Schneier on Security
A blog covering security and security technology.
« Another Password Analysis |
| Friday Squid Blogging: Jumbo Squid Teeth »
February 20, 2009
The "Broken Windows" Theory of Crimefighting
Evidence of its effectiveness:
Researchers, working with police, identified 34 crime hot spots. In half of them, authorities set to work—clearing trash from the sidewalks, fixing street lights, and sending loiterers scurrying. Abandoned buildings were secured, businesses forced to meet code, and more arrests made for misdemeanors. Mental health services and homeless aid referrals expanded.
In the remaining hot spots, normal policing and services continued.
Then researchers from Harvard and Suffolk University sat back and watched, meticulously recording criminal incidents in each of the hot spots.
The results, just now circulating in law enforcement circles, are striking: A 20 percent plunge in calls to police from the parts of town that received extra attention. It is seen as strong scientific evidence that the long-debated "broken windows" theory really works—that disorderly conditions breed bad behavior, and that fixing them can help prevent crime.
Many police departments across the country already use elements of the broken windows theory, or focus on crime hot spots. The Lowell experiment offers guidance on what seems to work best. Cleaning up the physical environment was very effective; misdemeanor arrests less so, and boosting social services had no apparent impact.
EDITED TO ADD (3/13): The paper.
Posted on February 20, 2009 at 12:03 PM
• 61 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
It just shows how important human psychology is in public safety and security. It wasn't the cleanliness itself but the change in perceptions caused by the changes. I live in a nice clean trailer park with standards for maintenance and yard care and we have much lower crime here than in my old trailer park which had much lower standards which were unevenly enforced.
My guess is that there should be a point of equilibrium. A little bit of the squeaky wheel...
That's interesting, but there's something about the police's behavior that disturbs me:
"... sending loiterers scurrying ..."
Since when is it illegal to stand around not doing anything? I find it sad that a word exists that's clearly intended to lend an air of illegitimacy to the perfectly innocent act of not having a specific purpose in mind whenever you're out of your house.
I guess a relevant question is whether the local reduction in crime resulted in a similar increase elsewhere? I.e., is crime reduced for society as a whole, or just transferred from one place to another?
I thought this had already been proven many times over, such as with the New York City Transit Authority in the 80s. It also can be said to reflect a "path of least resistance" model of crime. If I remember correctly, I thought you often criticize security controls that shift crime. Isn't this a perfect example of such a system?
Ah, yes, gentrification. Jakob Pagter and Davi Ottenheimer seem to get it.
I haven't RTFA yet (no time, have to go soon), but it looks like the effective thing was making it "look" nicer, not the increased policing.
This means its psychological... so it could just be shifting the crime, or it could actually be affecting the criminal psyche.
Jakob, I was wondering the exact same thing.
And this excerpt leads me to believe that may be somewhat true:
"A brick apartment building that once racked up 100 calls to police in a three-month period has, she said, had just one incident over the last six weeks. Gone, she noted, are the unregistered cars in the parking lot, the broken fence, and the code violations in the building - as well as problem tenants and crime."
If the problem tenants are gone, where did they go? I take it to mean that they moved somewhere else, died, or are in jail, with moving probably being the case for the majority.
However, there may be a chance that this strategy will shift crime somewhat, but also lead to an overall decrease, i.e. one are goes down 20% (per capita, let's say) but another goes up 15% (per capita), leading to an overall 5% decrease.
Of course, it could just mean that the crooks are avoiding the built-up regions that appear to be getting extra attention from the authorities. They just move elsewhere to avoid the hassles but they gradually return when the heat has died down.
That's what I'm talking about. NYC systematically cleaned the appearance of subways in the 80s and it reduced crime. I remember it well.
Here's a related article from the Atlantic in 1982
"But how can a neighborhood be "safer" when the crime rate has not gone down—in fact, may have gone up? Finding the answer requires first that we understand what most often frightens people in public places."
Malcolm Gladwell talked about something (related?) in "Tipping Point." Bottom line is that prosecuting the little crimes reduced the instances of the bigger crimes. I.e., catching people in NYC who would jump subway turnstiles seemed to discourage other crimes, like gun or drug crimes.
Premediated crimes of the "need to raise so-and-so much money for drugs" variety probably just shifts around.
