The Evolution of an Idea
While the twenty-first-century city has been declared spatially liberated, I argue that it remains place-based in much of its character. As chapter 1 contends, social science needs revised conceptual approaches and methodologies to chart this reality. In this book I attempt to meet the challenge, taking as my laboratory the social landscape and continued vitality of Chicago and guided by the inspiration of more than a century of careful scholarship.
The reader will be forgiven for suspecting my approach is in important respects a descendant of the well-known “Chicago School” of urban sociology, which many triumphantly consider dead.1 There are now LA Schools, NY Schools, French schools, the “new urban sociology” that originated in the 1970s, and undoubtedly more to come. Now is not the time to review claims of intellectual supremacy. Besides, Chicago was not even the first to seriously engage the idea of neighborhood effects, and I agree with Harvey Molotch’s exasperated reaction—“School’s Out”—to a recent Chicago-LA School shootout.2 My position is that there are many nuggets of insight worth preserving from our intellectual predecessors and that the goal should be to transcend tradition rather than view it as something to flee altogether. What matters is not whether a theoretical approach is old or new (or what “school” it belongs to) but whether it is productive for the question at hand. I thus discard what proved to be clearly wrong in the old Chicago School of urban sociology but preserve its key insights and draw upon newer schools and paradigms (e.g., political economy, social networks) when they are clearly right for the task. I also strive to recognize the changing nature of the city in global (or cyber) society in ways that take advantage of modern technology to collect new forms of data on community.
Although many publications have told the “Chicago School Story,” the threads have not been connected to the main plot in a systematic way that derives productive implications rather than just criticism. As a result I aim for breadth rather than depth about the Chicago School, painting it in broad strokes with an eye toward highlighting the long-term evolution of intellectual thought about neighborhood effects and extracting principles for moving forward.3 Although historical exegesis is not my goal, getting the sequence of facts right is central to any social inquiry. It helps that the facts are interesting and, for the pre–Chicago School history, often neglected. As one set of authors put it more generally (over seventy years ago), “The enthusiasm of social scientists often leads them to attribute greater originality to contemporary studies and less value to the old than is actually warranted by the facts in the case.”4 I thus begin the journey overseas where pioneers in criminology and epidemiology set the stage for modern social inquiry into the city.
Origins
The ecological concentration of a broad sweep of human conduct and general wellbeing has a venerable history that predates Chicago’s founding. Contextualized approaches appear to have originated with social statisticians intrigued by crime patterns. In 1833 André-Michel Guerry published Essai sur la Statistique Morale de la France and the cartographic method of presenting the statistical distribution of crime. Using judicial statistics based on the quarterly returns of the local public prosecutors, Guerry was one of the first researchers to empirically document the variation in crime rates within a country. Contrary to the then prevailing assumption derived from the notion of individual free will— that crime was randomly distributed—Guerry demonstrated that crime and other social behaviors occurred disproportionately in certain areas and at certain times. Guerry also analyzed the distribution of crime among geographical units (administrative departmental units) in relation to what are now standard sociological variables, such as education, income, urbanization, and density. He found that units with the greatest concentration of manufacturing industries had the highest rate of property crime. Like Durkheim, Guerry’s inquiry aimed to demonstrate that social facts influenced human behavior independent of individual biology and psychology. Guerry’s quantification of social data and ecological analysis served as a stepping stone for ecological researchers of the twentieth century, earning him the title of “the first of the social ecologists.”5
Adolph Quetelet followed in Guerry’s tradition and expanded the use of statistics to analyze differences in crime type between France, Belgium, and the Netherlands. Quetelet found that the greatest number of convictions for crime occurred in Southern France and in the southern part of Belgium and the Netherlands.6 Like Guerry, Quetelet found that there was a stronger “le penchant au crime” (propensity to crime) against persons in Southern France and a stronger propensity to crime against property in Northern France. Quetelet interpreted his data as supporting the concept of “social physics,” which was a milestone in the theory of sociological ecology according to Morris.7 Although Quetelet’s ideas are probably better known than Guerry’s, Quetelet was only marginally interested in the study of the spatial relations of people and institutions within a given habitat. Still, both Quetelet and Guerry were able to demonstrate in convincing fashion that crime was not a randomly distributed phenomenon but rather was concentrated disproportionately in the same areas over time.
