Fear of Crime and the Incivilities Thesis examines the psychological mechanism at the heart of broken windows theory: the proposition that visible physical and social disorder, or “incivilities,” raise residents’ fear of crime independent of actual victimization risk, and that this heightened fear, rather than disorder itself, is what drives the behavioral withdrawal from public space that the broader theory identifies as the proximate cause of neighborhood decline. This article traces the incivilities thesis from its origins in early 1980s survey research through its subsequent theoretical elaboration and empirical testing, examining the evidence distinguishing fear of crime from actual crime risk, the demographic and neighborhood-level factors that moderate fear responses to disorder, and the policy implications of treating fear reduction as a distinct target from crime reduction itself.
Fear of Crime and the Incivilities Thesis addresses a mechanism that Wilson and Kelling’s original 1982 essay invoked only implicitly through its Newark Foot Patrol Experiment evidence, discussed in Wilson and Kelling’s Broken Windows Article, in which residents reported feeling safer under foot patrol despite unchanged recorded crime rates. Subsequent researchers, particularly Wesley Skogan and Kenneth Ferraro, developed this perception-behavior gap into a formal theoretical account, the incivilities thesis, holding that disorder’s primary and most reliably documented effect operates through subjective fear rather than through the more contested direct pathway to serious crime examined in Physical Disorder and Crime and Social Disorder and Crime.
This article should be read alongside Core Principles of Broken Windows Theory, which situates fear as one intervening variable within the theory’s broader causal sequence, and Collective Efficacy Theory and Broken Windows, which examines how fear-driven withdrawal relates to the more structurally grounded collective efficacy mechanism that subsequent research has identified as the stronger predictor of serious neighborhood violence.
Origins and Development of the Incivilities Thesis
Early Formulations of the Fear-Disorder Link
The incivilities thesis emerged from a body of victimization survey research in the 1970s that first documented a puzzling empirical pattern: fear of crime among urban residents did not track closely with objective victimization risk, with some demographic groups, notably elderly residents and women, reporting disproportionately high fear despite comparatively low objective victimization rates, while young men, who faced the highest objective victimization risk, often reported comparatively lower fear (Garofalo, 1981; Skogan & Maxfield, 1981). This “fear-victimization paradox” motivated researchers to search for alternative predictors of fear beyond direct victimization experience, and visible neighborhood disorder emerged as a leading candidate.
Skogan and Maxfield’s (1981) early Chicago-area research was among the first to systematically document that resident-reported disorder, rather than recorded crime rates, was the stronger predictor of neighborhood fear, laying empirical groundwork that Wilson and Kelling’s essay, published the following year, would draw upon in developing its broader theoretical narrative. This sequencing is worth noting because it reverses a common assumption about the theory’s development: rather than the fear-disorder link being derived from broken windows theory, the empirical association substantially preceded and informed the theory’s popular articulation.
Skogan’s Systematic Elaboration
Skogan’s (1990) Disorder and Decline represented the most systematic elaboration of the incivilities thesis, formalizing a multi-stage model in which physical and social incivilities raise perceived risk of victimization, which in turn produces both psychological fear responses and behavioral avoidance, including reduced use of public space, increased home security investment, and, in more extreme cases, decisions to relocate out of the neighborhood entirely. Skogan’s model explicitly separated fear as a subjective emotional state from the concrete behavioral adaptations fear produces, arguing that both were independently important outcomes with distinct policy implications.
Skogan further proposed that the fear-disorder relationship operated largely independent of, though correlated with, the more contested disorder-to-serious-crime pathway central to Wilson and Kelling’s original causal chain, meaning that even in neighborhoods where disorder’s contribution to actual crime remained empirically uncertain, its contribution to residents’ fear and consequent quality of life could be documented and addressed on its own terms, a distinction that has proven durable in the subsequent literature and is examined further in Core Principles of Broken Windows Theory.
Distinguishing Fear of Crime from Objective Risk
The Fear-Victimization Paradox Explained
Ferraro’s (1995) influential synthesis proposed that the apparent paradox of high fear among low-risk groups, particularly elderly residents and women, could be explained by differences in perceived vulnerability rather than by irrationality or misperception of actual crime statistics. Elderly residents and women, Ferraro argued, rationally perceive themselves as less physically capable of resisting or escaping a criminal confrontation, and consequently experience heightened fear even at comparatively low objective victimization rates because the perceived severity of a potential victimization, given their vulnerability, outweighs its comparatively low probability in their risk calculus.
