Harvard Law Review Harvard Law Review Harvard Law Review

Harvard Law Review Forum

When Things Aren’t What They Seem: Context and Cognition in Appearance-Based Regulation

Responding to Adam Samaha, Regulation for the Sake of Appearance, 125 Harv. L. Rev. 1563 (2012)

I. Introduction

Appearances mean a lot in American society, more than we probably care to admit or deem appropriate. Using appearance as a justification for the government’s policy decisions is thus a provocative idea that demands scrutiny. Indeed, for the government to act on appearances alone — whether to change “reality” or not — is fraught with complexity and is problematic in its implications. Recognizing the high stakes, Professor Adam Samaha1 attempts to impose analytic and normative clarity by drawing on a mix of legal and social science scholarship. He offers a general framework for evaluating appearance-based government decisions and applies his logic to two concrete cases — campaign finance regulation and broken windows policing.

An important move in Samaha’s article is to take seriously the mechanism of the “self-fulfilling prophecy” and apply it to the appearance-reality connection.2 The “bank run” is a well-known example of a self-fulfilling prophecy that is triggered when appearances are the causal force driving a later meltdown. There are other ways to think about appearance and reality, however. Samaha uses the example of bridges to highlight the transparency problem: how reality can be insulated from appearance (a bridge can appear unsound but be safe, for example, or appear safe but lack structural integrity). Architects understand the importance of “designing in” the appearance of safety, even though believing that a bridge is safe does not make it safe. For this reason, the bridge mechanism is distinct from the bank, where appearance does drive reality. Clocks represent yet a third example, where reality “collapses into appearance” based on agreed-upon conventions. Using these conceptual distinctions, Samaha analyzes justifications for campaign finance reform (for example, even if a government is not corrupt, the appearance of corruption may present a problem) and the so-called “broken windows” revolution in urban policing. The latter has gained worldwide recognition based on the highly touted crime drop in New York City3 and the widely attributed role of broken windows theory in bringing about the decline.4

I applaud Samaha’s effort. The article is erudite and covers a vast terrain of literature that is creatively synthesized. Given the broad scope of his argument and the space constraints of my response, I aim for focus and draw out an overarching thesis that builds on Samaha’s insights and suggests how we might think of appearance-based regulation in future work. As an urban sociologist and student of crime, I do not have anything specific to say about campaign finance regulation, interesting though it may be. But I have studied “disorder” for many years, and disorder-based policing is arguably more problematic for policy because of how it bears on larger notions of institutional legitimacy, social order, race, and ultimately, the viability of a theory that seeks to tame urban anxiety. I thus train my sights on disorder in reframing broken windows theory and the potential unintended consequences of how it has been appropriated by policymakers. My thesis is twofold: (1) Samaha undertheorizes the role of racial context and its associated meanings in how we perceive or evaluate appearances. Specifically, he does not sufficiently incorporate the implicit biases and shared understandings that lead to heterogeneous views on how much disorder exists in a given environment. (2) Coupled with underappreciated competing explanations for the crime-reduction effects of broken windows policing claimed by policymakers, Samaha in turn undervalues the costs of appearance-based policing on target communities, risking an overestimate of how much reducing disorder is “intrinsically good” on aesthetic grounds even if it does not reduce crime. Despite the clear differences between street disorder and perceptions of government corruption, moreover, I hypothesize that the mechanisms underlying how we see or evaluate reality are broadly applicable to both phenomena.

II. The Context of Seeing Disorder

The most famous theory of urban disorder was articulated by the late Professor James Q. Wilson and his colleague Professor George L. Kelling.5 Their theory of “broken windows” was an intellectual and policy sensation for good reason — cues of disorder in public are highly visible and constitute a salient marker of urban spaces. Samaha does a good job laying out the theory and reviews a number of empirical studies that assess disorder and its relationship to crime. He makes the crucial distinction between theories of policing and theories of physical or social disorder. Putting a lot of cops on the street is distinct as a causal mechanism from cleaning up graffiti — you can have one without the other. Deterrence is the traditional justification for more police, but the disorder mechanism is about shaping perceived or actual disorder. Broken windows policing conflates the two such that there is no way to tell what effect the broken windows mechanism had on the New York City crime drop separate from the policing effect of deterrence — if in fact either of these mattered in the first place. Samaha correctly observes that the question is unresolved despite grand claims to the contrary by pundits and scholars alike.

