Stolen Lives: Redress for Slavery’s and Jim Crow’s Ongoing Theft of Lifespan

Quantifying State-Level Variation: Data

The primary lifespan data are measures of state-specific period life expectancy for non-Hispanic Black and White Americans, calculated as part of the County Health Rankings and Roadmaps (CHRR) dataset (University of Wisconsin Population Health Institute, n.d.), based on restricted-access lifespan data from the National Vital Statistics System.⁹ Because the long-term consequences for mortality of the COVID-19 pandemic are still unknown, this article uses pre-pandemic lifespans by using the 2021 CHRR release, which draws data from 2017 to 2019. Life expectancy measures in this dataset therefore represent the average lifespan of a hypothetical birth cohort whose mortality conditions, over their entire lives, were those of the same-race 2017–2019 population in their state. I treat Washington, D.C., as a state.

I use state-level lifespans, rather than more disaggregated geographies such as counties, for two reasons: first, because it is not clear that counties have any analog of state governments’ “critical role in defining racial categories and policing their boundaries” (Bruch, Rosenthal, and Soss 2019, 164); second, because some county-level estimates from the CHRR seem clearly implausible (for example, life expectancy above one hundred for both White and Black populations) and likely reflect life expectancy’s sensitivity to small-area variation. However, in an additional analysis reported in the appendix, I use county-level lifespans from a different source, the Institute for Health Metrics and Evaluation (Dwyer-Lindgren et al. 2022), which raise some different methodological challenges.

I analyze two outcomes that each imply different benchmarks for conceptualizing the lifespan loss associated with states’ participation in slavery and Jim Crow. First, I analyze state-level variation in Black life expectancy. This outcome implicitly uses Black life expectancy in other states as a benchmark (or baseline). Second, I analyze state-level variation in the White-Black life expectancy gap. This outcome uses White people in the same state as a benchmark.

As independent variables, I use four distinct measurements of the intensity of slavery and Jim Crow at the state level. First, I examine the proportion of the state population that was enslaved in 1860. This measure was previously found to be associated, at the county level, with lower Black, and higher White, life expectancy, net of various control variables (Reece 2022). Second, I examine counts of Jim Crow laws passed by each state before 1950. These laws, some seven hundred in total, were originally collected by Pauli Murray ([1950] 2016) and were digitized and extended by Althoff and Reichardt (n.d.). Third, I examine the quality of segregated Black schools. The school quality measure was also constructed by Althoff and Reichardt, who find large long-term economic consequences of both Jim Crow laws and Jim Crow school quality, using data on teacher salaries, student-to-teacher ratios, and term lengths in Black segregated schools during the Jim Crow era, all previously collected by David Card and Alan Krueger (1992). I reverse-code this measure to indicate inadequate quality, so that, like the other measures, it reflects the intensity of racial oppression.

Fourth, I examine the historic racial regimes index (HRR) developed by Regina Baker (2022), a composite index of measures of slavery, sharecropping, disfranchisement, and segregation—measures collectively spanning the antebellum period, Reconstruction, and Jim Crow. Baker uses this scale to assess the legacies of racial regimes for poverty across southern states, finding that a stronger historic racial regime is associated with worse Black poverty and, especially, Black-White economic inequalities.¹⁰

Finally, in the appendix, I additionally explore two additional measures designed to capture the intensity of racial terror at the state and county levels, respectively. These are, first, counts of Jim Crow violent incidents, collected by the Burnham-Nobles Digital Archive (Civil Rights & Restorative Justice Project, n.d.); and second, counts of lynchings of Black Americans between 1883 and 1941, collected by Charles Seguin and David Rigby (2019).¹¹ Both measures were collected from historical newspaper accounts, a source that raises distinctive measurement concerns, as described in the appendix, though these concerns are likely to be substantially greater for the broader measure of violence than for lynchings. Extrajudicial violence also bears a complex relationship with racial domination, in that it may occur most readily when other, more formalized and routinized means of domination are not in operation or not reliable (Troesken and Walsh 2019).¹² Yet, despite these weaknesses and calls to more comprehensively measure racial violence, much evidence based on these and other data sources show long-term consequences of racial terror (Cunningham, Lee, and Ward 2021).

