Book Review: Disasters and the American State by Patrick S. Roberts

In Disasters and the American State, Patrick S. Roberts has written a thoroughly persuasive account of the long and uneven development of what he calls the American “disaster state.” Roberts’ fundamental goal is to help us understand contemporary disaster politics, including how past politics and institutions have given rise to these politics. He draws on insights from the literature on American political development to provide us with this account, which emphasizes the role of historical patterns as well as idiosyncrasies in creating these politics, and their roles in shaping the American state. The book is very well written, provocative, and well researched. Anyone interested in American political development should find it compelling, and of course, disaster scholars may be especially interested.

Roberts argues that the federal government has been involved in disaster politics, albeit in rather ad hoc fashion, for most of the nation’s history. In the early republic, Congress was the dominant actor in disaster politics, and relief typically came in the form of tax relief or land grants rather than appropriations. Roberts offers an example in a major fire in Portsmouth, MA in 1806. In response to that fire, Congress suspended the collection of bonds in that town in order to relieve its financial burden. After the Civil War, Congress continued to predominate, but it began to make special appropriations for the relief of populations affected by disasters. Additionally, the pace of Congressional enactment of relief measured doubled compared to the antebellum period, and Reconstruction also saw the first bureaucratization of disaster relief as the Bureau of Refugees, Freedmen, and Abandoned Lands distributed aid in peacetime for disaster relief, among other things.

The Congress-centric story of federal disaster politics begins to weaken as the White House took on a stronger role in the field beginning, Roberts argues, with Calvin Coolidge’s response to the great Mississippi River flood of 1927. By this time, Americans expected more leadership from the White House in domestic affairs generally than they had in earlier periods (e.g. Lowi 1986), and Roberts shows that the realm of disaster response was no exception. Coolidge took unprecedented steps to relieve the Midwest: he established a special cabinet committee to coordinate the national response, and appointed Herbert Hoover (the Commerce Secretary) to direct the massive relief effort of the American Red Cross.

Roberts argues that the ad hoc nature of federal disaster response persisted until the mid-twentieth century. In the post-war period, however, the federal government began work to coordinate and systematize its disaster responses by passing the Disaster Relief Act of 1950. This Act provided that the president had broad, discretionary authority to declare what constituted a disaster eligible for federal aid, defined who in the federal government was to distribute the funds, outlined procedures for how those funds were to be allocated and distributed, and established that relief funds would only be sent to state and local governments, not to individuals. In this way, the Act contributed both to the shift toward presidential power and leadership, and to the layered, federalism-infused nature of disaster politics.

This era also marked another enduring shift in federal disaster politics: existential fear driven by Cold War led to the fusion of “disaster” and “emergency” and thus left disaster response tied up with civil defense, both socially and institutionally. Roberts shows that this amalgamation has largely persisted even as the threat of nuclear war has waned, and that the lingering “Cold War mindset” has been an important reason that disaster relief efforts today are frequently unsatisfactory to the public at large. That is, many agencies – especially FEMA – had dual missions: security and disaster response. Importantly, Roberts argues that the “security missions” of these agencies had major impacts on their ability to respond effectively to disasters. The creation of FEMA epitomized these competing strands: the organization was only loosely connected; a series of stovepipes separating various divisions and processes, and three distinct cultures divided the agency (civil defense, disaster relief, firefighting). The complexities led to the adoption of the “all hazards” approach, wherein “government will use the same plans, procedures, resources, and personnel to address all kinds of hazards and disasters” (p. 80). At different times, one of these missions necessarily took precedence over the other; the chief concern at any given time affected staffing decisions, attention from Congress, and public image. Thus the dual purposes of these organizations necessarily reduced their abilities to address either mission fully.

Roberts shows that the federal role in disaster relief has expanded greatly since the 1960s – but also that public satisfaction with federal responses is consistently quite low during this period. This is partly due to the fact that agencies were pulled in different directions by their dual missions, and did not have sufficient resources to address both missions. Another factor in the low satisfaction with federal disaster response is due to a general awareness that the government has an enlarged capacity to relieve federal disasters – that is, the fact of the growing “disaster state” itself inflates expectations of efficiency and efficacy in society.

