In a recent report, the department’sinspector general seems to have missed the purpose of the database entirely.…The inspector general focused on raw, unfiltered data to create national hype….The inspector general attempted to turn a deliberative, analytic process into anational mockery” (ibid.). However, the opacity of DHS’s deliberative process–especially when combined with the agency’s lack of credibility–undermined theeffectiveness of these defenses. Gaining public legitimacy for its allocationdecisions required not only “objective” calculation, but also transparency aboutthe details of the calculation process.

Claims and counter-claims continued among various branches of government. Inthe fall of 2006, the Congressional Research Service (CRS) delivered its own reporton the database that questioned the “phonebook” defense provided by DHS over thesummer. Citing Stephan’s argument that the “first day that al-Qaeda attacks aconvenience store chain times a dozen across the country ... we better have some ofthose things in the database so that we know what that universe of things is that wehave to worry about,” the CRS Report responded: “that assets may be critical undersome circumstances and not others … seems to conflict with the statutory definitionof critical infrastructures” (Moteff 2006). Moreover, the CRS report noted with acertain puzzlement that, from reading DHS documents, it was not clear what exactlythe National Asset Database actually had to do with the allocation decisions of theoffice in charge of the UASI program: “For the FY2006 cycle, ODP [Office ofDomestic Preparedness] indicated it had evaluated over 120,000 specific infrastruc-ture assets, but the National Asset Database only contains 77,069 assets. It wouldappear that ODP’s grant-making process operates independently from both theNational Asset Database and the National Infrastructure Protection Plan” (ibid).DHS had been hoisted on its own petard: while Foresman had (inaccurately) pointedto the massive numbers of critical assets in the NADB as the rational basis for DHSfunding decisions, the authority of these numbers crumbled when subjected to public

Theor Soc

scrutiny. To avoid the close inspection of its calculative process an agency needscredibility–something DHS sorely lacked.

Recalibration

Undersecretary Foresman’s assurance that “the process also continues to mature andimprove” did not assuage DHS’s critics. By the fall, Tracy Henke, the director of theoffice overseeing the UASI grants program, had resigned, and Chertoff hadpromised that DHS would rethink its method of allocating funds for urban areasecurity. By early 2007, it appeared that New York and Washington D.C. hadsucceeded in their efforts to reshape the UASI funding mechanism. In the nextiteration of the program, announced in January, urban regions were broken into twotiers with different systems of allocation for each. The six Tier I cities (Los Angeles/Long Beach; Bay Area; National Capital Area; Chicago; New York/New Jersey/Newark; and Houston) would receive $411 million in UASI funds, while 39 Tier IIcities would share $336 million (Department of Homeland Security 2007). NewYork would no longer be in direct competition for resources with small cities such asLouisville. Moreover, Chertoff reversed his position on the use of homeland securityfunds for local counter-terrorism personnel, and allowed up to one quarter of thefunds for Tier I cities to be used for what mayors and the Brookings Institute called“boots on the ground” security.

Meanwhile, DHS had significantly revised the National Asset Database–fromover 77,000 critical assets to around 2,100. Popcorn stands and petting zoos hadbeen removed. Chertoff explained the department’s more restrictive definition: “Thecritical assets are those which have, from an economic standpoint and otherstandpoint, a regional or national significance. A major dam, for example, that ifsomething were to happen to would have a catastrophic impact. That’s an exampleof a piece of critical infrastructure” (Chertoff 2007).

But the question of how much homeland security funding would be based on a“risk-based” methodology remained. Mayor Bloomberg testified to the SenateHomeland Security and Governmental Affairs committee, now chaired by JosephLieberman, in January: “Time and time again, our calls for fully risk-basedhomeland security funding have been ignored. Instead, we have seen large sums ofhomeland security money spread across the country like peanut butter. More than $3billion has been distributed in this irrational way so far” (Chan 2007). The issue wasnot only whether funds would be allocated through a risk-based method, but alsohow risk would be calculated. Bloomberg attacked DHS’s technique of calculatingrisk based on the vulnerability of critical assets rather than on the likelihood of astrike on economically or symbolically significant places. He drew a distinctionbetween “risks,” which are everywhere, and “targets,” which “are what the enemiesof this country will focus on” (ibid) in the next attack.

