INTRODUCTION
Robert J. Sampson’s Great American City: Chicago and The Enduring Neighborhood Effect (2012) returned this small-town boy to several salient formative memories of childhood. Visiting relatives in Chicago (the alluring and frightening metropolis about a ninety-minute drive from my downstate hometown) provided my earliest experiences with big city life. Although I became somewhat familiar with the west side where my mother’s Uncle Frank, a consummate blues pianist during the early years of the twentieth century, lived until his death in 1960, I was more familiar with the south side where Sampson focuses much of his attention. During the mid-1950s to early 1960s, I frequently (and unknowingly) answered Sinclair Drake’s and Horace Cayton’s 1945 invitation to readers of their classic study of Black Chicago (Black Metropolis) to “stand in the center of the Blackbelt—at Chicago’s 47th St. and South Parkway.”Footnote 1
Drake and Cayton, referring to Chicago during World War II, proudly describe a bustling urban center of shops, businesses, and people (overwhelmingly working class) headed to or from jobs, while Sampson discusses the wonderment of the area’s transformation from an urban slum during the late 1960s to a gentrifying connection of blocks during the early 2000s. Timewise, my own experience with Bronzeville, a neighborhood on Chicago’s south side, was betwixt and between. From the early 1950s to about 1962, my mother’s sister and her husband lived on the corner of 45th and Vincennes (pronounced Vin sens, a local would proudly correct you) a mere three blocks from Drake’s and Cayton’s designated center of Black Chicago. My recollections of Bronzeville span that period when, as a grade school lad, I would frequently visit (during the summer for weeks at a time), and when we cousins (lacking southern relatives) would take turns going to the City, reversing the annual treks to the South familiar to most African American children.
Bronzeville of the mid- and late 1950s still possessed many of the characteristics of which Drake and Cayton boast. Walking down 47th street, between Vincennes and South Parkway, one would encounter working adults, children, and teens, all (unlike the late twentieth century’s cavalcade of hoodies and sagging pants) dressed, perhaps more cheaply, but stylistically remarkably similar to other Americans across the nation. Shops and a variety of businesses thrived, but Bronzeville was clearly in transition to the devastated ghetto of disadvantage described by academics and journalists alike during the last four decades of the twentieth century. In many ways, the heart of Sampson’s complex argument can be understood in terms of the answer to a seemingly simple question: why did the thriving Black Metropolis hopefully envisioned by Drake and Cayton fail to materialize? As I read his impressive book, Sampson’s answer would entail two parts. First, his concerted interrogation of contemporary social scientists’ overwhelming emphasis on individual choice as the driving force of social change; he argues, “This book shows that a ‘kinds of people’ or compositional explanation of neighborhood disparities and social processes that turns on individual selection fails to answer the questions” (p. 23). Secondly, he would advance what I read as his central thesis:
“Neighborhoods are not merely settings in which individuals act out the dramas produced by…bigger external forces, but are important determinants of the quantity and quality of human behavior in their own right” (p. 22).
For Sampson, the overwhelming depopulation of middle-class households from Bronzeville occurring during the 1960s and 1970s follows a logic of spatial inequality and interdependence conforming to the neighborhood selection patterns he carefully describes. I refer to the strong version of Sampson’s arguments implicating individuals’ residential mobility as both a cause and a reaction to spatial context that creates and sustains interdependencies between choice and spatial inequality (p. 238). To make this point that neighborhood effects are equally paramount, Sampson personifies neighborhoods with human traits arguing, “neighborhoods choose people rather than the common idea that people choose neighborhoods” (p. 327). An alternative metaphor is that inter-neighborhood movement and neighborhood composition are analogous to health and auto insurance markets. During the sorting process, middle and high socioeconomic status (SES) households choose among a variety of neighborhoods, while low SES households are directed to the assigned risk pool where they remain stuck with no real choices.