But it sounds plausible that things such as assault, domestic violence, vandalism, etc. can be fueled by fear or despair, and that they can truly be reduced by making the environment less conductive to them.
re: Garrett G
If someone is going to regularly commit crimes, maybe they're also likely to commit misdemeanors. If you bust them for the small things, they are going to change their behavior - or be put in jail. This may be part of what leads to the reduction in crime.
Regarding the clean up of New York City: one aspect of Rudy Giuliani's policing efforts was a disregard of civil liberites for suspects.
Sept 11 came and squelched all investigations into this conduct. However, one can not discount its effectiveness in moving crime out of the city.
MLK Jr. was saying this some 50-odd years ago, while being pelted with rocks and racial slurs.
I sure am glad I live in a world where it takes over 50 years, but just one obvious report from the upper-class mind-wipes at Harvard, to drill it into the cops' heads.
We've long known this.
And it's not moving the problem to other places, although the loitering comment can contribute to that perception (and possibly to some of the reality).
It's a pyschological cue in people's minds.
People, in general, try to conform. Whether it's keeping buildings neat and tidy or listening to the minister on Sunday repeat the though shalt not steal...it reinforces good behavior.
It would also be a TERRIFIC way to try and have a postive impact on urban areas in this time of many other questionable "economic stimuli" being thrown around. Something similiar to the old CCC but focused on urban blight -- from cleaning up overgrown lots, renovating abandoned buildings, installing new sidewalks, rebuilding playgrounds, providing labor and low-cost loans for facade and common area improvements of private buildings. These are projects that could be shovel ready within weeks, and provide employment in areas traditionally economically depressed.
Back in the early 1990s I visited my sister at a camp she worked at just outside of Albany, NY. Since it was up on the edge of the Hudson Valley escarpment, there were a lot of radio and TV stations' towers. Since I was working in radio at the time, I took a walking tour of most of the sites.
Right away I noticed a difference in myself when I walked up to a building where the grass was trimmed and everything appeared ship-shape, versus a building that had grass and bushes growing all over, dirty windows, and security/motion lights that didn't work. In fact, I didn't explore very close to the buildings where everything was cared for, because I felt like someone must be watching me -- even though I knew without a doubt that there wasn't a single person within a mile or more.
When I got back to the station I worked at I related this, and we ended up engaging someone who lived near our transmitter/tower to mow the grass for us, keep things trimmed, and alert us to anything unusual. We never had a problem with any unwanted visitors.
Bruce, I've been to Lowell, and to it's "sister city" Lawrence, just a few miles to the northwest. It's amazing, the difference between the two.
In the '80's, Senator Paul Tsongas funneled federal money into his hometown of Lowell, renovating old cotton mills, making National Parks out of some of them, and such other efforts. Lawrence didn't get as much attention.
Now, granted, there's a branch of the UMass system in Lowell, unlike Lawrence, but even still, you drive through Lawrence quickly, with the windows up and the doors locked.
But in Lowell, you go to a Spinners game (the Red Sox single-A affiliate), maybe walk along the riverbank, perhaps catch a show at the Merrimack Repertory Theatre.
If you give people something nice, they tend to take ownership of it, and slow it's degradation.
@Chris G: I definitely agree with you.
I think it's a combo of factors:
a) catching people who commit both "small" and "large" crimes, and also
b) deterring people from "crime escalation" (movement from small to more serious crimes).
I suspect it's a case of "if others are smoking it's ok to light up".
That is if there is very visable neglect and petty crime in an area, others will feel it is ok to do likewise and you start getting a slow spiral downwards.
Give the area a good dusting off and lick of paint and suddenly it does not feel a safe place to commit a petty crime any more.
Likewise the more serious street crime will go from the area, but as others have noted it will move to another area simply because that is a "means of income" for some criminals not just a form of entertainment as it is for vandalisum and "rolling drunks".
Also there is "expectation crime" prostitution and drugs selling tends to happen in certain types of neighbourhood simply due to "expectation" leading to "fulfilment". That is if people think that the area looks like a place they can buy sex or drugs then fairly soon their enquires will lead to people supplying the requested commodity.
If the area looks like it will nolonger supply what is "expected" then customers will hesitate to go there and will look elsewhere and the suppliers will follow.