Despite such concerns, New York City officials seemed mostly pleased with thechange in DHS policy. Rep. King said: “I’m going to watch it, but based on theexplanation I’ve gotten, I think it’s going to work” (Barrett 2007). Presumably, thissatisfaction was due not to an improvement of the technical proficiency of riskassessment, but to the changes in the program that would likely provide more

Theor Soc

resources to New York City in the next round. Perhaps a combined technical-political solution that would satisfy DHS’s critics had been found.10

Conclusion

In the UASI controversy, a relatively new and weak bureaucracy responsible forhomeland security was unable to defend its allocation decisions on the grounds ofimpersonal calculation, and was forced to shift its policy in response to politicaloutcry. As critics of DHS issued public denunciations of the process through whichDHS was allocating resources, the agency sought to construct a new decision toolthat would fend off accusations of partiality and incompetence. In this sense, thecontroversy directs attention to the ways that legitimacy disputes are adjudicated andstabilized according to the norms of bureaucratic rationality, especially when theimplementation of such norms is a matter of uncertainty. Transparent and sharedstandards of measurement must be in place to make objective, “de-politicized”calculation possible.

More broadly, the UASI controversy raised a number of questions around thedefinition of “homeland security”: is it defense against terrorist attacks, or does itcomprise protection against a broader set of threats, such as natural disasters? Whatexactly is to be protected: population centers or “critical assets”? And who isresponsible for deciding on priorities–the federal government, states, or localjurisdictions? As a later Congressional Research Service report concluded: “It isclear … that risk management remains relatively immature in its application to thehomeland security field” (Massey et al. 2007). In other words, objective decisionprocesses could not be stabilized until a political agreement had been reached–among Congress, DHS, and big city mayors–as to the needs and norms of homelandsecurity. While the UASI controversy was staged as conflict between risk and pork,or rationality and politics, it is better seen as a case of political struggle over how todefine a concrete form of bureaucratic rationality in the context of high uncertaintyand an urgent sense of threat. In the end, although the debate died down, this was notbecause DHS had achieved a “more objective” allocation process, but rather becauseit had incorporated external critique into its methodology in order to create a moreresilient decision tool.

In the debate over urban security, experts played a central role in shaping whathomeland security would practically mean, by introducing the demand for anobjective decision tool. However, this demand did not close off political discussion;rather, it directed political struggle into the very construction of this device. DHS didnot have the power to impose its own understanding of “risk assessment” on otherpolitical actors, for two reasons: first, the agency itself was weak and internallydivided; and second, the federalist system of sovereignty grants significant power to

10 In 2008, DHS used a similar allocation scheme, and it was widely accepted. Chertoff tried to strike apolitical balance in his public statement about UASI funding decisions: “I don’t believe all the moneyshould go to the top five or six cities, because I don’t think that’s where all the risk is, but we think most ofit ought to go there, and then some of it ought to go into, you know, the next 30. And yet I get confrontedby people in city number 150 or 200 and they say, well what about my city; well, we have this here, wehave that here? And they’re right; you know, there is a limitless amount of risk out there.” (Chertoff 2008).

Theor Soc

states and localities. External critics from big cities strategically misunderstood thetool DHS had constructed, and emphasized the critique of “pork” in order to forcethe agency to recalibrate its risk assessment mechanism in a way that would be moreamenable to their interests.