Rawson shifted Guerry and Quetelet’s geographical focus of analysis from France to the distribution of crime in England and Wales. He argued that since “the employments of the people exert the most important influence upon their physical condition,”8 occupation should be an important determinant of their moral and social character. Dividing counties of England and Wales into four occupational types (agricultural, manufacturing, mining, and metropolitan London workers) on the basis of the 1831 census, Rawson found that crime prevailed to the greatest extent in metropolitan occupational areas. He found a low incidence of crime in the small mining counties and the mountainous districts of Northern England and Wales, and no differences in the crime rates between manufacturing and agricultural counties. Rawson’s strategy was one of the first examples of the typological approach in ecological analysis, which resurfaced over a century later in what came to be known as “social area analysis.”
As impressive as the work of these scholars was, a major turning point in the study of the nineteenth-century city came in the form of now classic works in London by Henry Mayhew9 and Charles Booth.10 Both scholars provided detailed documentation and early visual portrayals of the neighborhood context of multiple indicators of “pathology.” In the same vein as Clifford Shaw’s The Jack-Roller,11 Mayhew provides a unique picture of London in its historical and ecological setting. Based on scores of interviews and ecological analysis, Mayhew claimed that crime was learned and passed down in areas characterized by poverty, drunkenness, bad housing, and economic insecurity.12 Mayhew foreshadows the idea of “cultural transmission” put forth later by Shaw and Henry McKay, who documented the transmission of delinquent values across generations in areas characterized by social disorganization.13 Mayhew also linked the incidence of crime to the opportunities that areas characterized by taverns and lodging houses afforded thieves and prostitutes. In effect, he called for studies of crime to focus on the ecological distribution of environmental opportunities that are presented to potential offenders.
Booth’s empirical efforts were even more impressive and stand the test of time in terms of methodological detail and multimethod inquiry of social context. Indeed, perusing the many volumes of Life and Labor of the People of London is both a humbling and invigorating experience. Well before the advances offered by modern research technology, Booth painstakingly captured the wide ecological variability that Dickens’s London offered up. The Booth collection holds more than four hundred original notebooks from his surveys and observations, which include detailed assessment of the environments in which Londoners lived, as well as interviews with policemen on their beats, factory owners, ministers, and a diversity of city dwellers. Booth’s Inquiry mapped poverty and wealth street by street.14 It is worth noting as well that the oft-cited discovery of the etiology of cholera in London by the physician John Snow in the 1850s was inextricably intertwined with mapping the ecological unfolding of the disease.15 Taken together Booth and Snow represent an early example of neighborhood logic in action.
The Green Bible (a.k.a. Introduction to the Science of Sociology)
Although a social-ecological approach to cities flourished in Victorian London, the approach of sociologists at the University of Chicago brought neighborhood-centered research to the fore of the discipline during the early twentieth century. This is one of the most well-known turning points in sociology and, more generally, social science research.
Before World War I and well into the twenties, Robert Park, Ernest Burgess, and a number of their students took to the streets to investigate the ecological contexts of a wide range of social behavior.16 The famous concentric zone map, with its central business district forming the core of an expanding city, still fascinates and adorns the seminar room to this day at the University of Chicago. The “Green Bible” of Park and Burgess served as a touchstone to generations of scholars.17 The Chicago School proposed that cities were divided into numerous functioning “natural areas,” which exhibit distinct physical and cultural characteristics. Park’s human ecological framework focused on the interaction between the processes of human nature and the “metes and bounds” set by the physical geography of the city.18 Familiar ecological terms such as competition, invasion, succession, segregation, and symbiosis stemmed from his analysis of the interrelations between the biotic and cultural levels of community. The concentric zone map derived from Burgess’s hypothesis that social differentiation in cities was expressed ecologically as a radial expansion from the business district outward.19
Clifford Shaw, one of the most noted criminologists of the twentieth century, operated within the theoretical framework of Park and Burgess. The ideas put forth by Park and Burgess and a group of largely unsung social workers20 set the stage for the publication of Shaw’s Delinquency Areas, in which he proposed that “the study of such a problem as juvenile delinquency necessarily begins with a study of its geographical location.”21 To accomplish this, Shaw studied the distribution of delinquent boys and girls brought before the Juvenile Court of Cook County from 1900 to 1927. He utilized basic statistics to measure the delinquency rate (ratio between the number of delinquents and the total population of the same sex and age groups) and was one of the first American sociological researchers to demonstrate the marked variations within a major city. With his later associate Henry McKay, Shaw showed that the highest delinquency rates in Chicago were located in deteriorated zones in transition next to the central business and industrial district. Rates of delinquency decreased as distance from the center of the city increased, the exceptions being areas also characterized by industry and commerce. These findings led Shaw and McKay to conclude that delinquent behavior was closely related to the growth processes of the city as outlined in Park and Burgess’s masterwork, The City.