In or around 1961, my aunt Mary (a caterer with exquisite culinary skills) and uncle June (a member of one of the earliest waves of African Americans hired by the Chicago Transit Authority as bus drivers) would contribute to Bronzeville’s transition by buying a house above Sampson’s line at 79th street demarcating the black middle-class suburbs and the ghetto. Similar to most migrations, their participation in the middle-class depopulation of Bronzeville (famously stamped with causal significance in William Julius Wilson’s path-breaking The Truly Disadvantaged) contained push and pull motivations. The spiffy new house on Rhodes Avenue out past 84th Street, with its beautiful lawn and backyard supporting flower and vegetable gardens, surely appealed to my aunt and uncle who respectively had been born and raised in small towns in northern Illinois and southern Michigan/northern Indiana. However, explaining the depopulation of middle-class households solely by rational choice theory interacting with macro-level social change such as the Civil Rights Movement seems insufficient. The apartment they left in Bronzeville hardly conformed to extant academic and popular literature describing overcrowded housing in Chicago’s Black Ghetto of the period. The apartment’s convenient location to work and urban activities, its sunny kitchen filled with modern (1950s-style) appliances, good-sized bedrooms and bathrooms, and a spacious living room with lavish bay windows typical of the brownstones lining near neighborhood streets, should have retained many middle-class African American families. It should also have attracted others (especially young families unable to provide the down payment required for housing in suburbia). Understanding why this did not occur requires understanding why middle-income Blacks stopped moving in as much as why those already there left.
WHAT ARE NEIGHBORHOOD EFFECTS, AND HOW DO THEY WORK?
Sampson’s approach to a panoply of contemporary issues related to explaining the population composition of Bronzeville during the past sixty years is cautious, rigorous, and attentive to micro and macro variables. Meticulously careful to avoid creating black boxes, he asks, “what are the social pathways by which neighborhood effects are transmitted in the contemporary city” (p. 23)? His answer begins with the simple observation that neighborhoods are more than the totality of individual actions of residents, more than the sum of their parts. They are ruled by a spatial logic based on synergies created by characteristics of the neighborhood’s residents and the characteristics of neighborhoods in close proximity; both of which create a neighborhood’s reputation, and absent significant social intervention, its destiny.
Great American City is a brilliant analysis of urban life and sociological urban methods that merits the high praise meted out on the book’s back cover and in the Foreword by social science luminaries. My one major criticism pertains to an interconnected double omission. In Chapter 2, an otherwise excellent survey of the origins of urban sociology and ecological methods is quite disappointing with respect to its continuing Sociology’s long history of ignoring the signal contribution of W. E. B. Du Bois’s The Philadelphia Negro (1899), a work clearly influenced by research Sampson lauds highly—Charles Booth’s Life and Labor of the People in London (1885–1905)—and an exemplary exhibition of the use of descriptive and analytical methods applied to the neighborhood geography of a city that predates the Chicago School and could hardly have been unknown to its founders. Moreover, neglect of The Philadelphia Negro enables the omission of a discussion of Du Bois’s reliance on social psychological factors in disadvantaged minorities’ construction of individual and group identities to provide a micro-individualist foundation linking macro factors and individual choices to underpin neighborhood effects (more on this below).
To more carefully identify an example of a neighborhood effect, it is useful to examine three of Sampson’s most significant theoretical concepts: (1) social altruism, or the presence of other regarding behavior among neighborhood residents (p. 127); (2) collective efficacy, or “the linkage of cohesion and mutual trust among residents with shared expectations for intervening in support of neighborhood social control” (p. 37); and (3) social disorganization, or the inability of a community to realize the common values of its residents and maintain social controls (p. 22). Sampson’s definition of one of the Chicago School’s leading conceptual tools, social disorganization, is important because it does not assume residents of disadvantaged areas possess values different from the rest of society. The possibility that they do not, but find it more difficult to realize those values is there. Furthermore, the definition allows social disorganization to coexist with community organizations (e.g. gangs, dense kinship networks demanding resources that undermine individual social mobility) that are either antithetical to or a hindrance to establishing behavioral norms consistent with community values. Thus, although I suspect he would agree the concept is at least misnamed, Sampson correctly notes that some sociologists’ long-lived attacks against social disorganization’s use—ethnographers find plenty of social organization (gangs, kin networks, and churches, etc.) in high crime and poverty areas—conflates organization with control or efficacy.