The UASI controversy raises an interesting tension for the analyst of bureaucraticrationality. On the one hand, in this situation there seems to be no stable ground fromwhich a purely objective decision, outside of politics, can be made. So then–it seemsthat everyone is engaged in “politics.” But on the other hand, it is a distinctive kindof politics–one that is structured by the problem of objectivity. DHS’s capacity tomake legitimate decisions depended on its successful display of objectivity; andconversely, external challenges to such decisions required undermining the agency’sclaim to objectivity through the claim that such decisions were really shaped by“politics.” Each protagonist sought to draw a line between politics and bureaucraticrationality, and to place itself on the side of reason. From the vantage of the analyst,it is not a question of saying that this is all “just” politics, then, but rather that thevery basis of rational decision is an object of political struggle. From this case, wecan see that in contemporary struggles over security resources, the distinctionbetween “rationality” and “politics” does not exist a priori, but rather is defined inthe concrete political and technical struggle over the creation of a decision tool.

Acknowledgments The authors thank the New York University Center for Catastrophe Preparednessand the NYU International Center for Advanced Study for funding this research, and the Fritz Institute foradditional support. Mark Treskon provided excellent research assistance. Stephen Collier, Steven Epstein,Gil Eyal, Ted Porter, Akos Rona-Tas, Mitchell Stevens, and Caitlin Zaloom offered helpful comments onearly drafts of the article, as did members of the NYU Urban Studies Seminar, the UCSD CultureWorkshop, and members of the audience at an American Sociology Association panel on Risk.

References

Adut, A. (2008). On scandal: moral disturbances in society, politics, and art. New York: CambridgeUniversity Press.

Amadae, S. M. (2003). Rationalizing capitalist democracy: The cold war origins of rational choiceliberalism. Chicago: University of Chicago Press.

Baker, A., & Santora, M. (2004). First steps and first controversies for state security agency. New YorkTimes. February 14.

Barrett, D. (2007). U.S. worried about homegrown terrorists. Associated Press, January 5.Barry, D. (2003). The bull’s-eye vs. The pork barrel. New York Times. November 19.Beck, U. (1999). World risk society. Oxford: Blackwell.Carafano, J. J., & Heyman, D. (2004). DHS 2.0: Rethinking the Department of Homeland Security.

Heritage Foundation and Center for Strategic and International Studies Joint Report. December 13.Chan, S. (2007). Bloomberg Criticizes Security Fund Distribution. New York Times. January 9.Chertoff, M. (2005). Testimony by Secretary Michael Chertoff Before the House Homeland Security

Committee. Department of Homeland Security. April 13. http://www.dhs.gov/xnews/testimony/testimony_0034.shtm

Chertoff, M. (2006). New York, You’re Still No. 1. New York Times. June 7.Chertoff, M. (2007). Remarks by Secretary Michael Chertoff at a Press Conference on the Fiscal Year

2007 Homeland Security Grant Program. Department of Homeland Security. Office of the PressSecretary. January 5. http://www.dhs.gov/xnews/releases/pr_1168039350894.shtm

Chertoff, M. (2008). Remarks by Secretary of Homeland Security at the Center for Strategic andInternational Studies' Homeland Security Strategy Forum. March 20. http://www.dhs.gov/xnews/speeches/sp_1209147825509.shtm. Accessed September 30, 2008.

Theor Soc

Collier, S. J., & Lakoff, A. (2007). Distributed Preparedness: The Spatial Logic of Domestic Security inthe United States Environment and Planning D: Society and Space 25.

de Rugy, V. (2006a). Is Your Town Safe From Terrorists? Reason Online. June 8.de Rugy, V. (2006b). The Scarcity of Security. American Enterprise Institute. January10.Department of Homeland Security (2003a). FY 2003 Urban Areas Security Initiative (UASI) Grant

Program I. http://www.ojp.usdoj.gov/odp/docs/fy03uasigrant.pdfDepartment of Homeland Security (2003b). FY 2003 Urban Areas Security Initiative (UASI) Grant

Program II. http://www.ojp.usdoj.gov/odp/docs/UASIIIFY03GrantAppFinal.pdfDepartment of Homeland Security (2006a). DHS Introduces Risk-based Formula for Urban Area Security