Community environments were a focus of early research in the field of epidemiology as well, with an exemplar found in the early twentieth-century research of Joseph Goldberger and his colleagues on the vitamin deficiency known as pellagra.22 In a study of the distribution of pellagra in cotton-mill villages of the South, Goldberger and colleagues found that its contraction was related not only to individual-level socioeconomic status but also to the availability of nutritional foods in villages. They amassed an impressive array of both individual- and village-level data relating to food supply and malnutrition that included village-level measures of the prevalence of retail grocery establishments and home-provided foods, and contrasts in the type of agriculture in farm areas surrounding the villages. Their study also provided early evidence of an interaction between individual- and community-level health risks: in villages with fewer supplies of nutritious food, family income was “less efficient as a protective factor than in other similar localities with better conditions of food availability.”23 Goldberger’s work was widely influential and illustrative of a much broader tradition of epidemiological research on the ecological context of health that dates back to the early nineteenth century. His work stands as testament to the fact that the clustering of disease and the notion of “concentration effects” have long been with us even if their significance has been neglected.24 Neighborhood effects on health have been rediscovered and now make up a booming area of research.25
Social Disorganization Theory
In their major work published in 1942, Juvenile Delinquency and Urban Areas, Shaw and McKay extended Park and Burgess’s ecological theory of cities to include neighborhood characteristics and mediating social factors to explain their importance. They identified low economic status foremost, but ethnic heterogeneity and also residential instability as the three consistent predictors of delinquency rates. They discovered that high rates of delinquency in Chicago persisted in low-income, heterogeneous (usually immigrant) areas over many years. Moreover, the same Chicago neighborhoods characterized by poverty, residential instability, and high rates of crime and delinquency were also plagued by high rates of infant mortality, low birth weight, tuberculosis, physical abuse, and other factors detrimental to child development. Shaw and McKay argued that delinquency “is not an isolated phenomenon”26 and went on to document the close association of delinquency rates with a host of social problems.
In a similar vein, Robert E. L. Faris and H. Warren Dunham27 applied the ecological idea to mental health, showing that areas in transition with high rates of poverty had higher rates of hospitalization for mental disorders. Like Shaw and McKay, Faris and Dunham observed that high rates of adverse outcomes tended to persist in the same communities over time, despite the movement of different population groups through them. The teamwork of Faris and Dunham and Shaw and McKay formed the precursor to the general conceptual point and supporting maps shown by Drake and Cayton in Black Metropolis several years later.28
Guided by these findings, the Chicago School sociologists asserted that neighborhoods possess relatively enduring features and emergent properties that transcend the idiosyncratic characteristics of particular ethnic groups that inhabit them. Shaw and McKay specifically argued that criminal behavior was transmitted intergenerationally in neighborhoods characterized by “social disorganization.” The concept of social disorganization was defined as the inability of a community to realize the common values of its residents and maintain effective social controls.29 This definition later came to be operationalized in so-called systemic terms—that is, the disorganized community was viewed as suffering from a disrupted or weakened system of friendship, kinship, and acquaintanceship networks, which were seen as crucial for ongoing processes of socialization.30 Poverty, heterogeneity, and rapid population turnover were also thought by social disorganization theorists to reflect at the neighborhood level the larger processes of urbanization, industrialization, and social change emphasized by fellow Chicago School theorists such as Wirth. These neighborhood-level dynamics were thought to undermine personal ties, voluntary associations, and local institutions, which in turn were hypothesized to weaken the infrastructure necessary for socialization and social control, thus eventually leading to outcomes such as crime, delinquency, homelessness, and educational dropout. In this way the Chicago School theorists attempted to trace how macrolevel forces worked their way down to the local level in terms of community-level processes of social control and regulation.31