Each of these measures is defined for a distinct set of states, but their bivariate correlations, in the states where each respective pair exists, vary widely. Unsurprisingly given that it is constructed from other measures or analogs of them, the HRR is tightly correlated with most of the other measures: its correlation with (reverse-coded) Black school quality is 0.94, with the proportion enslaved in 1860 (which is one of its four components) is 0.89, and with the count of Jim Crow laws (a much simplified version of which is one of the HRR’s components) is 0.71.

Quantifying State-Level Variation: Results

Results of these analyses are summarized in table 1, which presents, for each measure, the difference between each outcome at the 75th and the 25th percentile values of that measure (with singular outliers omitted, as noted). Detailed, graphical results and summaries are presented in the appendix. State-level measures that include Washington, D.C.—the proportion enslaved in 1860, the count of Jim Crow laws, and the quality of Jim Crow schools—are reported with and without that observation because it is an extreme outlier of White life expectancy among Jim Crow states, with the highest state-level White life expectancy in the entire United States. Washington, D.C., has the fourth-lowest Black life expectancy in the United States, which does not make it an outlier among Jim Crow states. Additionally, the count of Jim Crow laws is also reported without Louisiana, which passed ninety-eight such laws, nearly 50 percent more than the next-highest state, Virginia, which passed sixty-six, an intensity of legal disfranchisement that was extraordinarily effective and damaging (Keele, Cubbison, and White 2021). These two outliers substantially affect the bivariate relationships summarized in table 1 when they are included.

Across measures, the relationship of the intensity of slavery or Jim Crow—when comparing states that had slavery in 1860 or Jim Crow laws—to Black life expectancy and to the White-Black life expectancy gap is somewhat inconsistent. For Black life expectancy, the proportion enslaved in 1860, poor-quality Jim Crow schools, and the HRR index all have a negative relationship, and the count of Jim Crow laws has a positive relationship. For the White-Black life expectancy gap, the proportion enslaved in 1860 and the count of Jim Crow laws have an unexpected negative relationship (as do both measures of violence analyzed in the appendix); the extent of poor-quality Jim Crow schools has almost no relationship; and only the HRR index has the expected positive relationship, which is driven primarily by that index’s positive association with White life expectancy. In general, the geographic variation in the White-Black life expectancy gap analyzed here is more heavily driven by variation in White life expectancy than by variation in Black life expectancy.

The inconsistency of these results is perhaps surprising in light of research finding county-level negative associations (net of controls) between Black life expectancy and the proportion of the population enslaved in 1860 (Reece 2022).¹³ More broadly, the research on the “long arm of slavery” finds enduring spatial and place-based differences, based on county-level and state-level histories of enslavement, in Black levels of, and Black-White disparities in, outcomes such as poverty (O’Connell 2012; Baker 2022), educational attainment (Bertocchi and Dimico 2012), arrest rates (Ward 2022), and harsh prison sentences (Gottlieb and Flynn 2021). County-level lynching histories predict contemporary use of corporal punishment in public schools, particularly for Black students (Ward 2022). States’ and counties’ enslavement histories predict their consequential cultural and institutional features like the political attitudes of their White residents (Acharya, Blackwell, and Sen 2016), their contemporary political suppression of Black votes and the paucity of their welfare state provisions (Williams, Logan, and Hardy 2021), and the late-twentieth-century growth of their incarceration apparatuses (Duxbury 2023), though this latter association partly reflects a complex history in which other forms of labor control predominated in the South in the early twentieth century (Muller 2021). This large research program pinpointing slavery’s long-term space- and place-based legacies (reviewed in Cunningham, Lee, and Ward 2021) is not without nuance and qualification, such as explorations of how demographic change enables and constrains the persistence of these relationships (O’Connell, Curtis, and DeWaard 2020) and analyses of mid-century racial terror as an intermediary mechanism between slavery and contemporary homicide that varied in intensity (Petersen and Ward 2015). Yet despite its growing complexity, this research program’s overall message is clear: slavery reverberates. Its echoes are so vast and all-encompassing that they are best approached on the terrain of ever-shifting metaphor: slavery has afterlives (Hartman 2008, 6); it leaves a wake (Sharpe 2016). From this perspective, the somewhat mixed results found here are unexpected.