In my view, however, Samaha undertheorizes urban social context and implicit meanings in thinking about disorder. My key argument questions a core assumption of the broken windows thesis: one broken window (appearance) does not necessarily lead to another broken window and in turn future crime (reality). Rather, it depends on where the broken window exists and the larger social meanings associated with that context. In short, it is not only the “disorder” itself that matters, but also how it is perceived as a problem, a process linked to potential bias and shared meanings, yet taken as largely unproblematic by most proponents of broken windows theory.6 I thus argue that Samaha’s framework can be usefully retooled.

Consider first an ironic fact. In the seminal experiment by Professor Philip Zimbardo in the 1960s that Wilson and Kelling cite in their original article, the abandoned car in the Bronx was ravaged within minutes by vandals. But in Palo Alto nothing happened for five days until the researchers themselves started smashing the car’s windows. Zimbardo even reports that in Palo Alto “one day when it began to rain a passerby lowered the hood of our abandoned car — God forbid the motor should get wet!”7 So even back then, context was everything — the same disorder, primed by design to be identical, nonetheless triggered vastly different responses such that we cannot divorce perceptions from their context and social meanings. The Bronx at the time was characterized by “anonymity of place,” with shared expectations of a kind of “no man’s land” that differed radically from Palo Alto’s more watchful eyes. If we are going regulate appearance, then, it seems important to step back and assess the process of perceiving or “seeing” this or that as disorder in the first place. This is a more heterogeneous process than Samaha seems to allow, with observers’ imputing meaning depending on the settings and interactions they are observing.

In a recent book based on over a decade of research in Chicago,8 I set out a theory of how cognitive perceptions are contextually shaped by cues that go well beyond the material aspects of the environment. Samaha cited an earlier paper from this project, where we reported comparisons of residents’ perceptions of things like graffiti, garbage, and people drinking in public with systematic observations based on videotapes and observer logs carried out in the same neighborhoods.9 Several surprising findings emerged with which Samaha did not engage, but which I think bear on his appearance-based framework. First, among those living in the same neighborhood, people differed significantly in how much they viewed disorder as a problem.10 Second, perceptions were systematically shaped by social position in society, such as education, class, age, and race. Race is particularly salient — whites see disorder as more of a problem than blacks, Latinos, and Asians, even when living in the same environment.11

Third and perhaps most important, I showed that immigrant concentration and the presence of a large population of African Americans lead all racial/ethnic groups to view disorder as problematic — regardless of how much observed or measured disorder there is in the neighborhood (which we meticulously assessed by videotaped observations and based on inter-rater agreement).12 This is the pernicious problem of implicit bias’s13 interacting with context, turning the tables on the common view that seeing is simply a function of what is there, optically speaking, within error and barring perceptual deficit. Just like memory is dependent on context, so too is what we think we “see.” In American society, concentrated populations of African Americans and immigrants, especially when poor, reflect highly salient and visible contexts that interact with similarly salient racialized beliefs.

My results hardly mean that concrete cues do not matter or are purely subjective. Neighborhoods with high concentrations of minority, immigrant, and poor residents have historically been hard hit by problems of crime and disorder. These persistent and very public processes are thus not mirages — they have deep roots in the concrete history of American stratification and segregation, and are not likely to lapse in the face of short-term change or contrary evidence. I argue that perceptions take on a new life and persist when reinforced through social interactions, institutional practices, and collective reputations. In particular, I show that shared perceptions of disorder — again, independent of observable visible cues (“reality”) — form a meaningful social property of the environment that influences both individual and neighborhood-level outcomes, including the future poverty level of a neighborhood (controlling for its past poverty and present actual disorder). I also show that an individual’s perception of disorder is most influenced not by the present level of observed disorder, his or her own characteristics, poverty, or even the racial composition of the neighborhood. What matters most is the intersubjective or shared prior beliefs of different residents formed years earlier — in other words, shared priors.