The theory of social disorganization did not go unchallenged. In Street Corner Society, William F. Whyte famously argued in 1943 that what looks like social disorganization from the outside is actually an internal organization.32 He discovered through extensive fieldwork an intricate pattern of social ties embedded within the social structure of the low-income Italian area of Boston’s North End; there were organized gangs and an integration of illegal markets with the routines of everyday life. Noting the relative nature of organization in the community, he maintained that the real problem of “Cornerville” was that its social organization failed to mesh with the larger structure of society around it. Whyte’s research came to be seen as a repudiation of the prevalent theory that slum communities were inherently “disorganized.” A bit later, circa the 1950s and 1960s, ethnographic research discovered thriving urban communities and ethnic enclaves where kinship and friendship solidarities flourished.33 Especially in poor urban neighborhoods, the evidence of dense social networks and local identification remained strong.34
Despite these criticisms, social disorganization theory survived and was even revitalized in later years, or reinvented, some might say, by the popular idea of “social capital.” Although there are many definitions of the term, social capital is typically conceptualized as a resource embodied in the social ties among persons—networks, norms, and trust.35 The connection of social disorganization to social capital theory was articulated by Robert Bursik to mean that neighborhoods lacking social capital, indicated by depleted social networks, are less able to realize common values and maintain the social controls that foster safety and efforts to promote social goods.36 Dense social ties thus play a key role in both social capital and disorganization theory. Perusing recent works on social capital reveals little that was not said in slightly different language by sociologists of the city from the birth of the discipline onward. Also like social disorganization, social capital as a concept has come under attack but continues to influence research on neighborhood effects.37
I will revisit some of these issues in chapter 7, but for now I would claim that what makes the Chicago School framework of the pre–World War II era of continued relevance to the study of neighborhoods is less its specific emphasis on disorganization (or, later, social capital) and more (a) its general emphasis on the characteristics of places rather than people, (b) its emphasis on neighborhood-level structural differentiation, (c) its notion of what we would now call “mediating social mechanisms,” (d) its focus on the concentration of multiple and seemingly disparate social phenomena, (e) its emphasis on the importance of dynamic processes of social reproduction over time, and (f) its recognition of larger macrosocial forces. One can admit to a concern over the concept of disorganization (which I do) while appreciating this broader Chicago School outlook.
Post-War Challenges
Theoretical work on neighborhoods leveled off and declined after the retirement of the Chicago School leaders and Whyte’s seminal critique. An empirical stumbling block also appeared at midcentury in the form of Robinson’s famous critique of the “ecological fallacy.”38 Robinson argued that individual-level relationships could not be accurately inferred from aggregate or ecological correlations. The main example he used was literacy rates and percent foreign born, where the state-level correlation was opposite (positive) that of individuals (negative). A backlash followed against ecological research, but for misleading reasons. Robinson’s mistake, and that of many readers, was to assume that ecological researchers only cared about individual-level inferences. Rather than arguing against ecological or neighborhood-level research, the right message was to make clear distinctions among units of analysis and to appropriately frame analytical questions, an early version of what we now call “multilevel” analysis. It follows that if the main goal is to explain rates of variation across neighborhoods rather than individual differences, Robinson’s critique does not hold. Moreover, worry about the ecological fallacy distracted attention from the “individualistic fallacy”—the often-invoked and also erroneous assumption that individual-level relations are sufficient to explain collective outcomes.