To explicate my understanding of Sampson’s use of these interrelated concepts, I return to Bronzeville nearer the mid-twentieth century. Already, during the 1950s, signs were indicating that “collective efficacy” and “social disorganization” were, respectively, eroding and ascending indicating that Bronzeville’s ability to attract and retain middle-class households was deteriorating. During 1957, emerging from a Walgreen drugstore where a speaker system blasting Thurston Harris’s R&B hit Little Bitty Pretty One forever imbued the feel of twentieth-century Black urban life in my neural paths, my cousin Bobby and I were mugged on 47th Street in broad daylight while dozens of passers-by ignored our predicament. The two tall African American teenage boys demanding our money were imposing to a couple of ten-year olds, but this being our second mugging, we had become streetwise. Although we were on our way to the movies and had a couple of dollars between us, we claimed to be returning from the movies and possessing only a few pennies. “Give it to me and you better have more next time,” the tough-looking boy with the cast on one arm threatened. The contempt we felt for these two—willing to strong arm two kids for a couple of pennies and gullible enough to fall for our evasion—was matched by our condemnation of the two White teenage brothers (soon destined for reform school) who had strong-armed us for a dime back in our home town earlier that year. However, the hometown mugging had occurred with no one else around. Had anyone else been present (and especially any adults as was the case on 47th Street), the two White teens would never have attempted the mugging. Moreover, if they had, the witnesses would have certainly intervened.
The palpable lack of social altruism demonstrated by the African Americans tending to their own business on 47th Street that Saturday afternoon long ago is consistent with the conclusions Sampson draws from the famous “lost letter experiment” (pp. 217–233). Sealed envelopes addressed and stamped were dropped on streets throughout Chicago neighborhoods to ascertain social altruism as measured by the percentage of letters subsequently dropped in mailboxes and reaching their addressees. However, despite Sampson’s assertion that returning a lost letter has no real costs for its finder, I am not convinced there are no unaccounted for income effects confounding the interpretation of return rates as measures of social altruism alone. The incentive to open a letter and ascertain its contents likely increases as the income of the letter finder decreases. Then once a letter has been opened, I suspect it is less likely to be returned. This suggests that altruistic behavior (as opposed to altruistic sentiment) does have a price and thus, depending on differences in neighborhood incomes, the behavior may vary more than the sentiment, making it difficult to disentangle income effects and altruistic behavior.
With this caveat in mind, I note the percentage of recovered letters differed significantly depending on the neighborhood where they were dropped. Unsurprisingly, letter return rates were lowest from neighborhoods that were poor and minority with return rates lowest at sites near high-rise public housing. Return rates were also positively and negatively correlated, respectively, with collective efficacy and moral cynicism (measured by factors such as belief in the legitimacy of laws and social institutions). On the basis of several statistical analyses that he recognizes do not provide foolproof evidence of a causal effect, Sampson argues, plausibly for the most part, that social altruism has a “neighborhood level path dependence that is influenced by long-term dynamics” setting “in motion actions and norms that govern future altruism” with effects on future events (p. 231). Examples of effected future events include homicide and teenage birth rates both intricately tied to neighborhood residents’ willingness to actualize other-regarding sentiments to intervene in support of neighborhood social control. Largely because of a high level of mutual trust, others will support interventions analogous to threatening to call police when observing children being mugged.
Great American City argues (again with sensible usage of correlation and principle component analysis to identify the most significant relationships) that collective efficacy—that important determinant of a neighborhood’s quality of life—is strongly and positively associated with the number of collective organizations operating within the neighborhood. In this regard (perhaps for reasons of sensitivity), Sampson’s discussion of local organizations and neighborhoods fails to adequately explore the links between neighborhood quality and churches. Churches, of course, are easily the most ubiquitous local organizations in neighborhoods of concentrated disadvantage. His expressed surprise upon finding a negative statistical relationship between church density and collective efficacy is perhaps due to the prominent role of churches and ministers during the Civil Rights Movement and the continuing role of many in social activity today. However, his surprise also reflects the fact that sociological literature generally ignores the traditional split among African American churches over the role of the church in civic affairs, with some eschewing politics.Footnote 2 Sampson says “cause” and “effect” in this negative correlation is not clear, but is likely connected to the finding that church density is also highest in areas of least “trust” (pp. 205, 207–209).