However, other research seems to be broadly more consistent with what is found here. One study finds that the direct economic consequences of enslavement had dissipated by 1940, whereas the consequences of Jim Crow—and thus also the indirect consequences of enslavement, by putting formerly enslaved families in what would become Jim Crow states—continue to this day (Althoff and Reichardt, n.d.). Another study suggests that the consequences of eradicating Jim Crow legal regimes for the survival of Black babies—a particularly consequential determinant of life expectancy—were immediate (Krieger et al. 2013).

To the extent that the results found here vary across measures, they raise a question about the extent to which parts of the legacy of slavery research program, in which studies typically rely on a single measure, may have a “file-drawer problem” (Rosenthal 1979). From the perspective of reparations programs, one implication of the results found here is that efforts to use such cross-state comparisons to calibrate metrics of lost lifespan probably need to commit to a measure on theoretical grounds, which may be difficult to justify.

Interpreting Geographic Variation in Lifespans

The premise of relating Black lifespans to state-level Jim Crow, slavery, and racial terror is that the variation across states in the intensity of these regimes captures the way that those regimes structure contemporary lifespans. That premise may not be true. Indeed, both baselines that are implicated in these analyses—Black lifespans in other states and White lifespans in the same state—are problematic from the perspective of identifying how racial regimes and histories of enslavement and Jim Crow structure Black lifespans today. The problems with the northern baseline are discussed in the remainder of this section; the problems with the White baseline are elaborated elsewhere (Wrigley-Field, n.d.) but amount to the fact that White mortality is hardly unaffected by racism and the legacies of slavery (Metzl 2019; Malat, Mayorga-Gallo, and Williams 2018; McGhee 2022).¹⁴

The act of comparing Black lifespans across states based on states’ intensity of slavery and Jim Crow implicitly imagines that other states were fully outside that system. This is not true. To deny it is not to deny the uniqueness of the Jim Crow regime or to imply that living under that regime was no different from living in the North; Jim Crow was unique in its harms. Yet there is something odd about the premise that, on the one hand, slavery in the South still affects southern states today, but on the other hand, the fact that the entire country was founded on slavery no longer deeply affects the North. Indeed, Black Americans’ advances in freedom—whether emancipation (Downs 2012) or northward migration (Black et al. 2015)—often came at the cost of shortened lives, in the respective contexts of abandoned Reconstruction and the brutal segregation of northern cities (Leibbrand et al. 2020), among other racist policy responses in northern migration destinations (Derenoncourt 2022)—despite substantial Black organizing for health (Long 2016; Nelson 2011). It seems perverse, then, to use the very places where lifespans were shorter, where Black migrants bought freedom (from Jim Crow) with freedom (their lives), as a neutral measure of lifespan free of Jim Crow’s echoes today.

This problem highlights a distinctive challenge of conceptualizing redress for lost lifespan: the lack of a natural metric, which creates the need for a baseline against which to compare even if no good baseline exists. Attempts at enumerating the scale of slavery’s theft of wealth are plagued with challenges, but they are conceptually clearer. One approach that some authors have taken is to focus on a single pivotal moment and ask, how much wealth was stolen just through this one injustice alone? (Logan and Darity 2020) This strategy does not presume that no other atrocities matter or deserve redress; it simply makes the problem of beginning an accounting more tractable. It would be useful if comparing Black lifespans in northern and southern states provided a similarly revealing, if imperfect, starting point. But the studies of wealth stolen in (in that example) one race riot in Arkansas do not assume that the wealth distribution is otherwise unproblematic, whereas using Black lifespans in northern states—or in southern states that enslaved fewer people or passed fewer Jim Crow laws—as a baseline builds in the assumption that those lifespans were not limited by slavery and Jim Crow.