These findings suggest that reputations are sticky and that the categorization of neighborhoods as “bad” or “disorderly” (or conversely, “good”) carries weight that is transmitted over time through institutionalized practices. The police are at the forefront of dividing up the city into easily understood categories shaped by race and class, and their own visible presence in the community can actively reinforce the priors of residents and further cement a neighborhood’s reputation as disorderly, potentially leading to further decline.14 This kind of self-fulfilling prophecy of stigmatization reflects what I call the “looking-glass neighborhood.”15 The social-psychological concept of the “looking-glass self” posits that a person’s individual self grows out of society’s interpersonal interactions and the perceptions of others. Moving up a level, I argue that the looking-glass neighborhood is one of the social mechanisms that organize the contemporary city and by implication its “appearance order.” Shared perceptions of disorder thus matter for reasons that extend far beyond the presence of physical cues in the environment: disorder is about much more than crime. It follows that acting on appearance will have heterogeneous effects that are shaped by the social context and collective meanings.

III. Social Transformation and the Missing Immigrant Story

Samaha argues that the popular attribution of New York City’s crime drop to broken windows policing is without strong scientific evidence.16 I agree but would go further and suggest that he take a deeper historical and comparative view (another kind of context) in his evaluative framework. No one really knows what caused the crime drop in New York City, just like no one knows what caused crime to drop around the country, in Canada, and in many places of the world that have never implemented New York City—style policing. Protagonists in the crime drop debate seem to gloss over longer-term macrostructural change. New York City, like other cities, has long-term cycles of crime that are independent of any particular police chief or government policy. Crime started to drop in New York City in 1990, before broken windows policing, and as far as I can tell the trajectory was not significantly altered by later policy changes.

Moreover, there is a clear counterhypothesis that policy debates largely neglect and that bears on my argument. In the last few decades, New York City, Los Angeles, Chicago, Boston, and many U.S. cities have been transformed socially, demographically, and physically (for example, through real estate development). Quite simply, New York City is not the same city as it was in the 1970s when it comes to the fundamental makeup of its population.17 Many scholars, including myself, have argued that massive foreign immigration is a contributor to the decline in crime and to the reshaping of the American urban landscape.18 Yet paradoxically, as I noted above, the concentration of immigration is associated with perceptions of increased disorder, resulting in the appearance-based scenario of hyperpolicing what may be a crime-reducing mechanism.19 The important point is that many participants in the crime decline and broken windows debate have operated in an ahistorical bubble and over-attribute causality to a single appearance-based regulation that is not only internally unclear (cops versus cleaning up disorder) but that also brackets the complex social reordering of the city from macrosocial forces like immigration that began before disorder policing and that are falling victim to appearance-based regulation in counterproductive ways.

IV. Aesthetics and the Unintended Consequences of Regulating Appearance

At the end of his article, Samaha pushes his framework to a normative and aesthetic level. He specifically notes that one can make a “clock-based” case for broken windows policing that is “no deeper” than aesthetics.20 Even if crime is unaffected, he reasons, if the neighborhood is cleaned up according to the tastes of “the mainstream, majority population,” the improved “optics” alone can be counted as a collective good, especially since “most people” find disorder discomforting.21 Samaha’s approach merges empirical and normative principles along the lines of what Professor David Thacher has called the “normative case study.”22 Using Jane Jacobs’s23 classic work as an exemplar, Thacher argued that bringing philosophical and normative considerations of desired behavior into the picture improves our empirical explanation of neighborhood-level processes. For example, he argues that Jacobs’s descriptions of city life:

[H]elped to clarify the things that are intrinsically good or bad for humans — features of city neighborhoods that are attractive not just because of the consequences they have for values we already understand clearly (as in the case of street eyes) but also because they reflect values we previously perceived only dimly if at all (as in the case of the vitality of cities and opportunities for contact they offer).24