@Adrian Lopez: "That's interesting, but there's something about the police's behavior that disturbs me: "... sending loiterers scurrying ..." Since when is it illegal to stand around not doing anything? I find it sad that a word exists that's clearly intended to lend an air of illegitimacy to the perfectly innocent act of not having a specific purpose in mind whenever you're out of your house."
It is unclear to me from the article whether the loiterers scurried because they police made them scurry, or whether they scurried voluntarily due to police presence. That makes a big difference.
It's no more illegal to stand around doing nothing around police than it is to do the speed limit or pass another driver with a police car around. Still, many people chose to slow down or not pass someone going 10 under the speed limit if a cop car is there.
@Adrian Lopez, HJohn
Loitering is, of course, the crime of failing to spend money. If you sit around doing nothing except paying for the occasional refill of your coffee, you're fine. But if you sit around doing nothing except drink coffee from a thermos you brought from home, you are loitering.
Did the two groups have equal resources? IOW, was the experiment simply a matter of changing how the cops allocated their time?
From what's presented in the article, it's not clear that "fixing the windows" is what's having an impact. It sounds like they're putting a lot of police on the street, arresting folks who are drinking in public, etc.
So you've got a lot of cops actively patrolling the street watching for disturbances, rather than just bouncing from emergency call to emergency call after the crime has already occurred.
It doesn't seem like a huge leap to say that increased, proactive, and obvious police presence would lead to reduced crime, and that windows and the fixing thereof might not be making much difference.
I would assume readers of Bruce's blog are IT security professionals, or crypto/math academics. I have never known someone in these fields to live in a trailer park--they tend to make significantly more money than most people do.
So I'm quite curious as to how you find yourself in a trailer and on Bruce's blog. Please share.
It seems, we'll need more of the good old "bLockwart" agein, don't we?
I still don't subscribe to the "disorderly conduct breeds crime" theory: My memory is, the all fascist ideologies wanted one thing from their subjects: cleanliness and order. From that I got the idea that demanding "orderly conduct" from citizens is actually more a technique to start subjugation at a low level of resistance by demanding something that does seem not more than just a matter of course. But as soon as the citizens have begun to obay and also have accepted the invasion of the state into the small matters of _conduct_ - you can teach thema almost every trick in the book ("Wollt ihr den totalen Krieg?" etc).
Obeisance is the first step. Fighting "loitering" is a good symptom of something going wrong.
"Fighting "loitering" is a good symptom of something going wrong."
Hmm, I'm not sure where you live but in London we have quite a few areas where gangs of "youths" have caused problems.
Initialy it is petty vandalisum such as graf-sigs on peoples fences etc then breaking glass in bus shelters and telephone boxes. Eventualy it leads to more serious crime, and areas become "no go" areas after dark.
A policy of "control areas" where the Police are allowed to move "youth" out or take them home has had some effect in reducing the problems and stoping it getting going in other areas.
Personaly I wish that the local Government would spend money on improving "youth" facilities so that they had something to do other than loitering on the streets, rather than waisting resources on "moving on".
As has once been observed "The Devil soon makes work for idle hands"...
Rather simplistic retort if you ask me.
It's less "disorderly conduct breeds crime", and more "poverty, many times even the relative perception of poverty, breeds crime". With a bit of Clive's allusions to mob mentality with "if others are smoking it's ok to light up" mixed in there.
This isn't about control or fascism, it's about the idea that people in less desperate or seemingly less desperate conditions FEELING less desperate, which would obviously in turn lead to less desperate measures being employed to fulfill whatever needs the petty crimes in question tend to fulfill, and again in terms of the actual 'broken windows' idea, if people *aren't lighting up anywhere around you, you are more likely to hesitate before you do, providing (possibly) just enough time for your conscience to step in (or in our generation's case, ADD to kick in, haha) and change your mind.
Also, I don't know what neighborhood you live in, but in East Pilsen in Chicago, people loitering (or, excuse me, 'standing around in groups minding their own business and not spending money') is very nearly a recipe for you (the passerby) getting your ass kicked and your money stolen. If you don't believe the hype, take your white bread self on a walk down 18th sometime. Surely extreme examples of this don't warrant a country wide ban on 'hanging out', but you use your best judgement in those situtations, and frankly when it comes to those situations (profiling being an inexact science by definition, at best) I tend to trust the beat cops in making the call as to who to bother and who to let alone, cause it really doesn't take a Harvard report to tell the difference between a bunch of kids hanging out and maybe smoking something they shouldn't be, and a group of gang-bangers who get their kicks off of kicking the shit out of you.