The analysis here explored a variety of measures of the intensity of slavery and Jim Crow to see whether they might offer clear targets for quantifying the degree of lost lifespan that should be redressed. In the end, the variable results across measures and the fact that these racial regimes, in addition to being particular to the South, were also integral to the United States as a whole, suggest that—though perhaps all American lifespans are shaped by slavery in some way—the most meaningful target may simply be the overall White-Black lifespan gap in the full United States.¹⁵ Before the COVID-19 pandemic, that gap was four years of life expectancy; during 2020 and 2021, it was nearly six years (Arias et al. 2022). In the remainder of this article, I consider what it could look like to take, as a target of redress, giving Black and White Americans equal access to time.

Can Wealth-Based Reparations Amount to Health-Based Reparations?

In particular, can the mechanism be wealth-based reparations? We know two things clearly: one major pathway that produces ill health and shortened lifespans for Black people in the United States runs through inadequate income and wealth; and income and wealth are not the only source of the health and lifespan inequities that Black populations face.

That much lifespan inequity runs through wealth is clear; research has found that wealth, income, and other economic resources account for substantial portions of racial inequity in survival (Sudano and Baker 2006) and health (Hayward et al. 2000), though others find that such economic variables account for little racial disparity in certain important health outcomes, such as cardiovascular conditions (Teitler et al. 2021). Three studies stand out for results that suggest that wealth differences may drive nearly all of White-Black mortality disparities. The first (Geruso 2012) finds that the vast majority of the White-Black life expectancy gap, excluding infant mortality, can be accounted for statistically by economic and demographic variables, although the underlying samples are small relative to the task of estimating mortality in small age groups. The second (Himmelstein et al. 2022) finds that wealth differences can statistically account for mortality disparities above age fifty. The third (Do, Frank, and Finch 2012) finds that self-rated health can be fully accounted for by socioeconomic status when the latter is comprehensively accounted for. Other research, however, suggests that study designs that do not account for the race-specific consequences of socioeconomic position (and stressors) can overestimate those conditions’ contributions to racial health gaps, and that such contributions are substantial but far from total (Brown et al. 2023).

The findings of that third study, among others, may also suggest that wealth-based reparations might improve health in ways that research employing standard economic measures fails to capture. Limitations of conventional measures are (among others) that they measure the economic resources of individuals but not their community contexts (Do, Frank, and Finch 2012) and that they typically reflect a single moment in time (Boen 2016). In contrast, class evolves over the life course (Phillips, Martin, and Belmi 2020) and Black and White individuals with the same point-in-time economic status have likely held it for different lengths of time (Do, Frank, and Finch 2012; Boen 2016) and different numbers of generations (Sharkey 2008). An adequate wealth-based reparations program would provide flexible protection at all stages of life: a cushion against poverty, a way of limiting deleterious exposures, and the possibility of making investments across all of life, all benefiting recipients’ descendants as well as themselves.¹⁶

Finally, some research finds that racial disparities in death during key midlife ages—the ages when most of the population-level lost lifespan is concentrated (Bor et al. 2022)—occur exclusively among those with relatively low levels of flourishing (Louie et al. 2021), a psychosocial construct capturing holistic well-being (VanderWeele, McNeely, and Koh 2019; Levin 2021), and it is plausible that large-scale wealth redistribution would sharply curtail the prevalence of low flourishing, such as by limiting poverty (Desmond 2023; Linares, Kandasamy, and Vladutiu 2022) and enabling education (VanderWeele 2017). Thus, across many distinct lines of research, the idea that wealth engenders health seems nearly undeniable, and research offers substantial direct and indirect evidence that a radical redistribution of wealth toward Black Americans would provide similarly radical improvements in Black health.