In my view, the finding of greater church density in African American and more disadvantaged areas really opens the question of what role churches are assuming in addressing the emotional, psychological, and civic life of parishioners. Despite relatively high church attendance and the fact that many ministers remain politically powerful, clearly, many churches are not aiding in the solution of basic problems such as under-performing schools, crime, and teen births. In other words, dense religious organization is not providing neighborhoods with adequate social capital to maintain social controls.
Lack of trust and dense and weak network ties—another important theoretical tool Sampson utilizes—assume a role in explaining the Church-disadvantaged neighborhood relationship. First, lack of trust (and cynicism) in such neighborhoods extends to the intent of churches and ministers. The view of rapper Brad Jordan (Scarface) excoriating the crass materialism and distrust of clergy exemplifies views I have heard from many African Americans spanning a wide band of the socioeconomic class spectrum: “They opening up these churches for some quick cash,… I follow no man for man be fallen,” (Jordan Reference Jordan1994).
Many of the churches comprising the statistics in Sampson’s data are very small congregations often operating as insular groupings viewing the larger community within which they are imbedded as outside the faith—hence, people with whom association will lead to spiritual corruption. Such congregations are often islands unto themselves exhibiting dense social ties within, but very weak ties without. The weak social ties of these insular organizations (as opposed to big tent churches) provide neighborhoods with little access to institutions outside the neighborhood. In this sense, the high density of churches in very disadvantaged neighborhoods is a misleading indicator of collective efficacy and social organization in much the same manner as presence of organized criminal gang activity is a misleading indicator of social organization, as Sampson defines it.Footnote 3 Although an insulated minister and a local gang leader may aid someone to obtain resources and get something done within their respective domains, their lack of ties with the wider political leadership structure throughout the city undermines effective social change even should they wish to see such change.
To identify an example of a neighborhood effect as promised above, I return to Bronzeville of the 1950s. The summer before Bobby’s and my mugging on 47th Street, I spent several weeks in Chicago, much of the time avoiding a vacant lot behind a supermarket at the corner of 45th and South Park Way where a gang of boys, ages eight to twelve, played. Whenever they saw me on the street, these boys would accost me with entreaties, then threats, that I would have to join their gang. My aunt and uncle had cautioned me not to associate with these gang members because they were “headed for trouble.” I improved my track skills running from those boys who informed me at every sighting that I was guaranteed an “ass whipping.” One afternoon, my aunt sent me to the store instructing me to go to the supermarket and not the small basement mom-and-pop store in the middle of the block that would allow me to avoid the vacant lot. Despite walking several blocks out of the way to approach the supermarket from a direction avoiding the vacant lot, I was spotted, then outrun and caught. Why, the lead boy demanded, was I avoiding the inevitable fact that I would have to join the gang or face beatings every time they saw me? I explained that I was far more afraid of what my aunt and uncle would do and think of me for hanging out with them than I was of getting beat up. Surprisingly, after some rough handling I was let go with no beating.
Clearly, I was fortunate to be backed by strong home values. But I suspect most, if not all, of these boys had parents who strongly disapproved of the gang. The point here of course is not that my aunt and uncle were likely all that different from the other parents. Needing merely to avoid joining the gang for a few weeks before returning home, I was in a completely different situation from the other boys. Had their area been my neighborhood, their school my school, I almost surely would have had to surrender and join the gang. A rap group, The Juvenile Committee (1993), explains how in low-income inner-city areas, neighborhood effects in the form of dangerous and overwhelming peer group pressure steer children away from comporting themselves in accordance with mainstream social norms:
you gotta come up an’ choose a side\ then you turn thirteen and get your own deuce\ five\ or you can try to be cool and go a different route\ but knuckleheads wanta see what you’re all about
Children locked in neighborhoods of concentrated joblessness are required to “choose a side.” However, choosing to present oneself in a manner calculated to bring rewards from mainstream society (in this case forsake being strapped with a twenty-five caliber handgun) would make one vulnerable to the opportunistic and violent behavior of many members of their peer group seeking to enhance their gangsta reputations at the expense of easy-looking prey.