> Since when is it illegal to stand around not doing anything?
Approximately 6 hundred years, according to the Wikipedia article.
> I find it sad that a word exists that's clearly intended to lend an air of illegitimacy to the perfectly innocent act of not having a specific purpose in mind whenever you're out of your house.
The exact wording varies between jurisdictions, but unlawful loitering generally requires more than simply "not having a specific purpose in mind." In the US, state laws as vague as that have been found to be unconstitutional, and replaced with more specific definitions of the harm that they are trying to prevent.
"It's no more illegal to stand around doing nothing around police than it is to do the speed limit or pass another driver with a police car around."
Once upon a time, on I-90 in South Dakota (west of Rapid, if you know that part of the country), I came upon a long line of cars in the right lane, all going 45.
I changed into the left lane and passed them at about 50.
Cost me an hour on the side of the road visiting with a state patrol drug-sniffing dog.
There was a write-up in The Economist within the past couple months about a similar experiment but with even better controls. The researchers took two, essentially similar corners, buildings, alleyways. One they carefully kept messed up - bits of trash, graffiti, etc. The other - which was nearby - they carefully kept clean. Petty crime was much lower in the clean place than the messy one.
I don't know if copyrights would allow me to post the article. Anyone (Bruce, it's your blog) know?
I live in a trailer park because I am a liqour store clerk. I ended up reading this blog because I'm a hobbyist programmer. That ended up here somehow.
I live near Lowell, MA and can say it's a nice city that's getting better. As someone said above, you drive through Lawrence w/ the windows locked.
I go to Lowell all the time and feel safe. I don't see lots of police around, but I sure do see the cleaned up buildings and whatnot.
Crime Prevention Through Environmental Design.
At least is an step toward evidence-based (no pun intended) crime control. As is the case with other forms of security, including computer security, there's seldom hard data behind policy decisions.
"Crime Prevention Through Environmental Design."
This has been hapening since 9/11 (the specific crime being terrorism).
Unfortunatly the results have so far not been good.
Instead of using natural materials in an esthetic manner they have tended to use soul less mass produced re-inforced concreate.
Instead of engendering a calm feeling of blending with the environment. It instills a "fortress mentality" which makes the majority of people uncomfortable and re-enforces a "them -v- us" feeling which deters people from what would normaly be public spaces.
Although instilling mild paranoia in people is desirable around certain facilities (prisons / military bases). It is undesirable around public buildings of Government agencies that need the trust an co-operation of the public.
It can as has been seen under certain past regiems induce feelings of oppression and foster civil dissent.
After a little thought I offer another reason why a clean neat environment has a lower crime rate.
If an area is run down dirty and visualy unappealing people do not notice further degradation as they cannot "see the new tree in the forest".
However in a neat clean environment people can get upset by something as mild as a neighbour not cutting their front lawn. In such an environment even the smallest of undesirable changes stand out and this inducess a level of response in the populas that makes further transgretions painfull for the casual offender.
Although I'm not in favour of a "Stepford Wives" community ethos it does have the ability to surpress small visable crimes...
> ... I'm not in favour of a "Stepford Wives" community ethos ...
Yes, Clive, I too baulk at murdering our wives!!
And we who are in other venues/life situations/not-so-simple twists of fate are not allowed to have aspirations? Or are we too stupid (in your apparent opinion)?
What do you care where we come from or what we do in the real world? We can think.
For the record, give me a million bucks (no, thanks, I'm working on earning it), and I'd pay off the mortgage and fix up the place a bit, leaving a lot left over for a retirement next egg, and a lot of good done around my neighborhood.
As it is, I'll earn my way there, and still do the good around the neighborhood.
But I wouldn't move. I like it here.
Pardon me, regarding Roboticus.
Everything else holds, though.
"Yes, Clive, I too baulk at murdering our wives!!"
You've got it in one, a faustian bargin, Stepford was a perfect place to have a family it was crime free with elegant houses run by genteel women being perfect house wives and lovers. Only one snag, first you have to sell your soul...
Sometimes the cost of fitting in to an apparently desirable social setting or society is much much higher than at first it appears.