This “dual mission” problem was temporarily abated in FEMA in the 1990s under the tutelage of James Witt, the FEMA chief appointed by Bill Clinton. Witt, Roberts argues, was able to clarify FEMA’s mission, improve customer service, and perhaps most importantly, consciously aligned FEMA’s gals with the reelection goals of Congressmen and the president (see also Roberts 2006). By refining and clarifying FEMA’s mission, and allocating its resources in accordance with that mission, FEMA became remarkably more effective and efficient – indeed, it went from laughing-stock to a genuine model bureaucracy.

Roberts argues that after 9/11, the quick absorption of FEMA into the new Department of Homeland Security, and the concomitant addition of terrorism preparedness to its core mission, is a prime example of how “government’s capability to deal with disaster could not keep up with the public’s rising expectations” (p. 126). Under Witt, FEMA had become a model agency largely because it was successful in narrowly defining its operational mission (that is, doing relatively few things, but doing them well). Public demand for protection from an amorphous terrorist threat, and perceptions of the agency’s power and ability to respond to such threat, outstripped its actual capacity for action.

Roberts argues that a reputation for competence and efficiency can be a source of bureaucratic autonomy and independence (however unstable). That is, FEMA’s reputation as the effective “all hazards” disaster agency allowed it to act independently. This uncertain reputational autonomy also left the agency exposed to destructive forms of politicization when it failed to live up to that reputation, such as after Hurricane Katrina. Here Roberts makes an important contribution to our understanding of bureaucratic autonomy. One of the (theoretical) virtues of administrative governance is political independence, that is, the making and implementation of policy on the bases of sound economic and scientific determinations made by experts, rather than on the bases of partisan or parochial politics.

Roberts closes the books with several important observations about the both the realities and potentialities of disaster mitigation in the U.S. First, that politicians’ claims and the public’s expectations about the government’s ability to prevent or ameliorate disasters greatly outstrip its actual capacity to do so. Second, that the disaster state (like the administrative state more generally) is extremely complex, being shaped by presidents, legislators, bureaucrats, federalism, the emergency management profession, and public expectations, all of which respond to their own separate incentive structures – and all of which change over time. Finally, he argues that the incentives of the chief actors tend to emphasize short time horizons (often no further than the next election), whereas many truly meaningful disaster mitigation strategies, such as building and zoning codes, take decades to implement and do not lend themselves to credit claiming by politicians, and thus are often cast aside in favor of more politically expedient measures. These observations should give pause to scholars, practitioners, and citizens concerned with disaster politics and policy in America’s future.

My main qualm with the book concerns the role of the public in the processes of social construction in Roberts’ narrative.

Roberts states that central puzzle of his book is “what is the role of the federal government in addressing disasters, and how has it changed? The answer is that citizens, members of Congress, disaster managers, presidents, and the media inadvertently shape what counts as a disaster and how much responsibility the federal government has in addressing it. This process of social construction occurs while various actors pursue their own interests, whether winning reelections, making promises to voters, managing organizations, reporting the news, or preparing for disasters.” (p. 176).

To this end, Roberts offers a great deal of evidence of these sorts of construction occurring in Congress (in the 19th century), and in the upper echelons of bureaucracies (in the 20th century). He also makes a strong case that media plays an important role in the construction of “disaster”, especially since the 1920s or so. For most of this developmental story, though, the role of average citizens – of the public at large – is unclear. Roberts acknowledges that most works in APD try to use public opinion polling and elections to measure public influence on development – yet also notes that “Public expectations are filtered through the news media” and that elections in which disasters are a major issue are “rare” (p. 189). These admissions suggest, to me, that this process is almost totally elite-driven. That is, this narrative suggests to me that the primary locus of the social construction of disaster was driven primarily by Congress and local, parochial elites in the 19th century, and in the 20th century, shifted from Congress to the presidency, and from local elites to the emergent national media. Today, even the absurdly high expectations of government’s capacity to respond to and mitigate disasters, held by much of the general public, has likely been driven in large part by media. Roberts’ extended discussion of “elite panic” (elite overreaction to the fear of crisis, social breakdown, and challenges to their authority, such as we saw in the days and weeks following Hurricane Katrina) underscores this interpretation.  None of this is to say that this is not social construction, rather only that even the modest role for the public that Roberts suggests is, I think, overstated.