This vulnerability-based explanation reframes the fear-disorder relationship as operating through a similarly rational, if subjectively weighted, risk-assessment process: incivilities are theorized to function as ambiguous environmental cues that vulnerable individuals, lacking other reliable information about a specific location’s actual danger, use as a proxy signal for risk, producing disproportionate fear responses among precisely those residents least equipped to manage a worst-case outcome (Ferraro, 1995). Warr’s (1990) related research on “sensitivity to risk” similarly found that fear responses to given levels of objective risk varied systematically by perceived seriousness of the feared offense, with fear of rape, for instance, raising women’s overall fear of crime disproportionately relative to its actual statistical frequency, a “shadow of sexual assault” effect that further explains gender differences in the fear-disorder relationship beyond incivilities alone.
Taken together, the vulnerability and sensitivity-to-risk explanations imply that any given level of observed disorder will generate systematically different fear responses across demographic subgroups within the very same neighborhood, meaning that neighborhood-average fear measures used in much of the aggregate incivilities literature may mask substantial underlying heterogeneity in how different residents actually experience and respond to identical physical and social conditions. This heterogeneity has practical implications for targeted fear-reduction interventions, discussed later in this article, which may need to be tailored to specific vulnerable subpopulations rather than designed around a single, undifferentiated neighborhood-wide fear response.
Distinguishing Formless and Concrete Fear
LaGrange, Ferraro, and Supancic (1992) further distinguished “formless fear,” a diffuse, generalized sense of unease about crime not tied to any specific situation, from “concrete fear,” specific worry about particular offenses in particular contexts, finding that incivilities predicted concrete fear considerably more strongly and consistently than formless fear. This distinction matters for measurement, since survey instruments asking broadly “how safe do you feel in your neighborhood” capture formless fear, while instruments asking about specific behavioral avoidance, such as willingness to walk alone at night on a particular street, capture concrete fear more directly tied to the incivilities thesis’s specific behavioral predictions.
This measurement distinction has practical implications for evaluating disorder-reduction interventions: programs that successfully reduce concrete, location-specific fear may not produce commensurate reductions in residents’ generalized, formless anxiety about crime, since the latter is also shaped by broader factors such as media crime coverage and macro-level social trust unrelated to any single neighborhood’s physical or social conditions (LaGrange, Ferraro, & Supancic, 1992).
The formless-concrete distinction also helps explain a recurring puzzle in the survey literature: national-level surveys frequently find that most respondents report their own neighborhood as comparatively safe while simultaneously reporting that crime nationally is a serious and worsening problem, a pattern difficult to reconcile if fear were understood as a single, undifferentiated construct closely tied to direct local observation. LaGrange and colleagues’ (1992) framework resolves this apparent inconsistency by treating the neighborhood-safety judgment as substantially concrete, grounded in residents’ actual observed local conditions, while treating the national-crime-trend judgment as substantially formless, shaped instead by national media coverage, political rhetoric, and broader cultural narratives about crime and social order largely independent of any individual respondent’s directly observed environment.
Neighborhood and Demographic Moderators of Fear
Racial Composition and the Social Construction of Fear
Building on the broader disorder-perception research examined in Physical Disorder and Crime, several studies have found that fear responses to observed disorder are moderated by neighborhood racial composition in ways that raise concern about the construct’s objectivity. Sampson and Raudenbush’s (2004) research found that residents’ fear and disorder perceptions were heightened in neighborhoods with higher proportions of Black and Latino residents independent of the neighborhood’s objectively observed physical conditions, suggesting that racialized assumptions about neighborhood danger, rather than disorder per se, partly drove fear responses in cross-neighborhood comparisons.
Chiricos, Hogan, and Gertz’s (1997) related research on the “racial typification of crime” similarly found that white residents’ fear of crime was significantly heightened by proximity to Black neighbors independent of local crime rates, a finding with direct implications for interpreting cross-neighborhood fear surveys and for understanding how the incivilities thesis’s core mechanism may be entangled with, rather than cleanly separable from, racialized perception, a concern examined at greater length in Broken Windows Theory and Race.