The way the economy is going, soon being a "liqour store clerk" may be one of the few jobs with long term security...
A lot of the ICT contract people I know are either "resting" or if they had a little foresight become public sector "permies" at much lower renumeration in the hope of weathering the storm to come.
I know that a number that bought houses in the past six years are now wishing that they had not. And one or two are trying to sell to avoid bankruptcy.
Again one or two with foresight sold their houses at the top of the market and moved to rented housing.
One guy I know sold on the high and moved into rented, has just bought a wreck of a house and is now living in a trailer on site and doing it up.
Others are doing what they can with what they have got.
Such is the state of the world we find ourselves in.
At least in the UK we have (in theory) some protection for the people unfortunate enough to get "down sized", but it's not a lot.
To quote the song,
"Oh the times they are a changing"
Well if you're really interested in programming, consider taking some courses at a community college. In a few years you could have a career in IT at 4x your current salary. I'm not saying there is anything wrong with working as a clerk (well, eventually convenience stores will be replaced with sophisticated vending machines, but I digress). I turned my "hobby" with the computer into a highly-paid career. I'm sure you could, too.
Sounds like these cops have discovered Heinrich's Pyramid and applied it to crime.
> If the problem tenants are gone, where did they go? I take it to mean that they moved somewhere else, died, or are in jail, with moving probably being the case for the majority.
It could simply mean that they no longer cause problems for their neighbours.
> However, there may be a chance that this strategy will shift crime somewhat, but also lead to an overall decrease, i.e. one are goes down 20% (per capita, let's say) but another goes up 15% (per capita), leading to an overall 5% decrease.
Every time a crime reduction strategy seems to be successful, the anti-Pollyannas trot out the weird "conservation of total number of crimes" theory, i.e. the crime was probably just displaced elsewhere because the junky still needs to mug someone to pay for his fix. This, not withstanding that police statistics show that overall crime rates do, in fact, change -- continuously.
At least one controlled crime reduction study actually made a pretty good effort to address this issue. This was the LAPD's "Operation Cul-de-Sac", which in 1990-1 attempted to reduce violent crimes in a South Central LA neighbourhood through urban design. LAPD carried out the operation, while NIJ independently conducted a controlled study of its effects. The operation saw a quite dramatic reduction in rates of violent crime (28% reduction in assault and battery, 85% reduction in murder), until it was cancelled due to political pressures, at which point the rates almost immediately reverted to their previous levels.
At the same time, they also monitored rates of violent crime in all the surrounding areas, to a considerable radius. They saw no statistically significant change at all in those areas, although allowing for errors could not rule out that possibly as much as 5% may have been displaced. Otherwise, however, these other potential crimes simply never occurred.
It's plausible - but the study does not cover enough ground to demonstrate a cause and effect.
Money spent tidying up the city is probably well spent anyway.
It's well known in criminalistics that
the number one motive for increasing the severity of crimes committed is
to conceal or extend the practice of previous, lesser crimes.
Hence, prevent the lesser crimes from repeating,
and any new crimes are neither concealing nor extending their practice.
Amusingly, an alternative meaning of "broken windows" is basically "spin", where a claim for the benefits of a policy "involve[s] an incomplete accounting for the consequences of an action".
For example in this case, maybe when you clean up a bad neighborhood, apartments are easier to rent, landlords can be choosy, and unpleasant tenants are forced to move elsewhere, taking some crime with them. Thus crime is not cured, just displaced.
It's difficult to be sure, but I think this is the report
Braga, Anthony A. Police Enforcement Strategies to Prevent Crime in Hot Spot Areas. Crime Prevention Research Reviews, No. 2. Washington, DC: U.S. Department of Justice, Office of Community Oriented Policing Services, 2008.
I dont know about the validity of "bw" theory; however I am personally convinced through personal experience that if graffitus are erased/cleaned/painted over the instant that they are discovered, then within a very short period of time they cease to appear at all. I further doubt that more than 1/3 of the graffiti producers move to other areas to continue the profession and that it is therefore absolutely reduced not merely displaced; and that they take up other pursuits (hopefully wholesome and profitable; or at least untroublesome, like watching TV) for entertainment.