Under such circumstances, the only decision available to my aunt and uncle that would prevent the gang inscription of their young kin was to move. They made that decision a few years later after my older cousin Ron was intensively recruited by a local gang of teenagers, many of whom later became part of the gang structure centered around the Black Disciples and Blackstone Rangers—two dangerous drug and criminal gangs essentially ruling the streets throughout Bronzeville during the 1960s and 1970s.
PERSONAL CHOICE AND NEIGHBORHOOD EFFECTS: MOVING TO OPPORTUNITY?
Probably, the book’s most sustained defense of the ubiquity and policy relevance of neighborhood effects is its discussion of the federal government’s “move to opportunity” experiment (MTO) conducted during the late 1990s. Families living in five cities—Baltimore, Boston, Chicago, New York, and Los Angeles—who resided in areas of concentrated poverty (census tracts with 40% or more households below poverty) and were below the poverty line themselves, were deemed eligible to apply for a housing voucher. To address the problem of selection bias, each applicant was randomly placed in one of three groups: experimental, section eight, or control. Those in the experimental category were offered vouchers that could only be used for housing in a neighborhood with less than 10% poverty; the section eight group were offered vouchers with no location restrictions, and the control group received no treatment. The counterfactual suggested by most discussants of the MTO is “whether an individual residing in a poor neighborhood would follow a different course if he or she in fact resided in a non-poor neighborhood” (Sampson 2012, p. 263).
Five measurable outcomes have been examined to operationalize the counterfactual “follow a different course”: adult economic self-sufficiency, mental health, physical health, education, and risky behavior. Statistical studies of the MTO data have produced mixed results with respect to the question of whether neighborhood effects matter. No significant differences between experimentals and controls have been found for adult economic self-sufficiency or physical health; significant positive effects on adult mental health; young female physical and mental health; or young female education and risky behavior (crime and delinquency). However, negative effects have been found for the physical health and delinquency of adolescent males, and there appear to be no effect on cognitive achievement.
In discussing the MTO’s less than sterling support of the importance of neighborhood effects (except for young girls), Sampson never disavows the obvious fact that individuals make choices. However, he stresses they do so within a variety of controlling contexts. One way of summarizing his detailed critique of the MTO’s methodology is to realize we cannot model many of these choices (schooling, criminal behavior, job seeking, and marriage) as simple linear processes where behavior in schooling (study habits or deportment) are the dependent variable and various individual and neighborhood traits are independent variables. To make sense of Sampson’s argument, it must be that choice-based behaviors are imbedded within a system of equations, and, importantly, at least one of the equations containing neighborhood effects and individual effects is nonlinear. In such a system, it will be highly problematic to speak solely of an individual effect. This is the point Sampson is making when he observes, “As poor MTO families move from one neighborhood to another, entire bundles of variables are also changing at once, making it difficult to disentangle change in neighborhood poverty from simultaneous changes in correlated structural and cultural features of the environment” (p. 265).
This point correlates well with the book’s persuasive argument against the quantitative empiricist position pushed by many social scientists, that methods utilizing randomized selection of participants into experimental and control groups constitute “real” science that is superior to alternative methods. Sampson is no methodological Luddite, and thus he vigorously supports the importance of studies (including the MTO) pursuing statistical analysis of properly specified randomized experiments. His second point, complementing the problems of simultaneity and nonlinearity among the variables, concerns model design and the fact that just as do alternative social science methods, randomized experiments always involve a host of implicit assumptions upon which the validity of any interpretation of the results depend. In this regard, I find two of the book’s points most illuminating.
An important weakness in the MTO data is analogous to a criticism made against “twin studies” that attempt to isolate the different effects of nature and nurture on some outcome such as IQ by studying identical twins reared apart. A criticism leveled at such studies is that the separated environments experienced by most such twin pairs are not sufficiently different to enable an experimental design capable of isolating the effects of genes and environment. Just so, Sampson marshals considerable evidence that the origin and destination neighborhoods of experimentals are (1) insufficiently different within the experimental group, and (2) insufficiently different when comparing the experimentals with the control group. Although experimentals did move to areas with lower poverty rates, according to Sampson, differences in origin and destination neighborhoods were differences of degree, not kind. Destination neighborhoods were still overwhelmingly segregated, composed of low-performing schools, and spatially embedded within contiguous neighborhoods of disadvantage far greater than the average American would ever experience. For example, among other risks, “MTO controls and experimentals both tended to end up in areas characterized by concentrated disadvantage, racial segregation, low social support, compromised public health, and low collective efficacy” (p. 274).