Prior Victimization and Media Exposure
Fear of crime has also been found to be strongly shaped by factors external to any given neighborhood’s observable conditions, including residents’ own prior victimization experience and their exposure to local crime media coverage. Chiricos, Padgett, and Gertz’s (2000) research found that local television news crime coverage independently heightened viewers’ fear of crime even after controlling for actual local crime rates and personal victimization history, indicating that media consumption operates as a distinct, disorder-independent fear pathway that can either amplify or, in principle, substitute for direct exposure to visible neighborhood incivilities.
This media-fear pathway carries important implications for disorder-reduction policy, since it suggests that even successful, well-documented disorder remediation may fail to produce proportionate fear reduction if residents’ fear is substantially shaped by broader crime narratives circulating through local and national media independent of their own neighborhood’s actual, improved conditions.
The relative contribution of media exposure versus direct disorder observation to overall fear levels appears to vary by residents’ own direct neighborhood experience: Chiricos, Padgett, and Gertz (2000) found that media effects on fear were strongest among residents with the least direct personal exposure to their local crime and disorder conditions, such as those who spent little time in public space, suggesting a substitution pattern in which residents without firsthand disorder observation relied more heavily on mediated crime narratives to form their fear judgments, while residents with more direct daily exposure to their neighborhood’s actual conditions weighted that direct experience more heavily relative to media coverage.
Longitudinal and Cross-National Evidence
Tracking Fear Over Time Within Neighborhoods
Much of the foundational incivilities research relied on cross-sectional survey designs capturing disorder and fear at a single point in time, leaving open questions about whether fear responds dynamically to changing disorder conditions within the same neighborhood over time. Taylor and Hale’s (1986) early panel research addressed this gap directly, comparing competing causal models of the disorder-fear relationship using longitudinal Baltimore survey data, and finding support for a model in which disorder’s effect on fear operated both directly and indirectly through residents’ perceived risk of victimization, rather than through a single, simple causal pathway. Their work represented one of the first attempts to formally test, rather than simply assume, the specific causal ordering the incivilities thesis proposed.
Later longitudinal work by Hale (1996), synthesizing the accumulated fear-of-crime literature across two decades, concluded that while cross-sectional disorder-fear associations were consistent and well replicated, genuinely longitudinal evidence establishing that within-neighborhood changes in disorder produced corresponding within-neighborhood changes in fear over time remained comparatively sparse, a gap Hale attributed to the practical difficulty and expense of repeated, comparable disorder and fear measurement within the same neighborhoods across multiple survey waves. This evidentiary gap has persisted even as cross-sectional support for the thesis has grown, a pattern paralleling the broader measurement challenges discussed in Physical Disorder and Crime regarding disorder’s temporal relationship to serious crime.
Innes’s (2004) signal crimes framework, discussed also in Core Principles of Broken Windows Theory, offered a further refinement relevant to the longitudinal question, proposing that certain disorderly events function as particularly potent fear-inducing signals precisely because they are perceived as novel or escalating relative to a neighborhood’s established baseline, implying that the same objective level of disorder may produce different fear responses depending on whether it represents a recent change or a longstanding, already-normalized condition, a dynamic that purely cross-sectional research designs are structurally unable to capture.
Fear of Crime Outside the American Context
British criminologists developed a parallel and partly independent tradition of incivilities research examining fear of crime in English and Welsh neighborhoods, generally replicating the core disorder-fear association documented in American research while also identifying some context-specific differences in its strength and mechanisms. Jackson’s (2004) research distinguished “experiential” fear, grounded in residents’ direct observation of local conditions, from “expressive” fear, reflecting broader anxieties about social change and moral decline that residents projected onto local disorder without direct experiential grounding, proposing that a meaningful share of reported fear of crime in British surveys reflected this expressive dimension rather than a direct response to observed neighborhood conditions.