Despite the Robinson detour, empirical research in the mid-twentieth century documented the continuing ecological differentiation of American cities in the Chicago School tradition. In an influential work of this period, Eshref Shevky and Wendell Bell developed three constructs to reflect social differentiation and stratification in urban, industrial society: social rank, urbanization/family status, and segregation.39 Using government census tracts as units of analysis, social rank was measured by the configuration of occupation, education, and rent; urbanization/family status by single-family versus multiple-dwelling units, fertility, and female labor force participation; and segregation by the proportions of racial and ethnic groups living in relative isolation. Typologies based on this scheme came to be known as “social area analysis,” reflecting the idea that social stratification was manifested in geographical areas. Although criticized for reasons that go beyond present concerns, many other independent studies of American cities during this postwar period largely confirmed the Chicago School prediction that spatial differentiation occurs along dimensions of socioeconomic, family, and ethnic status.40
Political Economy of Place and the State
Another important objection to the Chicago School model, circa 1970 onward, turns on the idea that neighborhoods are deeply shaped by extralocal processes having their roots in economic and political structures. Ecological conceptualizations of the black ghetto as a naturally occurring phenomenon came under withering criticism.
It is hard to overemphasize the early influence of the notion of a “natural area” that conceived neighborhoods as dynamic, adaptive systems driven by free-market competition that existed between businesses for land use and population groups for affordable and desirable housing.41 Borrowing concepts from Darwinian theory, Park and Burgess focused on the “balance of nature” and argued that natural forces were responsible for the initial distribution, concentration, and segregation of urban populations.
In the 1970s a Marxian approach to the urban question, often called the new urban sociology, was launched by scholars such as Manual Castells, David Harvey and Mark Gottdiener.42 At around the same time the “political economy of place” perspective emerged, culminating in the work of John Logan and Harvey Molotch.43 These interventions repudiated market-based assumptions and the biotic model, arguing that the Chicago School paradigm ignored capitalistic production and political forces beyond the borders of the local community. From this view, neighborhood inequality in American cities is shaped directly and indirectly by the logic of capital accumulation and the “growth machine” of the city, especially the collusion of public officials and businesses. Structural or institutional racism was also implicated in the persistence of racial segregation, a direct critique of the spatial assimilation model of the Chicago School that seemed to imply that ethnic ghettoes were temporary and thus that the black ghetto would disappear.
The political economy story was powerful. It claimed that the decline and rebirth of many central-city neighborhoods in the postwar era was facilitated not only by individual preferences or voluntary migration, but by incentives for suburban growth in the form of tax breaks for developers and private mortgage assistance, highway construction and urban renewal, economic disinvestment (or not) in central cities, and zoning restrictions on land use.44 For example, consider public housing and the legacy of urban renewal in the crucial period of inner-city decline. Bursik has shown that the construction of new public-housing projects in Chicago in the 1970s was associated with increased rates of population turnover, which in turn predicted increases in crime independent of the area’s population composition.45 Wesley Skogan argued that urban renewal and forced migration contributed to the wholesale uprooting of many urban communities; a good example is how freeway networks driven through the center of many cities in the 1960s destroyed viable, low-income neighborhoods.46 The state has also been posited as a causal force through the differential provision by government of city services, public housing, and human welfare resources to some and not other neighborhoods. Loïc Wacquant’s comparative analysis of American black urban ghettos (typified by those on the west and south sides of Chicago) and French banlieues maintained that the decline of American urban centers after the 1960s is explained by the withdrawal of state resources, or “desertification.”47
Thus with the purposeful segregation of low-income public housing, withdrawal of needed services, government subsidized development by the private sector, zoning, red-lining, blockbusting, or something so simple yet powerfully symbolic as gated communities with no sidewalks, it is no longer possible to think of neighborhoods as purely natural areas created by the aggregation of individual preferences alone. Government, business, and the wider political economy along with processes of place stratification thus matter to our understanding of what communities can and cannot supply. So too do globalization and international forces such as migration. From this view, neighborhoods may be seen as mediating extralocal social forces. This does not mean that neighborhoods lose analytic value any more than individuals do in an interconnected world. It means that they take on a more complex, situated role and that our job is to examine both the internal and articulating mechanisms that tie neighborhoods to the larger social order. The next intellectual intervention made such an attempt.
