Broken Windows Revisited | 1

Part 1 of a 3-part series that revisits a number of aspects of Broken Windows. This part looks at the general theoretical and empirical background.

Since its publication nearly 40 years ago Wilson and Kelling’s “Broken Windows” has become one of the most-influential and widely-adopted approaches to our understanding of crime and policing, particularly in America but also across a number of other Western societies, including in some small part, the UK.

Broken Windows: Wilson and Kelling
Broken Windows: The Atlantic Monthly

To understand how and why Broken Windows has come to have such a huge and lasting impact on our thinking about crime and policing we need to understand something about both its theoretical origins and underpinning and its empirical applications.

Theoretically, there’s nothing particularly innovative about Broken Windows.

It mainly draws on a range of well-known human ecological ideas and observations about crime popularised 40 – 50 years previously in the work of Chicago School theorists such as Park, Burgess, Shaw and McKay. More-specifically, it owes a theoretical debt to the latter’s work on Zones of Transition and, in particular, the notion of interstitial “inner city” zones in which crime flourished as a consequence of “socially disorganised spaces”.

The important thing Wilson and Kelling did, however, was take a relatively simple, well-known and largely uncontested proposition – that “Disorder and crime are linked” – and elevate it into both a more-general explanation for crime and, ultimately, a practical way of limiting or eliminating crime in an area.

They did this by not just arguing disorder and crime were linked, in the sense that disordered spaces were associated with crime (a proposition that has been tested both academically – through something like Keizer’s (2008) “disorder experiments” in the Netherlands – and experientially: pubs, clubs and fast-food outlets, for example, are  “disorderly spaces” where various forms of criminality – from violence to littering – are “commonly-known” to be more-likely to occur).

Rather, they argued disorder and crime were developmentally linked in a clear and unequivocal direction: social disorder was a cause of crime – a relationship neatly summed-up in one of their most famous observations:

If a window in a building is broken and is left unrepaired, all the rest of the windows will soon be broken…one unrepaired broken window is a signal that no one cares, and so breaking more windows costs nothing”.

To put this in more-conventional terms, relatively minor forms of social disorder, such as windows in a disused warehouse being broken and left unrepaired, leads to larger, more-serious, forms of criminal disorder – a general proposition that led to the obvious conclusion that the way to control crime was to prevent disorder occurring in the first place.

Wilson and Kelling’s equivocation about exactly how disorder could be prevented – they were, for example, highly-sceptical about the effectiveness of changes to policing strategies in American cities (from the mobile policing that had developed in the 1970’s, whereby police officers patrolled large areas from the comfort of their squad cars, to the selective re-introduction of “beat patrols” in places like New Jersey that saw officers “walking the beat” in smaller, more-localised, areas) – was not something shared by New York City Police Commissioner Bill Bratton. His interpretation was both selective (order maintenance was primarily a police role) and literal: the way to prevent both crime generally and major forms of criminal disorder specifically, was to clamp-down immediately, heavily and without exception on minor forms of disorder.

Order Maintenance Policing

Bratton’s introduction of Order Maintenance Policing (OMP), more-usually termed “Zero-Tolerance Policing” (ZTP), in the early 1990’s involved police officers stamping down hard on all kinds of minor social infractions – from fare-dodging to littering and street begging. The broad idea was that by controlling relatively minor social misdemeanours hitherto disorderly areas could be reclaimed by “law-abiding citizens” as ordered spaces. This, in turn, would lead to not only a drastic reduction in petty criminality but also, following Broken Windows, substantial falls in more-serious forms of criminality once an area was “ordered”. Or, as Sullivan and O’Keeffe (2017) put it, the idea that:

Proactive policing – which involves systematic and aggressive enforcement of low-level violations – is positively related to reports of major crime”.

Debates about Broken Windows have largely coalesced around Bratton’s empirical application and have led to a certain polarisation of opinion:

Proponents, for example, generally argue OMP has led to:

  • reduced rates of crime
  • a reduced fear of crime
  • the removal of criminal role models from sensitive areas
  • troublemakers, when caught young and for relatively minor misdemeanour crimes, are prevented from developing into more-serious career criminals
  • reinterpretations of the primary role of the police to one of “maintaining order”
  • an increase in legitimate neighbourhood activities once criminals are removed
  • prevention of the development of major crimes.
  • Critics, on the other hand, point to what they see as a range of weaknesses:

  • we can’t be certain disorder causes crime. Crime may cause disorder
  • crime has generally reduced in countries like Britain and America, regardless of whether OMP policing was in operation. In America, crime reduced at much the same rate in cities that employed OMP as those that did not: correlation, in this respect, does not equal causation
  • other social factors (an unrelated growth in the prison population, a fall in the number of young males in the population, fewer unemployed males…) explain the fall in crime in cities where OMP was applied
  • it targets the weakest, easiest to catch, prosecute and convict offenders, leaving major criminals largely untouched
  • labelling large numbers of people as “criminals” for relatively minor offences risks the creation of deviancy amplification spirals.
  • While this general debate remains largely unresolved – it has tended to devolve into a largely ideological “You pays your money and you takes your choice” selection of evidence – it’s clear that the empirical application of the theoretical propositions underpinning Broken Windows has raised a number of questions that require some sort of answer.

    Which is what we’ll do in subsequent posts in relation to both the general theoretical basis of Broken Windows (Part 2: does this interpretation of the general ecological model stand-up to closer examination?) and the question of whether the proposition that minor social disorder leads to major forms of criminality can be empirically justified.

    While you’re waiting, you might find this trailer for a film we recently made on the relationship between Space, Place and Crime useful (definitely) and informative (possibly).

    It does have quite a bit about Broken Windows and Zero Tolerance Policing, if that’s any help.

    Stay Updated

    Enter your email to be notified when we post something new:

    Archived Posts