Social Disorder and Crime examines the behavioral dimension of incivility within broken windows theory, encompassing public drinking, aggressive panhandling, street prostitution, open drug dealing, and loitering by unsupervised groups, and its theorized and empirically documented relationship to neighborhood crime rates. Unlike physical disorder, which concerns the observable condition of the built environment, social disorder concerns ongoing or repeated behavior, a distinction that carries significant implications for measurement, for the psychological mechanisms through which it is theorized to affect residents and offenders, and for the policing strategies it has historically justified. This article traces the measurement of social disorder from Wilson and Kelling’s original narrative account through survey-based and observational methodologies, and evaluates the evidence connecting social disorder independently to fear of crime, resident withdrawal, and subsequent criminal offending.
Social Disorder and Crime engages a body of evidence that diverges in important respects from the parallel literature on physical disorder, treated in Physical Disorder and Crime. Because social disorder involves observable human behavior rather than static environmental conditions, its measurement has depended more heavily on either direct behavioral observation, which requires researchers or trained observers to be present when the disorder occurs, or resident survey reports, which are vulnerable to the perceptual biases and potential circularity concerns discussed in Core Principles of Broken Windows Theory. This methodological difference has shaped which questions researchers have been able to answer with confidence and which remain more speculative.
This article should also be read alongside Broken Windows Theory and Policing and Order Maintenance Policing, since social disorder, more than physical disorder, has historically been the primary target of order-maintenance and quality-of-life policing strategies, and alongside Broken Windows Theory and Race, since social disorder enforcement has generated the theory’s most sustained and consequential civil rights controversies.
Defining and Categorizing Social Disorder
Behavioral Categories and Their Variation
The empirical literature has generally organized social disorder into several recurring behavioral categories: substance-related disorder, including public drinking, public intoxication, and visible drug use or sale; solicitation-related disorder, including aggressive panhandling and street prostitution; and congregation-related disorder, including loitering by unsupervised youth, unruly groups on street corners, and public disturbances such as loud arguments (Skogan, 1990; Sampson & Raudenbush, 1999). These categories vary considerably in the degree to which they involve an identifiable, potentially criminal act, public drug sale is itself a crime, whereas loitering is frequently lawful, a distinction with direct bearing on the legal basis available for police intervention and one examined further in Order Maintenance Policing.
Skogan (1990) further distinguished disorder that residents perceive as personally threatening, such as aggressive panhandling or public intoxication near one’s residence, from disorder perceived as merely unpleasant or aesthetically objectionable, such as loud but non-threatening groups of teenagers, finding that the threat-perception dimension carried greater explanatory weight for residents’ fear of crime than disorder’s mere presence or frequency. This finding suggests that social disorder’s signaling function may depend as much on residents’ subjective interpretation of threat as on any objective count of disorderly incidents.
This threat-perception dimension also appears to vary systematically by resident characteristics, with several studies finding that older residents, women, and residents with prior victimization experience report stronger threat perceptions in response to identical observed disorder than younger, male, or previously unvictimized residents (Ferraro, 1995). This heterogeneity in threat perception across resident subgroups complicates efforts to establish a single, generalizable disorder-fear relationship applicable uniformly across an entire neighborhood’s population, since the same corner gathering of unsupervised youths may register as alarming to one resident and unremarkable to another.
Measurement Challenges Distinct from Physical Disorder
Because social disorder is transient and behavioral rather than static, systematic social observation methodologies adapted from physical-disorder research face a distinct challenge: an observer passing through a block at a single point in time may miss social disorder that occurs at other hours, particularly late-night drinking, drug activity, or solicitation that clusters around specific times of day (Sampson & Raudenbush, 1999). Researchers have addressed this limitation through repeated-observation protocols, video-based continuous monitoring in select studies, and triangulation with resident survey reports and police-recorded disorder-related calls for service, though each method carries its own limitations, calls for service undercount disorder residents choose not to report, while survey reports may reflect residents’ broader anxieties rather than disorder specifically observed.
A further complication specific to social disorder measurement concerns the role of observer presence itself in altering the behavior being measured, a reactivity problem less pronounced for physical disorder, which does not respond to being observed. Researchers conducting in-person social disorder observation have noted that visibly marked research vehicles or uniformed observers can temporarily suppress the very behaviors, public drinking or open solicitation, that the observation protocol seeks to measure, potentially producing systematic undercounts in exactly the locations where social disorder is otherwise most prevalent. This reactivity concern has motivated growing interest in unobtrusive measurement alternatives, including analysis of anonymized calls-for-service data and, more recently, aggregated and de-identified location data, though each alternative introduces its own distinct measurement biases requiring careful methodological accounting.