Initiative Grants. January. http://www.dhs.gov/xnews/releases/press_release_0824.shtmDepartment of Homeland Security (2006b). Discussion of the FY 2006 Risk Methodology and the Urban

Area Security Initiative. Preparedness Directorate, Office of Grants and Training. www.ojp.usdoj.gov/odp/docs/FY_2006_UASI_Program_Explanation_Paper_011805.doc

Department of Homeland Security (2007). Overview: FY 2007 Homeland Security Grant Program. Officeof Grants and Training. January 5. http://www.ojp.usdoj.gov/odp/docs/fy07_hsgp_overview.pdf

Department of Homeland Security Office of the Inspector General (2006). Progress in Developing theNational Asset Database. OIG-06-40. June.

Eggen, D., & Sheridan, M. B. (2006). Anti-Terror Funding Cut In D.C. and New York: HomelandSecurity Criticized Over Grants. Washington Post, June 1.

Espeland, W. N. (1998). The struggle for water: Politics, rationality, and identity in the AmericanSouthwest. Chicago: University of Chicago Press.

Espeland, W. N., & Stevens, M. L. (1998). Commensuration as a Social Process. Annual Review ofSociology 24.

Esposito, R. (2006). No Icons, No Monuments Worth Protecting. ABC News. June 1. http://blogs.abcnews.com/theblotter/2006/06/no_icons_no_mon.html

Falkenrath, R. (2001). Problems of preparedness. International Security, 24(4), 147–186.Foresman, G., & Henke, T. (2006). Press Conference by Under Secretary George Foresman and Assistant

Secretary Tracy Henke on FY 06 Homeland Security Grant Program. Department of HomelandSecurity. May 31. https://www.dhs.gov/xnews/releases/press_release_0922.shtm

Fort Wayne News-Sentinel Editors (2006). No. 1 on al-Qaida’s List? After We’ve Had Fun With theNational Asset Database, Let’s Make it a Better Tool. Fort Wayne News-Sentinel. July 14.

Giddens, A. (1990). The consequences of modernity. Stanford: Stanford University Press.Jasanoff, S. (1986). Risk management and political culture: A comparative analysis of science. New York:

Russell Sage Foundation.Kettl, D. (1987). The regulation of American Federalism. Baltimore: Johns Hopkins University Press.Klinenberg, E. (2002). Heat wave: A social autopsy of disaster in Chicago. Chicago: University of

Chicago Press.Klinenberg, E., & Frank, T. (2005). Looting Homeland Security. Rolling Stone. December.Lakoff, A. (2007). Preparing for the next emergency. Public Culture, 19(2), 247–272.Levin, C. (2003). Investing in Homeland Security: Challenges Facing State and Local Government. Press

Release. May 15. http://www.senate.gov/~levin/newsroom/release.cfm?id=210974Lieberman, J. (2006). Lieberman urges DHS to explain why New Haven, SW Connecticut are Ineligible

for High Risk Homeland Security Grants. Press Release. February 3.Lipton, E. (2006). Come One, Come All, Join the Terror Target List. New York Times. July 12.Massey, T., O’Neil, S., & Rollins, J. (2007). The Department of Homeland Security’s Risk Assessment

Methodology: Evolution, Issues, and Options for Congress. Congressional Research Service.February 2.

Moteff, J. (2004). Risk Management and Critical Infrastructure Protection: Assessing, Integrating, andManaging Threats, Vulnerabilities and Consequences. Congressional Research Service. September 2.

Moteff, J. (2006). Critical Infrastructure: The National Asset Database. Congressional Research Service.September 14.

MSNBC (2006). Inspector: Homeland Security Database Flawed. MSNBC News Service. July 12. http://www.msnbc.msn.com/id/13822662/

New York Times (2006a). Pork 1, Antiterrorism, 0. Editorial. New York Times. June 2.New York Times (2006b). Risk Wins a Round Over Politics. Editorial. New York Times. January 4.O’Hanlon, M. (2006). Homeland Security Funding: Urban Area Grant Maze. Washington Times. June 9.Office of the White House (2003). National Strategy for the Physical Protection of Critical Infrastructures

and Key Assets. February.