I note the connection between this latter weakness in the MTO’s design and Sampson’s implied point that the constellation of effects set in motion by a household moving involves a system of equations with nonlinear components. Indeed, we may infer from Sampson’s discussion that he believes neighborhood effects are likely subject to threshold effects and not the continuous linear effects implicit in the statistical methods used to analyze differences in average values between experimentals and controls used in MTO studies. Threshold effects exist if neighborhood effects are similar across a range of poverty rates. Then, moving from a concentrated poverty neighborhood of 40% or more to one of 10% that is embedded within a larger area encompassing neighborhoods of concentrated poverty would have little or no effect on opportunities and outcomes. A decisive test would require moving experimentals to neighborhoods with bordering neighborhoods far different from those of origin.
The second of Sampson’s critiques of the MTO concerns his distinction between neighborhood “developmental effects” (how neighborhoods affect the behavior of individuals “no matter where they are”) and “situational effects” (variation in event rates across neighborhoods “regardless of who commits the acts (residents or otherwise)” (p. 264). An example of the latter is the greater propensity of residents and non-residents alike to participate in illicit activities in neighborhoods known to have a high level of street drug activity. However, it is the first class of effects I want to discuss. Sampson’s “developmental effects” are a category of what I call “socialization effects.” For purposes of this essay, I define the latter as modes of behavior defining a social identity constructed from an individual’s need for self-esteem within a specific social environment. Sampson’s critique of the MTO study may then be discussed in terms of the question: How adaptable are individual social identities at various stages of the individual’s life-cycle?
Sampson’s answer would appear to be, “not very adaptable,” at least for individuals who have reached post-adolescent stages of the life-cycle and forged more or less enduring social identities represented by various attitudes and modes of behavior important to their sense of self. His use of the term developmental effects may then be considered a demarcation indicating that socialization (an ongoing lifetime process) is in the pre-adolescent stage when its effects are most malleable should the individual’s social environment change. One way of thinking about this is in terms of “temporary” versus “permanent” behavior analogous to concepts introduced to adumbrate differences in consumption and saving patterns over the life-cycle in two famous theories of economic behavior. Did experimentals in the MTO study who moved to new neighborhoods come to believe they were a permanent part of the new environment, and would not return to the old, and therefore it would make sense for their behavior to adapt to the new neighborhood by adopting its norms as their new permanent guides of behavior? Or, did they see the new environment as a temporary abode and something that would not merit a wholesale adoption of new ways of being and self-presentation?
What Sampson is saying, though he is too polite to do so directly, may be stated colloquially, as I have heard many poor, working, and middle-class African Americans say: “You can take the boy out of the ghetto, but can you take the ghetto out of the boy?” And, for those who believe the answer to this theoretical question is yes, the policy question is, how large a change in opportunities (educational quality, peer group norms and behaviors, job prospects, and spousal networks, etc.) is required at various stages of the life-cycle? Sampson’s detailed overview of the rather modest changes in neighborhood opportunities available to experimentals exiting a neighborhood of concentrated poverty suggests strongly that the MTO has not answered these most important questions. In fact, we may confidently say, the MTO study has shown that voluntary movements of the poorest, most disadvantaged families from the most disadvantaged urban neighborhoods to environmental settings exhibiting modest improvements in poverty ratios, quality of academic performance, and racial-ethnic makeup, induces no improvement in adult self-sufficiency, modest improvement in some categories of adolescent and teenage girl schooling, and negative effects on the behavioral outcomes for adolescent and teenage boys.