Empirical Evidence on Social Disorder and Crime
Associations with Fear of Crime
The strongest and most consistently replicated finding in the social disorder literature concerns its association with fear of crime rather than with crime victimization itself. Skogan’s (1990) multi-city study found that resident-reported social disorder predicted heightened fear of crime independent of the neighborhood’s actual recorded crime rate, a finding replicated in subsequent research examining the “disorder-fear” relationship across varied urban contexts (Ferraro, 1995; Hinkle & Weisburd, 2008). This consistent disorder-fear association has proven considerably more consistent across studies than the more contested disorder-crime association itself, suggesting that social disorder’s clearest and best-supported effect operates through residents’ subjective sense of safety rather than through any direct causal pathway to subsequent offending.
Ferraro’s (1995) research further distinguished generalized fear of crime from more specific “constrained behavior,” the concrete adjustments residents make in response to perceived social disorder, such as avoiding particular streets after dark or curtailing children’s outdoor activity, finding that social disorder predicted both outcomes but that the behavioral constraint measure showed a stronger and more consistent relationship to disorder than generalized fear ratings alone. This distinction matters for the theory’s broader causal chain, since it is specifically the behavioral constraint response, residents’ literal withdrawal from public space, rather than fear as a subjective emotional state per se, that the theory’s core mechanism requires in order to translate perceived disorder into reduced informal social control.
Associations with Subsequent Crime
Evidence connecting social disorder independently to subsequent serious crime, controlling for structural neighborhood characteristics, has proven considerably more mixed than the disorder-fear association. Sampson and Raudenbush’s (1999) Chicago study, which included social as well as physical disorder in its systematic observation protocol, found that social disorder’s association with subsequent violent crime, like physical disorder’s, was substantially attenuated once collective efficacy and structural characteristics such as concentrated disadvantage were controlled, a finding that has informed the broader critique addressed in Criticisms of Broken Windows Theory.
Other researchers using different methodologies have reported more supportive findings. Hinkle and Weisburd’s (2008) study of Jersey City hot spots found that order-maintenance policing targeting social disorder was associated with reductions in serious crime at treated locations relative to comparison locations, though the authors cautioned that the mechanism might reflect general deterrence from increased police presence at disorder hot spots rather than disorder reduction per se, a distinction with significant implications for interpreting order-maintenance policing’s apparent effectiveness, discussed further in Broken Windows Theory and Policing. Kelling and Sousa (2001) similarly attributed part of New York City’s 1990s crime decline to sustained social disorder enforcement, though this city-level, pre-post attribution has faced the same confounding concerns raised against comparable claims for physical disorder, discussed in Physical Disorder and Crime.
Consistent with the moderation findings discussed in Core Principles of Broken Windows Theory, Markowitz and colleagues (2001) found that social disorder’s association with fear of crime was substantially stronger in neighborhoods with lower baseline social cohesion, indicating that social disorder’s effects, like physical disorder’s, are contingent on pre-existing neighborhood relational structure rather than operating as a uniform, context-independent signal.
The Confound of Shared Structural Causes
As with physical disorder, social disorder and serious crime frequently co-occur in neighborhoods characterized by concentrated poverty, residential instability, and reduced collective efficacy, raising the possibility that social disorder and crime are jointly produced by these shared structural conditions rather than standing in a direct causal relationship to one another (Sampson & Raudenbush, 1999; Harcourt, 2001). This confound problem is particularly acute for social disorder because behaviors such as open drug dealing may themselves be more accurately understood as low-level criminal activity embedded within the same underground economies that produce more serious offending, rather than as a categorically distinct “disorder” that precedes and causes crime, a conceptual ambiguity Harcourt (2001) identified as a fundamental weakness in the theory’s original disorder-crime distinction.