Theor Soc

Perrow, C. (2007). The next catastrophe: Reducing our vulnerabilities to natural, industrial, and terroristdisasters. Princeton: Princeton University Press.

Porter, T. M. (1995). Trust in numbers: The pursuit of objectivity in science and public life. Princeton:Princeton University Press.

Ridge, T. (2003). “Forward,” The Fiscal Year 2003 Urban Areas Security Initiative Grant Program II, i.Website. http://www.dhs.gov/xlibrary/assets/grants_audit_UASIIIFY03GrantAppFinal.pdf. AccessedJuly 20, 2010.

Roberts, P. (2005). Shifting priorities: congressional incentives and the homeland security grantingprocess. Review of Policy Research, 22(4), 437–449.

Rose, N. (1999). Powers of freedom: Reframing political thought. Cambridge: Cambridge University Press.Shenon, P., & Flynn, K. (2004). Mayor Tells Panel ‘Pork Barrel Politics’ is Increasing Risk of Terrorism

for City. New York Times. May 20.State of Missouri Department of Public Safety (2006). Missouri Among First in Nation to Overhaul

Homeland Security Database. Press Release. July 13. http://www.dps.mo.gov/dps/NEWS/Releases06/July/HomelandSecurityDatabase.htm

Stephan, R. (2006). Database is just the First Step. USA Today. July 21.Strathern, M. (Ed.). (2000). Audit cultures: Anthropological studies in accountability, ethics, and the

academy. New York: Routledge.Timmermans, S., & Berg, M. (2003). The gold standard: The challenge of evidence-based medicine and

standardization in health care. Philadelphia: Temple University Press.United States House of Representatives (2006). DHS Terrorism Preparedness Grants: Risk-Based or

Guess-Work? Testimony Transcript. Committee on Homeland Security. June 21.United States Senate (2005). Department of Homeland Security’s Budget Submission for Fiscal Year

2006. Testimony Transcript. Committee on Homeland Security and Government Affairs. U.S.Government Printing Office. March 9.

Wang, H. (2006). Homeland Security: Database Controversy Overblown. Chattanooga Times Free Press.July 15.

Weber, M. (2003). In S. Whimster (Ed.), The essential weber: A reader. New York: Routledge.Wildavsky, A. (1998). Federalism & political culture. New Brunswick: Transaction.9/11 Commission (National Commission on Terrorist Attacks Upon the United States) (2004). The 9/11

Commission Report: Final Report of the National Commission on Terrorist Attacks Upon the UnitedStates. Kean, T. H., Hamilton, L.H., Chair and Vice Chair. Government Printing Office.

9/11 Public Discourse Project (2005). Final Report on 9/11 Commission Recommendations. 9/11 PublicDiscourse Project. Kean, T. H., Hamilton, L.H., Chair and Vice Chair. December 5. http://www.9-11pdp.org/press/2005-12-05_report.pdf

Andrew Lakoff is Associate Professor of Anthropology, Sociology and Communication at the Universityof Southern California. He is the author of Pharmaceutical Reason: Knowledge and Value in GlobalPsychiatry (Cambridge University Press, 2005), and editor of Biosecurity Interventions: Global Healthand Security in Question (with Stephen Collier) (SSRC/ Columbia University Press, 2008) and Disasterand the Politics of Intervention (SSRC/ Columbia University Press, 2010).

Eric Klinenberg is Professor of Sociology at New York University. He is the author of Heat Wave: ASocial Autopsy of Disaster in Chicago (University of Chicago Press, 2002) and Fighting for Air: TheBattle to Control America’s Media (Metropolitan Books, 2007). He is currently writing a book about theextraordinary rise of living alone in cities, and, with Andrew Lakoff, conducting research on the problemof urban security.

Theor Soc