Given the exacting demands to prove one’s manhood placed on young males socialized in urban neighborhoods of concentrated poverty and adult joblessness during the past several decades, the MTO study’s negative findings with respect to this group are hardly surprising. I believe it was another Harvard sociologist, Lee Rainwater, who, commenting on disadvantaged urban populations that would have been the grandparents of those participating in the MTO, concluded that the modes of behavior and self-presentation learned in the poorest housing projects of St. Louis prepared children well to survive in that environment, but failed them miserably to succeed in the wider society. Rainwater, of course, without using the term neighborhood effect, was referring to the possibility that neighborhoods have effects that will likely follow their children to new neighborhoods.
About a decade after my summer evading the youth gang of 45th Street, by sheer chance, I encountered two survivors of that neighborhood (then age nineteen, like me) at the base camp of the 199th Infantry Brigade located outside the city Long Binh about eighteen kilometers North East of Saigon, South Vietnam. Observing and coming to know these two intelligent young men magnified my growing awareness of the costs and benefits of my growing up racially isolated in a small town in the north. Differences in our deportment, relationships with Whites, and worldview were striking. I was continually struck by their racial cynicism and opportunistic predatory attitudes; they, by what each considered my incredible naiveté in such matters. The one with the exceptionally exaggerated street presentation (pimp walk as we called it back then) guaranteed to provoke mistrust among sergeants and officers, I only knew a few weeks. The other, who circulated among my group of close knit friends for several months, finally paid a price for his penchant for breaking rules. Late one night (or early morning depending on perspective) the camp’s perimeter erupted with rapid machine gun fire and a voice yelling “they’re coming; they’re coming.” No one was coming. My friend had over breached the Army’s lax attitude enforcing regulations against marijuana smoking by off-duty soldiers coping with duty in a combat zone. Smoking while on perimeter guard duty, he had hallucinated an attack by Viet Cong dressed in black pajamas. His punishment was a transfer to more hostile territory. The last time I saw him he was part of a grungy procession from the “Big Red One” (1st Infantry Division) returning from the Pentagon’s latest tactical operation in jungle warfare and hiking through our camp. An M16 strapped behind his back, unshaven for what looked like weeks, and with dirt and mud caked upon his clothing, his eyes betrayed the wearied, forlorn look of a heightened constant death anxiety.
POLICY IMPLICATIONS
It is useful to discuss the policy implications of Sampson’s most impressive achievement in terms of a policy strategy and the tactics used to implement the strategy. With respect to a policy strategy, Great American City calls to mind an earlier generation of social scientists’ debate over the comparative merits of ghetto gelding versus ghetto dispersal as national policy.Footnote 4 Sampson’s argument for policy intervention (tactics) at the level of neighborhoods, rather than solely at the individual level, places him squarely in the camp of those who advocate public investment in improving the quality of life in poor people’s current neighborhoods.
This implication of the book’s general analysis is perhaps most salient when it addresses the evidence produced by the MTO experiments. It is important to stress again that Sampson does not dismiss the important role of individual choice and behavior. Rather, he argues that individual and neighborhood effects are distinct but complimentary. Their complimentary nature makes it extremely difficult, perhaps empirically impossible, to separate their effects, a proposition leading to the inference that Sampson’s critique of MTO shows neighborhood dispersal is not likely to improve the life chances of most individuals.
Because of socialization effects locked into individuals’ post-developmental sense of self and attitudes toward society, intervention directed at the level of an individual is unlikely to be effective unless dispersal of individuals assumes massive relocation of low-income minorities distributing them spatially. Thus, for example, poor African Americans who compose only about 3% of the U.S. population would be represented approximately at that percentage everywhere. Such a result would be prohibitively expensive, would provoke challenges from receiving communities (think of the school busing revolts of the 1970s and the turmoil caused by attempts to disperse residents of condemned public housing projects), and would provoke strong resistance from poor minorities resisting social and racial/ethnic isolation.
I find myself persuaded by Great American City’s tactical approach to urban policy. Policy must always work through choices made by individuals, Sampson writes, “but it should do so through experimental designs that randomly assign interventions at the level of neighborhoods or other ecological units, not individuals” (p. 264). Such policy interventions would measure the effects of neighborhood interventions by comparing treatment and control neighborhoods. Sampson’s example of such an intervention is the passing out of free needles to addicts to measure the effect on HIV rates across neighborhoods. I would add that randomly assigned interventions at neighborhood schools and employment programs should be at the top of the list.