I was also surprised to see very little discussion of the Army Corps of Engineers. Roberts’ account of the modern disaster state centers on FEMA, and for good reason. Yet the Army Corps is an important actor in the disaster state (and has been so for quite a long time), playing key roles in both mitigating and responding to disasters. Further, the Corps is, arguably, a very unique bureaucracy – one that may be able to exercise the sort of reputational autonomy Roberts observes in FEMA during the Witt era on a much more regular basis than other agencies (Adler 2012; Shallat 1989; see my previous discussion of this possibility here). Thus, I would like to have seen more thorough treatment of this important disaster agency, including its ability to act autonomously, and its role in constructing the meaning of disaster (and the meaning of “response” and “mitigation”) over time.

Finally, another point about which I’d like to see more work: as a scholar interested in American Constitutional development, I would have like to have seen more about the debates in Congress regarding the constitutionality of proposed relief measures. To be sure, this is largely beyond the scope of Roberts’ study, which focuses on the growth of the “disaster state.” His discussion of the construction of this state, especially before 1927, raises some very interesting questions about constitutional construct in a system of separated powers (or shared powers and separated institutions). For example, Roberts argues that in the first several decades of America’s existence, debates in Congress as to whether or not to provide relief for some particular misfortune (earthquakes, fires, etc.) often centered on questions of constitutional power – specifically, whether the Constitution actually empowered Congress to provide any such relief. Eventually, we see, precedent accumulated and debates over constitutionality presumably fade to the background and then disappear completely by the late 19th century. I would like to see much more about this process – the ins-and-outs of the arguments in Congress, how they changed over time, why they ultimately disappeared, etc. For now I’ll have to hope that some enterprising scholar will pick up this important question that Roberts suggested here.

All in all, Patrick Roberts’ Disaster and the American State is excellent – imminently readable, well sourced, and compelling; it is must-reading for individuals interested in disaster politics and policy, state capacity, or American political development.


Full citation to Roberts’ book:

Roberts, Patrick S. 2013. Disasters and the American State: How Politicians, Bureaucrats, and the Public Prepare for the UnexpectedCambridge University Press.

Other sources cited:

Adler, William D. 2012. “State Capacity and Bureaucratic Autonomy in the Early United States: The Case of the Army Corps of Topographical Engineers.” Studies in American Political Development 26: 107-124.

Lowi, Theodore J. 1986. The Personal President: Power Invested, Promise Unfulfilled. Cornell University Press.

Roberts, Patrick S. 2006. “FEMA and the Prospects for Reputation-Based Autonomy.” Studies in American Political Development 20(Spring): 57-87.

Shallat, Todd. 1989. “Engineering Policy: The U.S. Army Corps of Engineers and the Historical Foundations of Power.” The Public Historian 11(3): 6-27.

Book Review: The Sympathetic State by Michele Landis Dauber 1

Over the summer I read the wonderful book, The Sympathetic State: Disaster Relief and the Origins of the American Welfare State. The author, Michele Landis Dauber is a professor of law as well as the Bernard D. Bergreen Faculty Scholar at Stanford University. While I have never met Professor Dauber, she and I have been traveling in overlapping scholarly circles through the Law and Society Association for several years. When I moved from property and takings into studies of disasters and law, I quickly stumbled upon her work. When the University of Chicago Press released her book last year I bought it and put it in my pile of work-related reading. I finally got to it — and wish I had gotten to it earlier.