In a similar vein, Samaha argues in regard to broken windows policing that reducing disorder can be evaluated on aesthetic grounds (or as an intrinsic good). He states: “The policy and its success are defined by what people perceive. What you see is what you get.”25 But once again, what we see is not necessarily what we get. Ugliness more than beauty, we might say, is “in the eye of the beholder,” rendering problematic the assertion that “[p]eople will perceive largely the same thing,”26 especially when it comes to aspects of urban public life where meanings are contested and interact with the criminal justice system. If race and class are embedded in our historical understandings of place and disorder, they will inevitably shape not only our perceptions, but also the policy responses that follow. In the looking glass, the police are part of society and thus inextricably linked to the enforcement of broken windows policy along with the shared expectations and implicit biases we all share as citizens. Therein lies the rub.

Returning to the seedbed of broken windows policing, there can be little doubt that changes in police practices in New York City have had profound consequences for disadvantaged groups, which Samaha hints at but does not dwell on in his analysis.27 As Professor Jeffrey Fagan and colleagues have shown, street stops of African American citizens grew dramatically in New York City and remain far in excess of what can be explained by crime.28 The distance from the policing of disorder to the disproportionate policing of minorities and minority (or immigrant) neighborhoods — even with no racial animus or ethnic prejudice on the part of individual officers — is short and theoretically predicted by the theory I have sketched. While it may be true that broken windows policing makes the majority of New Yorkers feel safer, those benefits need to be weighed against costs to the social fabric of targeted communities and populations.

A potentially large cost is the moral and legal cynicism that is corrosive in many highly disadvantaged, minority, and heavily policed neighborhoods, which in turn undermines the legitimacy of legal institutions.29 In an effort to gauge social norms about moral rules and the experiential relevance of law in the everyday lives of residents, we asked residents of Chicago to report their trust in police and agreement with statements such as “Laws were made to be broken” and “It’s okay to do anything you want as long as you don’t hurt anyone.” In communities with high levels of intersubjectively shared cynicism of police misbehavior and the perceived irrelevance of legal rules, violence is higher.30 This association is reinforced by disadvantage, suggesting a feedback loop whereby concentrated disadvantage is corrosive with respect to other-regarding norms. This process thus feeds cynicism, which reinforces behavior that is outside of the law (for example, using violence to settle disputes), in turn further undermining trust and notions of the just community. For this reason legal cynicism helps explain the persistence of violence in certain communities and racial disparities in violence.

The late Professor William Stuntz cogently summarizes the horns of the appearance-reality dilemma for the police. He argues that broken windows policing can have enormous social benefits: “[T]he perception that the streets are safe could lead to greater law-abiding street traffic, which in turn would lead to the reality of safer streets.”31 That mechanism is important and is properly accounted for by Samaha. I would add that crime reductions redound disproportionately to minority communities. But Stuntz goes on to say:

Or such tactics may send the signal that young men of the wrong race or ethnicity are automatic targets for the police, and hence that the police are a hostile presence in the community. That signal could have large social costs: If the police and, through them, the criminal justice system, come to be seen as illegitimate, the norms of law-abiding behavior could unravel, with the streets becoming less safe, not more so.32

Despite this dilemma, Samaha cites the popularity of broken windows policing with most of the public and suggests the policy has been oversold by proponents.33 This is an important point. Perhaps the appearance of appearance regulation has led the public to enter into a kind of Faustian bargain with the police — accepting an intrusion into public spaces and personal lives they would not otherwise tolerate and to which a significant portion of the citizenry, namely minorities, bear the predominant costs. This bargain may not hold if the appearance-based theory of broken windows is dethroned as the canonical account of the city’s crime decline.