Yet much research also finds substantial health and longevity gaps between Black and White Americans net of their economic position; indeed, some studies find that these gaps are largest among the wealthy (Colen 2011) and the most highly educated (Farmer and Ferraro 2005; Bell et al. 2020). Such findings have historical echoes: a study based on detailed collection of death certificates in the Carolinas across the twentieth century (Logan and Parman 2014) finds that, in the early twentieth century, Black workers in higher-status occupations had higher mortality than those in lower-status occupations—the inverse of the White pattern. Wealth does not protect against all forms of discrimination; indeed, it can intensify risk of certain kinds of discrimination by bringing Black people into settings in which few other people are Black (Colen et al. 2018; DeAngelis 2022) and where racial boundaries might be most actively defended (Christensen and Timmins 2023). These encounters are not benign; the evidence that experiences of discrimination can harm health through stress pathways is pervasive and convincing (Williams and Mohammed 2009; Lewis, Cogburn, and Williams 2015; Goosby, Cheadle, and Mitchell 2018), particularly for mental health outcomes (Paradies 2006), whose impacts on lifespan are indirect but likely meaningful, and whose impacts on the quality of life are clear.

Wealth-based reparations would not directly address one particular kind of discrimination with obvious import for health: discrimination by medical providers (Smedley, Stith, and Nelson 2003; Spencer and Grace 2016; Nong et al. 2020; McClure et al. 2020; Bavli and Jones 2022; Green et al. 2023; Brown et al. 2023), makers of medical devices (Kadambi 2021), and police and security guards in medical settings (Saadi and Ray 2023),¹⁷ leading Black patients to have worse diagnoses, treatments, and outcomes.¹⁸ Such mechanisms may explain the finding that Black Americans do not receive the same health returns on socioeconomic status that White Americans do (Boen 2016). Moreover, equivalent wealth will not necessarily provide the same health purchasing power to Black and White Americans; Black people typically pay more for goods and services of all kinds because of segregation (Williams, Priest, and Anderson 2016). Discrimination also limits the economic returns to wealth (Shapiro, Meschede, and Osoro 2013), particularly via smaller returns on housing investments (Thomas, Mann, and Meschede 2018).

To what extent do these lines of research bear on the question of what would happen if a major wealth-based reparations program were enacted? Research necessarily tells us in what ways wealth protects and fails to protect in the current context, in which racial wealth gaps are vast. It cannot directly tell us what wealth would or wouldn’t be able to buy in the vastly different context of a truly large infusion of wealth into a sizable portion of the U.S. Black population, an instance of the “SUTVA [stable unit treatment value assumption] problem” in causal inference (Morgan and Winship 2014, 48–55). For example, if wealth currently fails to protect because some of its benefits are offset by creating greater exposure to certain kinds of discrimination, it might protect more fully in a context in which wealthy spaces were more often Black spaces. More broadly, wealth redistribution could indirectly improve health by redistributing political power and might disrupt health consequences associated with being at the bottom of a social hierarchy, and positive feedback loops between increased health and improved political participation and democracy could result (Lynch 2023; Rodriguez 2018).

Consider neighborhoods. Neighborhoods are powerfully structured by income and wealth (Florida and Mellander 2015) and also powerfully structured by race independent of income and wealth (Taylor 2019; Coates 2014). Neighborhoods also seem to have some genuinely causal consequences for health, some of which flow through, for example, differential exposures to polluted air (Wodtke et al. 2022).¹⁹ To what extent would wealth-based reparations disrupt Black Americans’ differential exposure to pollutants? The answer seems to turn on the extent to which wealth allowed Black residents of the most polluted neighborhoods to move to safer ones (Pollack et al. 2023)—moves that, in the current context, are directly limited by discriminatory housing practices (Christensen, Sarmiento-Barbieri, and Timmins 2022)—and the extent to which a greater concentration of wealth increased Black communities’ ability to fight for their current neighborhoods to become less polluted (Currie, Voorheis, and Walker 2023; LaVeist 1992; Williams and Collins 1995, 377–78). In terms of this example, one way to cast the challenge posed in this article is this: Would articulating the lifespan gap as an independent target of reparations policy suggest that wealth transfers be supplemented with reparative environmental policies? If so, the deep links between place and health suggest that seeking to equalize lifespans might finally allow people to realize the twin mobility rights long denied by America’s racial regimes, the right to move freely and the right to stay where one is and have it be safe.