The Truly Disadvantaged and “Concentration Effects”
In the latter part of the twentieth century, highly visible publications brought neighborhood-based research to the forefront once more. This time there was an even clearer focus on macrolevel forces that the early Chicago school researchers had tended to gloss. The best-known effort to accommodate the large-scale social changes seen in the latter decades of the twentieth century is William Julius Wilson’s notion of the “concentration effects” that arise from living in a neighborhood that is overwhelmingly impoverished.48 The Truly Disadvantaged marks another turning point in the study of neighborhood effects after the classic studies of Victorian London, the paradigmatic intervention of the Chicago School, and the political economy critique.49
Wilson argued that the social transformation of inner-city areas in the decades of the 1970s and ’80s resulted in an increased concentration of the most disadvantaged segments of the urban black population—especially poor, female-headed families with children. The consequences of these various ecological distributions are profound because they mean that relationships between race and individual outcomes are systematically confounded with important differences in community contexts (in fact, chapter 5 will demonstrate the virtual absence of white low-income communities, not just in Chicago but nationwide). Wilson argued that this racialized concentration of poverty and joblessness resulted from macrostructural economic changes related to the deindustrialization of central cities where low-income minorities were disproportionately located.50 These changes include a shift from goods-producing to service-producing industries, the increasing polarization of the labor market into low-wage and high-wage sectors, and the relocation of manufacturing out of the inner city, all of which can be linked to global economic trends. According to Wilson, the exodus of middle- and upper-income black families from the inner city also removed an important social buffer that could potentially deflect the full impact of prolonged joblessness and industrial transformation.51 The increasing stratification among blacks differed significantly from the environment that existed in inner-city neighborhoods in previous decades. Wilson argued that income mixing within communities was more characteristic of ghetto neighborhoods during previous decades, whereas inequality among communities today has become more pronounced as a result of the increasing spatial separation of middle- and upper-income blacks from lower-income blacks.52 The result was the “social isolation” of the ghetto poor from mainstream America.53
Douglas Massey and Nancy Denton picked up the thread of Wilson’s work in 1993, but focused on racial segregation as a primary causal variable.54 They describe how increasing social differentiation caused by economic dislocation interacts with the spatial concentration of a minority group to create a set of structural circumstances that reinforce the effects of social and economic deprivation. In a segregated environment, economic shocks that cause a downward shift in the distribution of minority income not only bring about an increase in the poverty rate for the group as a whole but also cause an increase in the geographic concentration of poverty. This geographic intensification of poverty occurs because the additional poverty created by macroeconomic conditions is spread unevenly over the metropolitan area.55 Thus the greater the segregation, “the smaller the number of neighborhoods absorbing the shock, and the more severe the resulting concentration of poverty.”56 Segregation by race and poverty, then, is the key causal force in Massey’s conceptual scheme. Along with The Truly Disadvantaged, American Apartheid stands out as one of the dominant pieces of scholarship at the closing of the twentieth century. Although these books posited different causal mechanisms, both were powerful statements on neighborhood effects.57
The Ideology of Community Lament
No discussion of neighborhood effects can be complete without confronting the concept of “community.” Across the entire history of intellectual thought about neighborhoods, a persistent refrain has been what Robert Nisbet aptly described in the mid-twentieth century as the “ideology of lament”—a widespread concern that something essential had been irrevocably lost by the apparent shredding of community bonds.58 As introduced in chapter 1, the notion of “community lost” has morphed over time to take on different forms but the basic message is seemingly always the same: community is in decline and must be recovered.