This conceptual ambiguity has practical consequences for how order-maintenance policing programs are evaluated: crackdowns targeting open-air drug markets, classified by their proponents as disorder enforcement, are difficult to distinguish empirically from conventional drug-crime suppression efforts that predate and operate independently of broken windows theory, meaning apparent crime reductions following such crackdowns cannot straightforwardly be attributed to the disorder-reduction mechanism the theory specifies rather than to more conventional deterrent or incapacitative effects of drug enforcement itself. Braga and Weisburd’s (2010) review of hot-spots policing evaluations noted that many of the most methodologically rigorous studies showing crime reductions from disorder-focused policing involved interventions, such as problem-oriented crackdowns on specific drug locations, that combined elements of both disorder enforcement and conventional crime suppression in ways the underlying evaluations could not fully disentangle.
Policing and Regulatory Responses to Social Disorder
Order-Maintenance Enforcement Strategies
Social disorder, more than physical disorder, has historically been the primary object of order-maintenance and quality-of-life policing strategies, since addressing behavioral disorder typically requires direct police contact with individuals engaged in the disorderly conduct, whether through warnings, citations, or arrest for offenses such as public intoxication, disorderly conduct, or loitering statutes (Kelling & Coles, 1996). This enforcement approach is examined in detail in Order Maintenance Policing and Zero Tolerance Policing and Broken Windows, which distinguish more discretionary, relationship-based order-maintenance approaches from stricter zero-tolerance enforcement models.
Within order-maintenance strategies, departments have varied considerably in how they operationalize social disorder enforcement, ranging from informal, warning-based approaches in which officers exercise substantial discretion to overlook minor infractions in favor of relationship-building with residents, to formalized ticketing and arrest quotas that leave officers comparatively little discretionary latitude. Research comparing these variants has generally found that more discretionary, community-engagement-oriented approaches produce fewer of the disparate-enforcement outcomes documented in stricter zero-tolerance models, though discretionary approaches also generate less consistent, more officer-dependent enforcement patterns that complicate systematic evaluation of their crime-control effectiveness.
Civil and Regulatory Alternatives
Beyond criminal enforcement, municipalities have developed civil and regulatory alternatives targeting social disorder, including civil citations for public drinking, business improvement district ambassador programs addressing panhandling and loitering through non-police personnel, and problem-oriented policing partnerships with social service providers addressing the underlying conditions, such as untreated addiction or homelessness, that frequently underlie visible social disorder (Braga & Weisburd, 2010). These regulatory alternatives have gained traction partly in response to concerns about the discriminatory enforcement patterns associated with criminal order-maintenance policing, discussed in Broken Windows Theory and Race, and partly in response to evidence that criminal sanctions for low-level disorder offenses frequently fail to address underlying causes such as substance dependency or mental illness.
Business improvement district ambassador programs, in particular, have expanded substantially in American downtowns and commercial corridors since the 1990s, employing uniformed but unarmed personnel to address low-level disorder, provide directions, and coordinate with social service outreach, functioning as a hybrid model that addresses order-maintenance goals without relying primarily on police contact or criminal sanction (Braga & Weisburd, 2010). Evaluations of these programs have generally found reductions in resident-reported disorder complaints within program boundaries, though researchers note that such programs may simply displace disorder to adjacent, non-covered blocks rather than eliminating it, a displacement concern that parallels similar critiques leveled against criminal order-maintenance policing itself.
Social Disorder, Race, and Discretionary Enforcement
Disparate Enforcement Patterns in Order-Maintenance Policing
Because social disorder enforcement depends heavily on officer discretion in deciding which observed behaviors warrant intervention, empirical research has consistently documented racially disparate enforcement patterns in cities adopting order-maintenance strategies. Fagan and Davies’s (2000) analysis of New York City stop-and-frisk data found that stops for low-level disorder offenses were concentrated in predominantly Black and Latino precincts at rates substantially exceeding what those precincts’ recorded crime rates alone would predict, suggesting that neighborhood racial composition, independent of crime or disorder levels, influenced enforcement intensity. This pattern has been replicated in several other jurisdictions adopting similar strategies, reinforcing concern that social disorder’s inherently discretionary, behavior-based definition makes it a uniquely vulnerable target for racially disparate policing relative to crimes with more objectively verifiable elements.
Beckett and Herbert’s (2010) ethnographic and quantitative research on Seattle’s civility ordinances similarly found that social disorder enforcement disproportionately targeted homeless individuals and people of color, describing a broader pattern of “banishment,” in which municipalities used civil and criminal disorder statutes to exclude marginalized populations from commercial and residential areas rather than to address the underlying conditions generating the observed disorder. These findings, examined at greater length in Broken Windows Theory and Race, complicate any straightforward policy endorsement of social disorder enforcement, since the practical implementation of such enforcement has proven difficult to disentangle from disparate treatment of already marginalized groups.
