PART I
Building Leviathan
CHAPTER ONE
Social Insecurities
Private Data and Public Culture in Modern America
Sarah E. Igo
How are Americans known by their state, and with what ramifications for individual privacy and political culture? Asked urgently today, the question surfaced as early as the first censuses of the population.1 But it captured broad public attention in the decades defined by the Depression and World War II, provoked by the U.S. governmentâs newâor at least newly openâmethods of tracking its people. Honed during the Philippine-American War at the turn of the century and refined through the Bureau of Investigationâs domestic surveillance activities during World War I, such techniques were in this era extended to a much broader swath of the citizenry.2 From birth certificates to passports, administrative tracking was becoming part of the bureaucratic everyday.3 In the New Deal, it would come wrapped in the guise not of social order but of social benefitâindeed, social security.
What scholars have termed the administrative state entered citizensâ lives in new ways and to novel ends in the 1930s.4 It ballooned further during World War II, when the scale of government activity came to dwarf the New Deal programs âthat had seemed gargantuan only a few years earlier.â5 The state had been a locus for fears about centralized authority since the first days of the American republic, of course. But the state understood as administrator or bureaucrat was a product of the twentieth century.6 As federal agencies loomed larger in Americansâ lives, they also became a focal point for reflecting on individual privacy. How much knowledge about its own citizens ought a government possess? And what would an administered society mean for the people caught in its net?
These questions became less abstract with the passage of the Social Security Act of 1935. A landmark piece of legislation, still considered âthe most expansive and important social welfare program in the United States,â the act ushered in old-age and unemployment benefits for a large segment of the population.7 Less noticed, it also marked the U.S. governmentâs first widespread use of personal information to identify and administer specific individuals, in the form of the Social Security number (SSN). The SSN was an essential mechanism of the ambitious new program, which as reformer and social scientist Sophonisba P. Breckinridge put it in 1935, âcontemplates the participation in all of our lives of the federal, state, and local governments and puts, for the first time, a degree of validity into the expression âAmerican standard of life.ââ8
Standard here referred to a minimum threshold for subsistence, but it implied a kind of standardization common to large-scale administrative projects. Unprecedented though it was in scope, Social Security was in step with a set of identification and documentation practices well advanced by the early decades of the twentieth century.9 Indeed, its planners drew from other nationsâ experiments with administering citizensâ identities, particularly those of France, Britain, and Brazil.10 âSeeing like a state,â in James C. Scottâs influential formulation, hinged on making citizens âlegibleâ and thereby amenable to the designs of officials and planners.11 The expansion of âpaper identitiesâ was thus intertwined with a mode of governance able to register and recognize specific persons.12
States were never the only authors of this documentary impulse. Life insurance outfits and credit agencies were two of the powerful private entities driving the creation of what we would now call âpersonally identifying information.â13 Through the efforts of private as well as public agencies, modern Americans were becoming deeply enmeshed in webs of bureaucratic verification. A columnist for an Atlanta newspaper wryly testified in 1942 that âevery law-abiding citizen todayâ had âhis vest pockets . . . crammed with credentials,â including âa draft registration card, a social security card, a driverâs license, a hospitalization card, an insurance card, a gasoline ration book, a sugar ration book, a finger-print identification card, a shopperâs credit card,â and so on. âPractically all of these items stress the fact that I am me and nobody else; without them, I would officially cease to exist,â he quipped.14
For this columnist, Social Security cards were just one piece of a âthoroughly classified, documented, and cross-indexedâ modern existence.15 Yet these cards warrant special scrutiny for the fashion in which the numbers imprinted on them bound data to entitlements and individuals to the stateâenlisting Americans in their own bureaucratic visibility by making manifest the benefits of identification. The federal governmentâs numbering of individuals, and the potential tracking it permitted, did not escape public notice. Quite apart from discussions over Social Securityâs substantive merits, this feature of its operation engendered sharp questions from a strange set of bedfellows: the Republican opposition as well as African Americans, labor unions, working women, and religious groups. But we must not read backward from our anxious contemporary stance toward identity documents; nor should we assume that state surveillance loomed large for most citizens in the 1930s and 1940s. Concern about Social Security numbers in that era, while evident, competed with another view, in which the nine digits were broadcast, even cherished, as proof of membership in a newly generous polity. This was, we might say, legibility with benefits.
The proud claiming of a Social Security number, a bureaucratic instrument of the expanding welfare state, may today strike us as strange. That dissonance compels us to recognize the ways that the New Deal state has been remembered differentlyâboth in our partisan political culture and in our scholarly accountsâthan it was experienced at the time. The lived history of the SSN reveals that earlier Americansâ relationship with their identity cards, and with the agencies that tracked their affairs, diverged markedly from our own. It also helps us appreciate the fungibility of the very contents of the public and the private. Understanding how, in those decades of depression and war, âprivateâ data could advance a public claim or identity requires an imagination tempered by time.