Today, lamentation for community lost is as strong as ever but with a twist. Calls for a return to community values and neighborhood governance have been voiced across an unusually diverse spectrum. Whether from elite politicians on the Left or Right, local government officials, communitarians, private foundations, real estate developers, or social scientists, the appeals to community have been many. For example, the move to community-based approaches has penetrated broadly in the foundation world and criminal justice policy, from well-known efforts to increase community policing to community-based prosecution policies and community corrections.59 Although it is not surprising that philanthropic foundations have embraced community as an ideal, real estate developers are also on board. Taking heed of modern discontent with suburban sprawl and anonymity, the “new urbanism” has promoted visions of architecture that promote neighborliness, local interaction, and common physical space in an attempt to restore elements of community.60 Even the World Bank has adopted concepts of community and social capital to alleviate poverty around the globe.61 The communitarian perspective also landed at both the Clinton and second Bush White Houses, and themes of community organization serve as a backdrop to the Obama administration.62
Whatever the ultimate source, once again in our intellectual history there has emerged a widespread belief that a return to “community” is needed.63 Indeed, the appeal of community has never lapsed, and the idea of a shared vision and collective approach to solving human problems is a deeply social yearning. The problem is that much of the discussion around community is normative and nostalgic rather than analytic, impeding progress on how research should proceed.64 Thus despite widespread interest across multiple disciplines, the nature, sources, and consequences of community in contemporary society remain ambiguous and largely disconnected from serious inquiry. Appropriation of social capital and communitarian ideals, for example, tends to romanticize the idea of community rather than pose hard questions subject to empirical scrutiny. If community has come to mean everything good, as a concept it loses its analytical bite and therefore means little.
We must also be careful to incorporate the possibility of a darker side to communal life and reflect on empirical evidence that a generation of community-building efforts came up empty-handed.65 In particular, what do we stand to lose by a return to community and the idea of “community organizing” and control—what does such a communal life potentially deny?66 As Thomas Sugrue’s research on postwar Detroit has taught us, neighborhood associations, while often a force for good, were nonetheless exploited by whites to forcibly keep blacks from moving into white working-class areas by resorting to arson, threats, and violence.67 This is not what communitarians mean by neighborhood cohesion. Justice and inequality must in the end be part of discussions of the idea of community (chapter 9). Put differently, the shared content of what passes through social networks matters as much as their structural configuration—cohesion or efficacy cannot be “read off” from the simple density of networks.
Ultimately, then, the equation of neighborhood with traditional community has done the field a disservice, for with this equation neighborhoods seem to have declined by definition. As described in chapter 3, by taking an alternative approach we can escape prior traps of community ideology yet retain what is most relevant—variability in the spatial social organization of everyday life. With this move neighborhoods have not declined by definitional fiat; moreover, aspects of what we think of as community are variable over time and space and, depending on the problem, still conceptually relevant. As I will later argue, shared understandings constitute a major aspect of community that need not be about private ties and therefore can be analytically applied in the contemporary city.
Neighborhood Effects at the Turn of the Twenty-First Century
By the dawn of the twenty-first century the literature on neighborhood effects was enormous. When my colleagues and I attempted a comprehensive review circa 2000 we discovered hundreds of studies and since then hundreds more have appeared.68 It was impossible to review them all then and would be even more so now. We did, however, summarize in broad form a core set of themes that have emerged—what we termed “neighborhood facts”: first, there is considerable social inequality between neighborhoods, especially in terms of socioeconomic position and racial/ethnic segregation. Second, these factors are connected in that concentrated disadvantage often coincides with the geographic isolation of racial minority and immigrant groups. Third, a number of crime- and health-related problems tend to come bundled together at the neighborhood level and are predicted by neighborhood characteristics such as the concentration of poverty, racial isolation, single-parent families, and to a lesser extent rates of residential and housing instability. Fourth, a number of social indicators at the upper end of what many would consider progress, such as affluence, computer literacy, and elite occupational attainment, are also clustered geographically (as shown in chapter 1).69 I think this broad characterization still stands.
While these facts may seem relatively straightforward and noncon-troversial, their theoretical implications are not. The reason is that while the urban ecological tradition has yielded a treasure trove of correlates and the knowledge base of facts is considerable, the social mechanisms and dynamic processes accounting for neighborhood effects have remained largely a black box. We have known for at least a hundred years the demographic correlates of all sorts of community indicators of general wellbeing, especially the aggregated characteristics of individuals. But by focusing on correlates of outcomes at the level of community social composition—most notably poverty and fixed categories of race—prior research has tended toward a risk-factor rather than an explanatory approach. Why, to take the most notable example of the past several decades, does concentrated poverty (which is, after all, the concentration of poor people) matter? Wilson has promoted the idea of social isolation as a social process mediating the effects of concentrated poverty, while others have emphasized moral cynicism and the erosion of collective efficacy. Consistent with the original Chicago School, the larger idea is that if neighborhood effects are not merely the reflection of individual characteristics then presumably they stem from social-interactional and institutional processes that involve collective aspects of community—emergent properties, in other words.