Dauber’s argument is that disaster relief and the “disaster narrative” is central to the development of the American welfare state. What she means by “disaster narrative” is especially interesting to me as these narratives (i.e., stories we tell about disasters and their victims) have major impacts upon public perceptions that feed into public policy. For Dauber, the disaster narrative she is primarily interested in is one developed during the New Deal that “fashion[ed] a story against the backdrop of an existing moral economy that would constitute a compelling case for aid” (p. 11). This “compelling case for aid” may have started in narratives around natural disasters, but Dauber argues that during the Depression it was stretched to include other social problems. Unemployment, according to Dauber, became a “disaster” that had to be addressed through legislation. The chronic unemployment that characterized The Great Depression is often attributed to drought, but quite likely is more closely linked to the mechanization of farming. Whatever its underlying causes, it was framed as a “disaster” by the Roosevelt administration in order to provide a moral basis for providing aid through Congressional legislation for those migrant workers, farm laborers and sharecroppers who found themselves deprived of the jobs they had customarily held. Dauber examines the development of this disaster narrative in US welfare policy beginning at the founding, and moving through the New Deal era. Her argument is wells documented and very persuasive. Indeed, Dauber’s account of how the attempt to stretch “disaster relief” to include unemployment at the national level is one of the great achievements of this book.

In addition, making extensive use of archival work, Dauber provides some interesting content analysis to support her view that not only did a disaster narrative develop in national political institutions, but that non-political actors such as citizens who had suffered unemployment and other economic hardships, deployed the disaster narrative in their attempts to receive assistance. She uncovers the characteristics and structure of this narrative through extensive research not only into governmental documents, but also materials such as the letters requesting aid from President and Mrs. Roosevelt.

Along these lines, one of Dauber’s most interesting research strategies includes a careful content analysis of letters to Eleanor Roosevelt. Hundreds of thousands of letters were written to Mrs. Roosevelt during the Great Depression. This large body of material, combined with other official documents such as the census, provides Dauber with a rich dataset for exploring the way those seeking assistance imagined what requests would be granted. Both they way these letter writers conceived the government as well as their own situation are on display in these letters. Dauber explains the writers of these letters were ordinary people seeking assistance, who, she writes are “[u]nlike the judges, politicians, lawyers, artists, writers, and PR men [discussed earlier in the book] the writers…had only those persuasive resources they could mobilize over the kitchen table,” (p. 187). Dauber closely analyzes a random sample of 529 of these letters, then did additional analysis on a subset of 267, for whom she was able to find additional information. Fascinatingly, Dauber uncovers what she calls “the underlying moral order” that informed the writing of these letters. She uses them, as she explains, as “accounts” – which has a fairly specific meaning in social science: “as efforts by social actors to explain otherwise discrediting or inexplicable situations as a way that will make sense to their audience and purge their behavior of troublesome aspects” (p. 191).

As Dauber further explains, such accounts provide indicators of “shared understandings about what constitutes a good explanation of behavior, and indeed of what requires an explanation in the first place” (p. 191). In these letters to Mrs. Roosevelt, then, we learn that many, many of the letter writers felt the need to explain not only their current situation but why they could not support themselves. They did, indeed, provide excuses for seeking help. For the social scientists among my readers, Dauber explains that historians have treated these letters differently, seeing them as evidence of the hardship and need of the writers, or as autobiography of sorts (p. 192-193, citing Robert McElvine and Robert Cohen). Dauber uses the letters with a different goal in mind. Drawing on the insights of C. Wright Mills, she sees “a vocabulary of motives.” So many letters employ such similar strategies that, Dauber explains, the letters contain strong evidence of “a widespread expectation that these were the excuses and justifications that would be ‘honored’,” (p. 192). In other words, these were the reasons for the situations that were understandable and would elicit sympathy from the audience for the letters.

What makes her analysis particularly interesting is that Dauber pays attention to both the details in the letters and the details of the letters. Among other things, we learn they are very uniform in length (running to about 400 words). They often contain the same basic information (current employment status, and a very brief description of the difficulties faced by the family). By cross-referencing the letter writers’ names with the 1930 census data, Dauber could determine whether their home was classified as urban or rural, which supplemented the basic information already in the letters such as the occupation, gender and basic family structure of the writer. Dauber went a step further and located the manuscript census form for the letter writers when she could find them (she was successful over half the time). For these writers, Dauber was able to add great detail yet, such as the value of their homes or monthly rental costs, whether there was more than one generation of the family living in the home, and even whether a family owned a radio (p. 194). Such information allowed her to determine how unique these writers were, while also making clear the limitations her data contained.