Farrall, Jackson, and Gray’s (2009) subsequent synthesis of British fear-of-crime research reinforced this expressive-fear framework, arguing that fear of crime functions partly as a vehicle through which residents express broader concerns about social cohesion, moral order, and neighborhood change, concerns that visible incivilities can trigger or symbolize without incivilities necessarily being fear’s most proximate or complete cause. This British theoretical tradition, developed largely independently of but substantially convergent with the American incivilities literature reviewed above, reinforces the broader pattern identified throughout this silo that the fear-disorder relationship, while consistent, is considerably more psychologically and socially complex than Wilson and Kelling’s original 1982 formulation suggested.
Policy Implications of Treating Fear as a Distinct Target
Fear-Reduction Strategies Independent of Crime Reduction
Because the incivilities thesis identifies fear as a partially independent outcome from actual crime risk, several jurisdictions have developed fear-reduction strategies distinct from conventional crime-control policing, including improved street lighting, sightline-clearing landscaping interventions, and visible community policing presence designed explicitly to reassure residents rather than primarily to deter or apprehend offenders (Skogan, 1990). Evaluations of such reassurance-oriented interventions have generally found measurable fear reductions even in the absence of corresponding measurable crime reductions, supporting the incivilities thesis’s core claim that fear and crime, while related, respond to at least partially distinct interventions.
The British “reassurance policing” model, developed partly in response to the signal crimes framework discussed above, represents one of the more systematically evaluated fear-reduction strategies distinct from American zero-tolerance approaches, emphasizing sustained, visible foot patrol targeted specifically at locations residents identified as producing disproportionate anxiety rather than at locations selected purely by recorded crime statistics (Innes, 2004). Evaluations of reassurance policing pilots in several English police force areas found meaningful reductions in resident-reported fear following implementation, with the strongest effects concentrated in exactly the locations residents had identified as anxiety-inducing prior to intervention, lending support to the underlying signal crimes premise that targeting specific, resident-identified fear triggers produces more efficient reassurance than broader, undifferentiated patrol strategies.
Balancing Fear Reduction Against Enforcement Costs
The recognition that fear can be addressed independently of crime reduction has also informed critiques of aggressive order-maintenance and zero-tolerance policing strategies, discussed in Zero Tolerance Policing and Broken Windows, on the grounds that such strategies incur substantial civil liberties and community-relations costs, examined in Broken Windows Theory and Race, in pursuit of fear reduction that might be achievable through less coercive means, such as environmental design improvements or community policing reassurance strategies that do not require expanded police discretion over resident behavior. This policy tension, between the incivilities thesis’s original justification for order-maintenance enforcement and less coercive alternative pathways to the same fear-reduction goal, is examined further in Alternatives to Broken Windows Policing.
A further consideration relevant to this policy balance concerns fear’s own capacity to generate secondary harms when addressed through coercive rather than reassurance-oriented means. Several researchers have noted that aggressive order-maintenance enforcement can, paradoxically, increase certain residents’ fear rather than reduce it, particularly among residents who themselves belong to demographic groups disproportionately subject to police stops and are consequently more likely to experience heightened police presence as a source of anxiety rather than reassurance (Fagan & Davies, 2000, discussed at greater length in Social Disorder and Crime). This paradoxical fear effect suggests that the relationship between policing intensity and resident reassurance is not monotonic, and that beyond some threshold, additional enforcement presence may begin to undermine rather than support the very fear-reduction goal that originally justified it, a possibility with direct relevance for calibrating order-maintenance strategies discussed throughout this silo.