V. Conclusion

Cognitive processes and perceptions of appearance have traditionally been overlooked in urban sociology, where structuralism rules, and apparently in legal scholarship.34 Samaha has done the latter field a great service by interrogating the justifications of appearance-based government regulation, and his analysis can and should inform social scientists as well. While my analysis suggests that Samaha is on to something fundamental, I would like to see him go even further and interrogate what many in the scholarly and policy world at present consider as largely “untroubling,” namely the very project of government, and especially the police, trying to “cure urban anxiety.” The inescapable historical reality is that in its present-day form, broken windows policing, disorder, space, and race (and most recently, immigration) are bound up together, a synergistic interaction that has produced arguably negative, even if unintended, social consequences. As Professor Richard Sennett argued over forty years ago, urban life is inherently anxiety-producing, such that disorder is not so much a thing to be eradicated but to be understood and negotiated.35 Whether one agrees with Sennett or not, our perceptions of disorder and the consequences of acting on it are fundamentally social in nature rather than fixed in meaning.

Legal theorists of regulating appearance like Samaha would therefore profit, in my view, from considering such arguments and the larger social structure of how cities are organized and ultimately understood in culturally and historically structured terms rather than just optical terms. I suspect that while a different set of manifest mechanisms are involved, the appearance of corruption and the entire area of campaign finance reform would also benefit from a deeper consideration of social meaning and the possibility that our very perceptions of what is malfeasance are inextricably linked to prior shared beliefs, structured relationships, and implicit biases. Samaha has provided us a needed set of conceptual tools for excavating the appearance-reality connection, and I look forward to his next incursion into these meaty matters.


1. Adam M. Samaha, Regulation for the Sake of Appearance, 125 Harv. L. Rev. 1563 (2012).

2. Id. at 1578. Professor Robert K. Merton is usually identified most closely with the concept of the self-fulfilling prophecy, which he wrote about in Social Theory and Social Structure 475—90 (1968). But the idea was given earlier form in the “Thomas Theorem,” which holds that if men define situations as real, they are real in their consequences. See William I. Thomas & Dorothy Swaine Thomas, The Child in America: Behavior Problems and Programs 572 (1928).

3. See, e.g., Franklin E. Zimring, The City That Became Safe (2011); see also Jeffrey Fagan, Franklin E. Zimring & June Kim, Declining Homicide in New York City: A Tale of Two Trends, 88 J. Crim. L. & Criminology 1277, 1277 (1998).

4. Samaha, supra note 1, at 1630. For background on broken windows theory, see generally George L. Kelling & Catherine M. Coles, Fixing Broken Windows (1996).

5. James Q. Wilson & George L. Kelling, Broken Windows: The Police and Neighborhood Safety, Atlantic Monthly, Mar. 1982, at 29, 29—38.

6. Although pursuing a different line of argument, there is legal scholarship that takes seriously social meaning and norms with respect to policing. See, e.g., Tracey L. Meares & Dan M. Kahan, Law and (Norms of) Order in the Inner City, 32 L. & Soc’y Rev. 805, 805—32 (1998). On social meaning and law, see generally Dan M. Kahan, Social Influence, Social Meaning, and Deterrence, 83 Va. L. Rev. 349 (1997), Lawrence Lessig, The Regulation of Social Meaning, 62 U. Chi. L. Rev. 943 (1995), and Bernard E. Harcourt, Language of the Gun (2006).

7. Philip Zimbardo, Anonymity of Place Stimulates Destructive Vandalism, The Lucifer Effect, http://www.lucifereffect.com/about_content_anon.htm (last visited Mar. 29, 2012).

8. Robert J. Sampson, Great American City: Chicago and the Enduring Neighborhood Effect 121—48, 355—85 (2012). Chapter 6 of the book also provides an intellectual history of the concept of urban disorder.

9. Robert J. Sampson & Stephen W. Raudenbush, Seeing Disorder: Neighborhood Stigma and the Social Construction of “Broken Windows,” 67 Soc. Psychol. Q. 319 (2004).

10. Id. at 328.

11. Id.

12. Id. at 329-30.

13. Implicit bias is not conscious prejudice. In one example of implicit bias that works beneath the conscious radar screen, participants instructed to act as police officers in an experiment were quicker to correctly “shoot” armed offenders if they were black compared to white. Shared expectations that blacks are more likely to be violent are implicated in explaining this finding. See Joshua Correll et al., The Police Officer’s Dilemma: Using Ethnicity to Disambiguate Potentially Threatening Individuals, 83 J. Personality & Soc. Psychol. 1314, 1321—28 (2002).