Finally, conceptualizing lost lifespan as a target of redress might serve an additional, expressive function: the social goods that follow from framing an injustice as something that needs repair (Russell 2022; Walker 2013).²⁰ The experience of the 9/11 Victim Compensation Fund, as recounted by Gillian Hadfield (2008), may be instructive. Many families filed for compensation much later than expected, and legal experts interpreted this as reflecting disorganization or an expectation of a greater monetary payoff through a lawsuit. Interviews with families, however, revealed that their reluctance in fact reflected a series of nonmonetary goals that they weren’t sure could be achieved by the very large settlements being offered to them: information obtained through depositions, accountability for their losses, and political reforms that would reduce the chance of similar tragedies. The analogy may not hold. In the 9/11 case, people who had experienced a loss of great public concern had to decide “whether to resist the public characterization of their legal interests as private and monetary” (Hadfield 2008, 648); yet because the White-Black wealth gap is already socially defined as private, wealth-based reparations would seemingly reframe it as public. Lifespan-based reparations, of course, might also reframe that widely misunderstood disparity (Deyrup and Graves 2022) as well. Nevertheless, the 9/11 example illustrates that one yardstick for evaluating programs of repair is how fully they redress—or even acknowledge—the fullest extent of human freedoms that America’s history of enslavement still tramples.

What Other Avenues Might There Be to Redress Lost Lifespan?

Proposals and policies that aim to redress short lifespans can adopt multiple strategies. They can target short lifespans directly, aiming to lengthen them. They can also aim to increase time freedom while people are still alive, perhaps granting greater control over time in total. Proposals can also be universal (indirectly racially equitable) or targeted, and targeted proposals can also be outcome-oriented or process-oriented (participatory). Whether universalist policies can count—in the context of an articulated motivation grounded in naming a specific harm (Walker 2013; Táíwò 2022)—as a type of reparation, or whether they can instead function as a meaningful alternative to reparations (Reed 2016; Taylor and Reed 2019) is, to say the least, debatable (Klein and Fouksman 2022; Russell 2022); but it is, perhaps, notable that the quintessential universalist American health program, Medicare, grew out of the civil rights movement (Smith 2016). In the next section, I provide some examples of types of policies that might provide some redress for stolen time, noting more and less targeted versions of policies that adopt a similar strategy to recovering time that has been taken. These possibilities, discussed in turn, are typologized in table 2.

One possibility involves a New Deal–style expansion of the Black medical and public health workforces and community health infrastructure. A community health infrastructure (Chowkwanyun 2022) could include creating permanent bases of funding and other support for community vaccination initiatives founded during the COVID-19 pandemic to expand their work into other health arenas (Faherty et al. 2021; Bile et al. 2022), perhaps drawing on models of infrastructure built around community organizations that developed to help HIV/AIDS patients (Kayal 1993); it could also draw from examples of medical-community collaborations to address the maternal health crisis for Black birthing people (Hardeman et al. 2020). Measures to radically expand the Black medical workforce could perhaps constitute a specific form of redress for the 1910 Flexner Report, which led to the closure of all but two Black medical schools; one consequence is the estimated loss of between ten thousand and thirty thousand Black medical graduates between the schools’ closure and 2019 (Campbell et al. 2020) and an unknown number of practitioners of adjacent medical professions (McLemore 2022, 51). To succeed, such efforts would need to also address factors that drive would-be Black medical and health workers out of their chosen professions, such as the “overpolicing” of Black medical residents (Ellis et al. 2023).