Analytical Tables
Table 1. Key Studies in the Development of the Incivilities Thesis
| Study | Contribution | Key Finding |
|---|---|---|
| Garofalo (1981) | Identifies fear-victimization paradox | Fear does not track objective victimization risk |
| Skogan & Maxfield (1981) | Early Chicago fear research | Disorder predicts fear more strongly than crime rates |
| Skogan (1990) | Formalizes incivilities thesis | Multi-stage disorder-fear-withdrawal model |
| Ferraro (1995) | Vulnerability-based explanation | Perceived vulnerability explains fear-risk paradox |
| LaGrange, Ferraro & Supancic (1992) | Formless vs. concrete fear distinction | Incivilities predict concrete fear more strongly |
| Sampson & Raudenbush (2004) | Racialized disorder perception | Fear heightened by racial composition independent of conditions |
| Chiricos, Padgett & Gertz (2000) | Media effects on fear | Local news coverage independently raises fear |
Table 2. Distinct Fear-Reduction and Crime-Reduction Policy Levers
| Intervention Type | Primary Target | Coercive Element | Evidence of Fear Reduction | Evidence of Crime Reduction |
|---|---|---|---|---|
| Street lighting improvement | Fear, perceived risk | None | Moderate to strong | Weak to moderate |
| Sightline-clearing landscaping | Fear, perceived risk | None | Moderate | Weak |
| Visible community policing | Fear, reassurance | Low | Moderate to strong | Weak to moderate |
| Order-maintenance enforcement | Disorder, fear, crime | High | Moderate | Mixed |
| Zero-tolerance enforcement | Crime, disorder | Very high | Weak to moderate | Mixed, contested |
Conclusion
The incivilities thesis identifies the most consistently replicated finding in the broader broken windows research program: visible disorder reliably raises residents’ fear of crime, a relationship documented across numerous cities, methodologies, and decades, even where disorder’s independent contribution to serious crime remains considerably more contested. This consistent fear-disorder association, however, has proven more complicated than early formulations suggested, shaped by demographic vulnerability, racialized perception, media exposure, and the specific distinction between formless and concrete fear, each of which qualifies any simple, universal account of how disorder produces fear.
For policy purposes, the incivilities thesis’s central contribution has been to establish fear as a legitimate, independently measurable target of neighborhood intervention, distinct from crime reduction itself, opening space for less coercive fear-reduction strategies that do not depend on the more contested and civil-liberties-implicated disorder-enforcement approaches examined elsewhere in this silo. The remaining challenge, addressed in Alternatives to Broken Windows Policing, is determining which fear-reduction strategies achieve genuine, lasting reassurance without simply displacing the underlying anxieties that media exposure, racialized perception, and demographic vulnerability continue to sustain independent of any single neighborhood’s physical or social conditions.
Related Articles
- Core Principles of Broken Windows Theory
- Social Disorder and Crime
- Collective Efficacy Theory and Broken Windows
- Broken Windows Theory and Race
- Alternatives to Broken Windows Policing
References
- Chiricos, T., Hogan, M., & Gertz, M. (1997). Racial composition of neighborhood and fear of crime. Criminology, 35(1), 107–131.
- Chiricos, T., Padgett, K., & Gertz, M. (2000). Fear, TV news, and the reality of crime. Criminology, 38(3), 755–786.
- Farrall, S., Jackson, J., & Gray, E. (2009). Social Order and the Fear of Crime in Contemporary Times. 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.
- Garofalo, J. (1981). The fear of crime: Causes and consequences. Journal of Criminal Law and Criminology, 72(2), 839–857.
- Hale, C. (1996). Fear of crime: A review of the literature. International Review of Victimology, 4(2), 79–150.
- Innes, M. (2004). Signal crimes and signal disorders: Notes on deviance as communicative action. British Journal of Sociology, 55(3), 335–355.
- Jackson, J. (2004). Experience and expression in the fear of crime. British Journal of Criminology, 44(6), 946–966.
- LaGrange, R. L., Ferraro, K. F., & Supancic, M. (1992). Perceived risk and fear of crime: Role of social and physical incivilities. Journal of Research in Crime and Delinquency, 29(3), 311–334.
- Sampson, R. J., & Raudenbush, S. W. (2004). Seeing disorder: Neighborhood stigma and the social construction of “broken windows.” Social Psychology Quarterly, 67(4), 319–342.
- Skogan, W. G. (1990). Disorder and Decline: Crime and the Spiral of Decay in American Neighborhoods. Free Press.
- Skogan, W. G., & Maxfield, M. G. (1981). Coping with Crime: Individual and Neighborhood Reactions. Sage Publications.
- Taylor, R. B., & Hale, M. (1986). Testing alternative models of fear of crime. Journal of Criminal Law and Criminology, 77(1), 151–189.
- Warr, M. (1990). Dangerous situations: Social context and fear of victimization. Social Forces, 68(3), 891–907.
- Wilson, J. Q., & Kelling, G. L. (1982). Broken windows: The police and neighborhood safety. The Atlantic Monthly, 249(3), 29–38.