14. Early police researchers called this “ecological contamination,” whereby all persons encountered in a “bad” neighborhood are viewed as possessing the moral liability of the neighborhood itself. Sampson, supra note 8, at 133.

15. Id. at 365.

16. Samaha, supra note 1, at 1630—31; see also Bernard E. Harcourt, Illusion of Order: The False Promise of Broken Windows Policing (2001) and Zimring, supra note 3 (questioning whether policing of disorder itself has anything to do with the crime decline in New York as compared to changes in arrest practices, as arresting for misdemeanor crimes is not a policy designed to reduce physical signs of disorder).

17. See generally Richard Harris, The Geography of Employment and Residence in New York Since 1950, in Dual City: Restructuring New York 129 (John H. Mollenkopf & Manuel Castells eds., 1991); Philip Kasinitz et al., Inheriting the City: The Children of Immigrants Come of Age (2008).

18. John MacDonald & Robert J. Sampson, The World in a City: Immigration and America’s Changing Social Fabric, 641 Annals Am. Acad. Pol. & Soc. Sci. 6, 14—15 (2012); see also Robert J. Sampson, Rethinking Crime and Immigration, Contexts, Winter 2008, at 28, 33.

19. For an example of how immigration enforcement inevitably targets persons whose appearance signals Mexican descent, and the circumstances under which it might be constitutionally permissible, see United States v. Martinez-Fuerte, 428 U.S. 543, 563 (1976). See also Jeffrey Fagan & Tracey Meares, The Arizona Solution, Slate.com (May 24, 2010), http://www.slate.com/articles/news_and_politics/jurisprudence/2010/05/the_arizona_solution.html. I thank Professor Jeffrey Fagan for bringing these sources to my attention and for providing comments.

20. Samaha, supra note 1, at 1632.

21. Id.

22. David Thacher, The Normative Case Study, 111 Am. J. Soc. 1631 (2006).

23. Jane Jacobs, The Death and Life of Great American Cities (1961).

24. Thacher, supra note 22, at 1641.

25. Samaha, supra note 1, at 1633.

26. Id.

27. See id. at 1634.

28. Jeffrey A. Fagan et al., Street Stops and Broken Windows Revisited: The Demography and Logic of Proactive Policing in a Safe and Changing City, in Race, Ethnicity, and Policing: New and Essential Readings 309 (Stephen K. Rice & Michael D. White eds., 2010); Andrew Gelman, Jeffrey Fagan & Alex Kiss, An Analysis of the New York City Police Department’s “Stop-and-Frisk” Policy in the Context of Claims of Racial Bias, 102 J. Am. Stat. Ass’n 479, 479—89 (2007). Street stops in New York City now exceed half a million a year, four-fifths involving minority men. See NYCLU Analysis Reveals NYPD Street Stops Soar 600% over Course of Bloomberg Administration, NYCLU (Feb. 14, 2012), http://www.nyclu.org/news/nyclu-analysis-reveals-nypd-street-stops-soar-600-over-course-of-bloomberg-administration.

29. On the crucial link between policing and legitimacy, see David J. Smith, The Foundations of Legitimacy, in Legitimacy and Criminal Justice: International Perspectives 30 (Tom R. Tyler ed., 2008).

30. Sampson, supra note 8; David S. Kirk & Andrew V. Papachristos, Cultural Mechanisms and the Persistence of Neighborhood Violence, 116 Am. J. Soc. 1190, 1190—1233 (2011).

31. William J. Stuntz, Terry’s Impossibility, 72 St. John’s L. Rev. 1213, 1217 (1998).

32. Id.

33. See Samaha, supra note 1, at 1631.

34. But see supra note 6.

35. See generally Richard Sennett, The Uses of Disorder: Personal Identity and City Life (1970).