It is from this idea that in recent decades we have witnessed another turning point in the form of a renewed commitment to uncovering the social processes and mechanisms that account for neighborhood (or concentration) effects. Social mechanisms provide theoretically plausible accounts of how neighborhoods bring about change in a given phenomenon.70 Mechanisms are by and large a theoretical claim about explanation—mechanisms can only rarely be observed or manipulated causally in an experiment. Rather, social mechanisms make up the hypothesized links in the pathway of explanation from a theoretically manipulable cause to an outcome. The goal is to develop indicators of the sets of practices, meanings, and actions that reflect hypothesized mechanisms. This commitment has led to programs of research designed to directly assess theoretically motivated social processes and cultural properties that vary at the neighborhood or extralocal level (e.g., through spatial dynamics of diffusion) and thus go beyond the simple aggregation of individual traits. My colleagues and I have referred to this shift in focus as the “process turn” in neighborhood-effects research; this book and the PHDCN-related studies as a whole stand as a renewed effort in this direction.71
There have been counterdevelopments as well, most notably a shift in some quarters to individual-level accounts of social life, highlighted by increased concerns about causal inference and what has become known as “selection bias.” In fact, much of the newer literature has redefined neighborhood effects in ways that have led to stylistic modern convention, whereby the analytic focus is the direct effect of the neighborhood on some individual behavior. While important, this leaves much of interest out of the picture. From the European epidemiologists to the Chicago School and its followers, the main unit of analysis and inference has not been the individual but rates of social behavior that varied by neighborhood-level cultural and social structure. A key intervention that shifted the unit of interest to the individual came from Christopher Jencks and Susan Mayer, who can be considered the foremost source of late twentieth-century anxiety over the dilemma posed by individuals selecting environments. In a widely cited set of critiques, they asked the question: How do we know that the neighborhood differences in any outcome of interest are the result of neighborhood factors rather than the differential selection of adolescents or their families into certain neighborhoods?72 They concluded that we did not know the answer and that the only way to address the question rigorously was to conduct controlled experiments.
As a result, recent years have witnessed a vigorous debate over the role of experiments in social research in general and neighborhood effects in particular. Many have responded to the call for experiments and more emphasis on the estimate of single causal parameters. The belief that experiments promise a superior research strategy for assessing causality has become dogma in many quarters and has led to a kind of experimental hegemony of thought. The most widely promoted efforts of this new turn hail from criminology, epidemiology, and the Moving to Opportunity (MTO) housing studies. MTO randomly allocated housing vouchers to poor families to induce movement to low-poverty neighborhoods, motivated by the idea that causal neighborhood effects on later outcomes could be estimated in an unbiased manner. But I argue in chapters 11 and 15 that the MTO experiment, while seductive, does not answer important theoretical questions that I ask, including how individuals select themselves into neighborhoods and how individual perceptions (and in turn selection itself) are influenced by neighborhood characteristics. MTO also randomized individuals rather than neighborhoods to a treatment, and the offer of a voucher does not tell us the social mechanisms that account for why neighborhoods have effects on individuals, if in fact they do. Causal explanation requires theory and concepts that organize knowledge about (typically) unobserved mechanisms that bring about the effect, a challenge that simultaneously demands a base of rigorous observational evidence.
This book therefore advances the processual and social-mechanistic turn of the late twenieth century in ways that respect causal logic but that take context seriously and steer clear of individually reductionist accounts. Experiments can tell us a lot and experimental logic— especially what James Woodward calls a “manipulability theory” of causal explanation—helps improve our thinking about observational evidence.73 But causal questions are not the only ones, and the experimental method alone is ill-equipped for the study of neighborhood-level phenomena and macrolevel social processes that unfold over long periods of time. This is not to say that individual choice or causal inference are unimportant—to the contrary, I describe in the next chapter an analytic approach to neighborhood effects that draws on the lessons of past research to engage contextual explanations at the level of both neighborhood (and higher) processes and individual selection.