Legal Challenges to Discretionary Disorder Enforcement
The discretionary character of social disorder enforcement has generated sustained legal challenge, most prominently in Floyd v. City of New York (2013), in which a federal district court found that the New York Police Department’s stop-and-frisk practices, substantially justified by reference to order-maintenance and quality-of-life policing rationales, violated the Fourth Amendment’s prohibition on unreasonable searches and the Fourteenth Amendment’s equal protection guarantee through racially disparate application. The Floyd litigation and its aftermath, examined in detail in Stop-and-Frisk and Broken Windows Policing, established significant legal constraints on how aggressively municipalities can operationalize social disorder enforcement, requiring documented, individualized reasonable suspicion rather than generalized disorder-reduction rationales for street-level stops.
Subsequent litigation in other jurisdictions has extended similar Fourth Amendment scrutiny to municipal loitering and civility ordinances, with several courts finding that vague statutory definitions of disorderly conduct or loitering afford police excessive discretion inconsistent with due process requirements (Natapoff, 2018). This body of case law has pushed several police departments toward more narrowly defined, behavior-specific disorder statutes and away from broad discretionary loitering ordinances of the kind common in the early order-maintenance policing era.
Social Disorder and Vulnerable Populations
Homelessness, Mental Illness, and the Criminalization of Disorder
A substantial share of behavior classified as social disorder, including public intoxication, loitering, and public urination, is disproportionately associated with unsheltered homelessness and untreated mental illness rather than with predatory criminal intent, raising concerns that order-maintenance enforcement effectively criminalizes poverty and disability rather than addressing genuine threats to public safety (Herring, 2019). Natapoff’s (2018) research on America’s misdemeanor system found that low-level disorder offenses disproportionately generate court debt, warrants, and collateral consequences, including employment and housing barriers, for economically vulnerable defendants, consequences that persist regardless of whether the underlying disorderly conduct posed any genuine risk to public safety.
Herring’s (2019) study of San Francisco’s complaint-driven homelessness enforcement found that resident 311 complaints regarding social disorder disproportionately targeted visibly homeless individuals in gentrifying neighborhoods, illustrating how social disorder enforcement can function as a mechanism of neighborhood-level exclusion tied to broader processes of urban change, a dynamic examined further in Broken Windows Theory and Gentrification.
Alternative Response Models
In response to these criminalization concerns, numerous municipalities have piloted non-police alternative response models for social disorder involving homelessness or behavioral health crises, dispatching trained civilian crisis responders or outreach workers rather than police officers to address public intoxication, mental health crises, and related disorder (Herring, 2019). Early evaluations of these alternative models have generally found reductions in arrest rates for low-level disorder offenses without corresponding increases in recorded crime, though researchers caution that rigorous, randomized evaluation of these programs remains limited relative to the scale of their recent municipal adoption, an evidentiary gap examined further in Alternatives to Broken Windows Policing.
The scale of alternative-response adoption has grown considerably since the early pilot programs of the 2010s, with several major American cities now operating dedicated non-police crisis response units covering some portion of citywide social disorder calls. Comparative evaluation across these programs remains complicated by significant variation in program design, including differences in dispatch criteria, responder training, and the availability of follow-up services such as shelter placement or substance treatment referral, making it difficult to attribute observed arrest reductions cleanly to any single program feature rather than to the broader shift away from police-first response itself.