These sorts of limitations always exist in any dataset, though it is not always the case that we can see them as well as Dauber can present them because of the multiple data sources she has. For example, Dauber reports that she was more likely to find information on rural letter writers than urban, and certainly more likely to find those who had not changed their name since the 1930 census. She points out that while there may be bias in her subsample of 267 letters, she is able to take these limitations into account. The one limitation I wish she had discussed more seems to be the middle class status of most of these letter writers. In terms of bias in their perception, the fact that they are not the poorest members of the population leads me to wonder whether there would be a class difference in the way requests for assistance would be made; that is, would the moral economy of social welfare provision look different if they were addressing not their own needs, but more destitute victims of unemployment and economic hardship? That said Dauber does, indeed, do a very good job in both noting various issues and carefully not overstating her findings. I would not be able to raise the question about the class status of these letter writers had she not so carefully parsed both the language of the letters and the writers. Her findings do provide us with a remarkable sense of what the shared understandings of these letter writers were, and why they believed they deserved help from the government. After carefully examining them, Dauber concludes her analysis of the letters:

These letters are replete with evidence of a shared sense of entitlement, if a peculiarly American strain. In order to qualify for help in this system, the test is neither loyalty nor need but losing out through no fault of one’s own… [M]ost writers had an employed bread-winner. They asked not for food or shelter but often for quite frivolous things; nice clothing, vacations, summer camps, help paying off chattel mortgages so that they would not their rugs and furniture to repossession. They wanted to get out of debt, start businesses, and send their children to college. While many asked for loans, most expected to receive a gift and expected to be able to spend it as they saw fit, free from the scrutiny of gossipy neighbors and the interfering do-gooders down at the local relief board.

From the perspective these writers, disaster had struck, the government’s obligation to bring relief was activated and the citizen correctly sought to enforce the social contract.  (p. 222)

Ultimately, Dauber says that these letters demonstrate a “broadly held understanding of the American moral economy of social welfare provision” (p. 223). This is quite likely true. One of the things, however, I wanted to understand better was whether these, often middle class letter writers, maintained the same beliefs about others who had been much less fortunate. Did they see families with no breadwinner that were suffering from the weather-related disasters (including the drought) or economic hardships such as a bottomed-out market in soybeans or corn – did these same writers extend their belief about this moral economy of social welfare provision to the much less fortunate? Based upon earlier chapters in her book, the link between the unemployed and a “disaster” was pecularilarly difficult to produce inside of governmental institutions and at the national level. Yet here she shows that many middle class individuals embraced it.

My questions, not answered by this book, are these: if Dauber is correct, how did middle class individuals, suffering and requesting aid by writing to the President and his wife, perceive the needs of others? How did their perceptions concerning social welfare provision fit into the moral economy of social welfare provision for the much more poor? How did their beliefs about aid structure their voting behavior? And in turn, did this influence the way legislators structured our welfare policy? This strikes me as the next step in this work, to link the letters back into the policymaking process. Broadly, it is an issue that scholars interested in poverty and social welfare in the United States have taken up, but with Dauber’s work we have new insights into the way disaster relief played a role in the development of the New Deal social welfare programs and insight into the understandings of social welfare prevalent among some of the members of the society at the time. By reconnecting the public’s perceptions in these letters to the policymaking process, we may be able to understand even more fully the contemporary politics of welfare as well as the continuing power of disaster relief narratives.

Full cite to Dauber’s Book:

Dauber, Michele Landis. 2013. The Sympathetic State: Disaster Relief and the Origins of the American Welfare State. The University of Chicago Press (Chicago, IL).


Cohen, Robert, ed. Dear Mrs. Roosevelt: Letters from Children of the Great Depression. Chapel Hill: University of North Carolina Press, 2002.

McElvaine, Robert S. Down and Out in the Great Depression: Letters from the “Forgotten Man.” Chapel Hill: University of North Carolina Press, 1983.

Mills, C. Wright. “Situated Action and Vocabularies of Motive.” American Sociological Review 5, no. 6 (1940): 904-13.

Scott, Marvin B., and Stanford M. Lyman. “Accounts.” American Sociological Review, 33, no. 1(1968): 46-62.