Evidence suggests that expanding the Black doctor pool could be extremely consequential for improving Black health. Research finds county-level correlations between Black primary care physicians and Black life expectancy (Snyder et al. 2023) and finds that Black newborns are twice as likely to survive the days surrounding birth if they have a Black physician (Greenwood et al. 2020). Particularly compelling is a unique experiment based on an Oakland clinic that randomly assigned doctors of different races to Black male patients. Analysis of doctors’ notes suggested that the Black doctors had higher-quality interactions with their Black patients than other doctors did, and understood their health struggles more holistically in context of their life circumstances. Most strikingly, the study found results about patients’ willingness to embrace preventative measures that, if scaled up to the population, suggested that having enough Black doctors could reduce men’s Black-White gap in cardiovascular mortality by 19 percent (Alsan, Garrick, and Graziani 2019). Notably, however, the context in which Black doctors make such a significant difference is one in which Black men suffer so many risk factors for cardiovascular disease; in a context that successfully addressed the nonmedical sources of this disproportionate disease burden, Black doctors would have fewer lives to save.

This type of policy would be relatively targeted; more universalist versions could be policies such as Medicare for All or other single-payer health system, expanded enforcement of pollution laws, or measures designed to limit gun deaths. Despite the targeted nature of an expansion of the Black medical workforce specifically, any such expansion would likely also produce a model for similar expansions directed at the rural medical workforce, which might considerably improve White health and longevity as well (Simpson 2020; Gujral and Basu 2019).

A different approach would be to try to increase time freedom among the living. Universalist possibilities abound: fair scheduling laws (Schneider and Harknett 2019) and other workplace protections; efforts to aggressively reduce the administrative burden required to receive social services (Herd and Moynihan 2019; Herd et al. 2023; Jackson 2021); targeted improvements to transit infrastructure or expansion of work-proximity housing for populations with particularly burdensome commutes (Roberto 2008; Preston and McLafferty 2016); and policies, like Sunday closing laws, that promote access to predictable and shared time (Rose 2016), to name a few. A policy that could be implemented in a targeted or a universalist form would be a paid sabbatical program (Day 2019). Higher education faculty sabbaticals are, among other things, a form of work time that nevertheless has salutary effects on well-being (Davidson et al. 2010); a reparative sabbatical could, instead, be a paid period free of any work or governmental obligations, with the right to return to work guaranteed (as with leave, from some employers, under the Family Medical Leave Act, for example), in which sabbatical-takers could pursue whatever they chose.

Programs and policies that reduced time burden might also have multiplier effects through improving health and thus ultimately lifespans, as suggested by, for example, Cynthia Colen and colleagues’ (2023) systematic framework identifying time use as a social exposure that patterns health inequities. Lack of time is a major driver of unhealthy behaviors (Strazdins et al. 2011; Venn and Strazdins 2017; Covert 2022; Jabs and Devine 2006) and of stress (Mullainathan and Shafir 2021), and consistent, long-term reductions in time scarcity might plausibly shape the subjective time valuations that also influence health behaviors (Daugherty and Brase 2010). Parents’ and kin’s greater control over time in key moments in children’s and other loved ones’ lives might also produce spillover benefits to health (Stearns 2015).

A final category is participatory. Redress might take the form of a dedicated reparations fund, distinct from individual payments, meant to target lost time in particular. A collaborative process among those deemed eligible for individual payment, whether African descendants of slaves or some broader constituency (Jones 2021), could allow recipients to decide collectively what would do the most to return their time to them. Such a process could be modeled on examples of participatory budgeting (Fung and Wright 2011) and draw on lessons learned in analyses of victim compensation funds, such as the importance of recipients feeling like active participants (Feinberg 2005, 274–75). No one can better determine what would return time to the people it is being stolen from than those people themselves. Moreover, participatory processes could, perhaps uniquely, remain open to identifying—and seeking to ameliorate—further, potentially overlooked, dimensions of harm arising from slavery and Jim Crow that can still be redressed, for survivors and descendants, today.²¹

On the other hand, there may be something unsatisfactory about seeking to redress a population’s lost time through a process that demands of that same population that they take part in a time-consuming process of political participation (on the time demands of participatory democracy, see Cohen and Rogers 2003). Moreover, even though the strength of a participatory process is precisely its open-endedness, that very characteristic can be hard to maintain in practice. “The democratic process begins by defining the democratic body” (Demsas 2023) and the geographic scale at which participatory processes take place would likely circumscribe what kinds of options they could realistically consider.