Analytical Tables
Table 1. Social Disorder Categories and Their Empirical Associations
| Disorder Category | Example Behaviors | Association with Fear of Crime | Association with Subsequent Serious Crime | Typical Legal Basis for Intervention |
|---|---|---|---|---|
| Substance-related | Public drinking, intoxication | Strong, well-replicated | Mixed, attenuated by structural controls | Public intoxication statutes |
| Drug-related | Open-air drug sale or use | Strong | Moderate, confounded with underlying crime | Controlled substance statutes |
| Solicitation-related | Aggressive panhandling, prostitution | Strong | Weak, poorly isolated | Municipal ordinances, solicitation statutes |
| Congregation-related | Unsupervised youth groups, loitering | Moderate | Weak | Loitering ordinances, curfew laws |
| Disturbance-related | Loud disputes, public disturbances | Moderate | Weak | Disorderly conduct statutes |
Table 2. Comparing Social and Physical Disorder Along Key Theoretical Dimensions
| Dimension | Social Disorder | Physical Disorder |
|---|---|---|
| Measurement approach | Behavioral observation, resident survey | Systematic environmental observation |
| Temporal stability | Transient, time-of-day dependent | Relatively stable over time |
| Primary responsible actor | Individual engaging in behavior | Property owner or municipality |
| Strongest documented effect | Heightened fear of crime | Attenuated once structural controls applied |
| Primary intervention type | Police contact, civil citation, service referral | Code enforcement, physical remediation |
| Discriminatory enforcement risk | High | Low |
Conclusion
Social disorder occupies a theoretically central but empirically contested position within broken windows theory. The evidence most consistently supports social disorder’s association with residents’ fear of crime, a relationship replicated across multiple cities and methodologies, while evidence for social disorder’s independent causal contribution to subsequent serious crime, once structural neighborhood conditions are held constant, remains considerably weaker and more context-dependent than the theory’s original formulation implied. This pattern parallels, and in some respects reinforces, the analogous findings for physical disorder examined elsewhere in this silo.
Because social disorder enforcement necessarily involves direct police contact with individuals, its policy implications carry stakes that purely environmental interventions targeting physical disorder do not, implicating questions of discretion, discrimination, and civil liberties that the remaining articles in this collection, particularly Broken Windows Theory and Race and Stop-and-Frisk and Broken Windows Policing, examine in detail. The gap between social disorder’s well-supported association with fear and its more contested relationship to serious crime remains a central point of tension in contemporary evaluations of order-maintenance policing strategy.
Related Articles
- Physical Disorder and Crime
- Fear of Crime and the Incivilities Thesis
- Order Maintenance Policing
- Broken Windows Theory and Race
- Zero Tolerance Policing and Broken Windows
References
- Beckett, K., & Herbert, S. (2010). Banished: The New Social Control in Urban America. Oxford University Press.
- Braga, A. A., & Weisburd, D. L. (2010). Policing Problem Places: Crime Hot Spots and Effective Prevention. Oxford University Press.
- Fagan, J., & Davies, G. (2000). Street stops and broken windows: Terry, race, and disorder in New York City. Fordham Urban Law Journal, 28(2), 457–504.
- Ferraro, K. F. (1995). Fear of Crime: Interpreting Victimization Risk. State University of New York Press.
- Floyd v. City of New York, 959 F. Supp. 2d 540 (S.D.N.Y. 2013).
- Harcourt, B. E. (2001). Illusion of Order: The False Promise of Broken Windows Policing. Harvard University Press.
- Herring, C. (2019). Complaint-oriented policing: Regulating homelessness in public space. American Sociological Review, 84(5), 769–800.
- Hinkle, J. C., & Weisburd, D. (2008). The irony of broken windows policing: A micro-place study of the relationship between disorder, focused police crackdowns, and fear of crime. Journal of Criminal Justice, 36(6), 503–512.
- Kelling, G. L., & Coles, C. M. (1996). Fixing Broken Windows: Restoring Order and Reducing Crime in Our Communities. Free Press.
- Kelling, G. L., & Sousa, W. H. (2001). Do police matter? An analysis of the impact of New York City’s police reforms. Civic Report, 22, Manhattan Institute.
- Markowitz, F. E., Bellair, P. E., Liska, A. E., & Liu, J. (2001). Extending social disorganization theory: Modeling the relationships between cohesion, disorder, and fear. Criminology, 39(2), 293–320.
- Natapoff, A. (2018). Punishment Without Crime: How Our Massive Misdemeanor System Traps the Innocent and Makes America More Unequal. Basic Books.
- Sampson, R. J., & Raudenbush, S. W. (1999). Systematic social observation of public spaces: A new look at disorder in urban neighborhoods. American Journal of Sociology, 105(3), 603–651.
- Skogan, W. G. (1990). Disorder and Decline: Crime and the Spiral of Decay in American Neighborhoods. Free Press.
- Wilson, J. Q., & Kelling, G. L. (1982). Broken windows: The police and neighborhood safety. The Atlantic Monthly, 249(3), 29–38.