Alternatively I am convinced that if graffitus are left untouched they breed (opinion - the people who did it brag to their friends when they drive past it and then their friends have to "one-up" them with more/larger/better/louder works) and eventually become a tremendous nuisance (for example the Cochem bridge over the Mosel river. Some of the first graffiti I had ever seen in Germany was on the side of the bridge ["bullen raus"] in 1998. I went back again in 2001 and it was still there and it had brought a lot of friends. Seeing it depressed me and made me think much less highly of Cochem; which I had previously idolized as the prettiest place in the world).
I am also of the opinion that an area where there is a lot of graffiti on the walls becomes a place where more nefarious crimes begin to occur. So I tend to believe in the "bw" theory.
I heard an interview on Australian radio a few months ago in which they reported the results of planting trees and flowers in run-down areas overseas -- nothing to do with the cops.
The average response was a 7 per cent reduction in crime relative to comparable areas with no plantings. The experiment was done properly with careful controls.
I was thinking, it makes a good rejoinder to Microsoft's often-claimed claim that MSWindows is targetted more often than the Un*ces because it is more popular ... maybe it's targetted more because it's broken and ill-maintained...
It looks like the actual study referred to in the article was probably: Braga, Anthony A., and Brenda J. Bond. "Policing Crime and Disorder Hot Spots: A Randomized Controlled Trial." Criminology 46.3 (August 2008): 577-607.
Freakonomics... they also like to cite legalized and easy-to-obtain abortions as playing a major role in crime reduction.
Also, when WSJ contributors are asked to write articles regarding income gaps vs. crime rates, they also tend to find a number of non-income/poverty related issues to blame.
The sad fact is, whether other factors play a major role or not, income gaps do tend to fall into a pretty steady ratio with crime rates, a great example of this is Washington DC, or even Japan (well, Japan is a good example, but obviously being culturally light years apart from us, it's hard to use as a concrete comparison, but they do have an insanely low rate of crime overall, and their income gap as a country is one of the lowest in the world).
The insane crime-rate slide of the 90's was a fluke that may have had a number of other observable factors that seem to contradict the idea, but it shouldn't be used to discount the very real notion that poverty tends to breed crime. A notion that even Freakonomics' somewhat off-the-wall idea of abortion could even serve to illustrate in a few roundabout ways.
Some cost/benefit analysis please.
I am skeptical about any simple conclusions that environment changes magically reduce crime.
There are so many variables to control and the more controlled something is the more expensive it is to produce.
If you are a scientist, there is an incentive to produce attractive results which will attract attention and funding.
Feynman was asked about the applications of a certain piece of research and he said "there aren't any". The reply was "well, then we will not get any funding". Feynman disliked this and favoured accepting the lack of funding that might result from simply stating the truth
Roger, your "Approximately 6 hundred years" retort cracked me up. Good one.
@Vader: I bet you think tagging is "art," too.
@Clive Robinson and Extreme Misspelling: You are entirely right. Loitering and the anti-social behavior that follows is a world-wide problem, *if* people let it be a problem. ""The only thing necessary for the triumph of evil, is for good men to do nothing." [Edmund Burke or whomever.]
We have the Lowell experiment on the railroad I work for, on a much smaller scale and with no outside funding: If someone keeps the grass mowed across from a certain old depot, the area stays relatively clean. If the grass gets high, we have to pick beer bottles out of it before the next mowing.
Does it just move crime to another place? Maybe. Maybe the miscreants will just move to a place where they're tolerated, because people in that place believe stopping them would be "fascist ideology" or some such drivel.
"Does it just move crime to another place? Maybe."
That is a question that needs to be answered before we can say if the idea realy works or not (depending on your chosen success criteria).
"Maybe the miscreants will just move to a place where they're tolerated... "
Hmm, if they do move on I'm not sure I would use the word "tolerated", for the likley places they would go "feared" is more likley.
Although individuals can be quite happy to live in their own mess, there are probably very few who are happy to live in other peoples mess (otherwise they would be selling houses in rubbish dumps / tips).
I've seen this effect directly, and think it stems from a simple principle "Evil Flourishes When Good Men Do Nothing". If you take a direct responsibility for fixing a problem and making reparations, the criminality will decline - criminals work in shadows & are cowards at heart. Light a few candles & of course they will move on.
Schneier.com is a personal website. Opinions expressed are not necessarily those of BT.