The Early Days of Tracking
Franklin D. Roosevelt signed the Social Security Act on August 14, 1935, in the midst of the worst economic crisis in American history.16 Intended to provide benefits for the elderly, dependent, and unemployed through a payroll tax, it pledgedâin a wordâsecurity for millions of Americans.17 A vast scholarly literature now examines the impulses behind, the architects of, and the ideological assumptions built into Social Security and other institutions of the New Deal state. Considerably less attention has been paid to the by-products of the new administrative system, and particularly the assigning of unique identifying numbers to citizens. What did this intersection of numbering and state building mean for the Americans newly in Social Securityâs embrace?
Given its exclusion of certain classes of workersâagricultural laborers and domestics, and thus African Americans, most prominently (and deliberately)âSocial Security was not a national system in the sense that it covered all citizens or residents.18 Initially, only those in commercial and industrial employment, roughly 60 percent of the nationâs paid workforce, were encompassed by the program.19 Nevertheless, the legislationâs reach was unparalleled, establishing something akin to a ânational enumeration system.â20 It also differed in kind from most prior state ventures to gather information from Americans. The U.S. Census, although it aggregated reams of personal information, made no decisions pegged to particular individualsâ data. The Social Security system was designed to do just this: track specific workersâ payroll contributions over their entire lifetimes in order to pay out appropriate benefits. Not only, that is, did Social Security need to enlist millions of workers into the program; it also had to keep those individuals in its sights for decades to come.
As a result, SSNs raised in an early form the dilemmas of a society organized around the collection and maintenance of what the agency itself described as âconsiderable personal and confidential information.â21 It was an issue tailor made for partisan combat, and Social Securityâs opponents did not squander the opportunity. Republican operatives seized upon the issue of state-issued identification numbers, whipping up fears of regimentation and improper state invasion into Americansâ private lives.
This was the point of a colorful political stunt engineered by the publisher William Randolph Hearst and the Republican National Committee (RNC) chairman John D. M. Hamilton on the eve of the 1936 presidential election. Not only would workers hand over to the state a stash of sensitive private information, includingâthe RNC (falsely) claimedâoneâs religious and union affiliations, âphysical defects,â and marital status. They would soon also be required to wear âdog tagsâ listing their Social Security number. A central exhibit in the Republican campaign against Social Security was a fabricated photograph of the offending item, pictured on a chain around a young manâs neck.22 âIf the Roosevelt administration is returned to power, we shall see two groups of citizens in this nation,â thundered the RNC chair at a rally in Boston: âthose who are numbered and those who are not numbered.â The former were the unlucky â27 million men and women who will be forced to report to a politically appointed clerk, every change of their residence, every change in their wages, every change of their employment.â For at least some in the crowd of twenty thousand at the Boston Garden who responded with ârepeated waves of applauseâ to Hamiltonâs invocations of police cards and state surveillance, this was the road to European-style despotism.23 As Americans watched developments unfold in Hitlerâs Germany, associating Social Security with other forms of state coercion was a charge with some potency.
The episode fits neatly with conventional wisdom about Americansâ reflexive antistatismâtheir jealous resistance to infringements of their individual libertyânot to mention American historiansâ received view of the New Deal as cementing the modern liberal-conservative divide.24 Indeed, even before the Republican attack, the Social Security Board (SSB) was highly sensitive to the public relations of numbering the population, certain it served a people who had âalways been fearful of anything that might suggest the loss of some personal freedom through formal records of identities.â25 Thus the board scrupulously avoided the term registration to describe the enrollment effort, instead favoring enumerationâan attempt to assimilate the new practice with the long-standing one of census taking. It also insisted, somewhat disingenuously, on the âentirely voluntaryâ nature of applying for an account number.26 Moreover, that number, it was stressed, was for the holderâs convenience and not for identification. In response to the RNC âforgery,â planners stated emphatically that Social Security did not âintend nor had it ever intended to issue identification disks to American workers.â27 Finally, the SSB described information about marital status and union ties as âmatters private in their nature and of no legitimate concern to the Federal Government,â adding that âno such questions would be asked now or at any time in the future.â28
Evidently, the board believed that it was tightly constrained by the public culture in which Social Security was taking root and needed to tread carefully. Each decision it made regarding the rollout of the SSN was carefully weighed not just for its administrative implications but also for its political ones. Internal debates over how best to track Social Securityâs beneficiaries, vigilant attention to questions of public reception, and strenuous avoidance of fingerprinting or anything that resembled âregistrationâ all point to a bureaucracy focused on exerting the lightest touch possible.29 But was the Social Security Boardâor the RNC, for that matterâcorrect in its estimation of the American public? Evidence suggests that citizen...