Wednesday, April 28, 2010

Sin-city and crime generators

What I learned in Vegas...

A funny thing happened to me in Las Vegas last week. No, it's not what you're thinking! It's not really about clear-headed thinking at all. It's about some fuzzy thinking that replaces coherent understanding about preventing crime.

I've said before the CPTED and Design Out Crime folks don't talk much about the theories behind their work. A few of those thorny theory issues showed up in my blogs last year, such as ROTO (Research-On-The-Obvious).

There is another side to that coin. I'm referring to the fuzzy thinking of some academicians who study crime-and-place and their kin (crime mappers, police crime analysts).

Let's remember the main purpose of theory, at least in the empirical sciences, is to provide a plausible and testable explanation of why something happens, in this case crime. Events surrounding that explanation - where and when crime occurs - are merely descriptive symbols of the main event, not the actual theory that explains "why" the event happens in the first place.

From this point of view, CPTED and Design Out Crime have very few actual theories - only descriptive symbols.


Descriptive symbols are useful. Describing where malaria starts killing people may help us isolate the outbreak and target treatment. Symbols may point us in the right direction toward explanation. But they do not explain the biology of malaria - and that is what kills us! Mosquito nets may help us and we need them. Ultimately, we need to be working toward a cure. We mustn't rest with nets. Same with crime.

What about gambling, casinos, and sin-city?

As I walked around Las Vegas last week I began thinking about all the casino's, the craziness, and all the activity. In Advanced CPTED there is something called "crime generators". These are locales in the city that tend to generate crime opportunities. Large casinos fit the crime generator tag. If mappers scoured the crime patterns, no doubt they'd find lots of hotspots with pick-pocketing, assaults, drug dealing, and fraud in and around the casinos. Or perhaps casino security displaces it to venues just outside the Strip? Either way, no doubt someone would come up with a "theory of casino crime locations", which isn't really a theory at all.

This is exactly what happened with the Broken Windows "theory". Crime went down in New York. Unfortunately for Broken Window Theory, as Professor Zimring showed in his book, The Great American Crime Decline, crime went down in places with no Broken Window programs. That's because descriptive symbols won't really stop crime in the long run. Only prevention based on sound theories can do that.

Yes, theories matter.

Does that make casino's crime generators? Casinos no doubt "generate" direct and indirect crime. But if we could calculate per capita crime rates around casinos, would they be any more crime-prone than other high activity places?

Do activity generators = crime generators?

Probably not in the case of large political rallies (except if you find yourself in Thailand this week). How about in the case of large football games with thousands of drunken fans? Those probably do generate similar (or more) assaults, except if we limit alcohol sales and the home team wins.

That's the problem with descriptive symbols versus real theory. They tell us something about where, and little about why.


It might be simple to conclude sin-city Vegas is also crime city. But FBI crime stats don't show that. We can argue all we want about unreported crime, gambling impact on family life, and so forth (those things are true and serious). Yet according to the 2008 FBI crime rates, Las Vegas wasn't even in the top 20 cities for total violent crime nor in the top 50 for total property crime cities.

So much for crime generators.

Personally, I'm no fan of the Strip in Las Vegas. There are no doubt many tragic realities that arise in that famous (infamous?) epicenter of self-absorption.

Whatever those realities, we must not confuse descriptive symbols that plot when and where with actual theories explaining why crime happens. Descriptive symbols may help us target crime and temporarily reduce it with 1st Generation CPTED and Design Out Crime. But they won't help us prevent it in the long run. Nor will they replace proper and robust theories that help us build safer places - including entertainment meccas like Vegas - in the years to come.

6 Replies so far - Add your comment

todd said...

Mr Saville,

I've heard alot about Vegas but no information about Vegas PD.

Question: Have you researched Vegas PD for budget, size, patrol districts, investigative units, casino and business cooperation, community involvment, State and Federal Support? Do you think any of these factors or combination of factors could lead to the city being relatively safe? If you have which factor(s) seems to be most important?

GSaville said...

Hi Todd, thanks for that question. It's a good one and I'm not sure I can do it justice. But I'll give it a try...

I'm not sure I was singing the prevention praises of Sin-City. If it seemed so, my apologies.

My point in the blog was to question whether crime-and-place theories about crime "generators" like casinos really tell us anything useful about the causes of crime.

But in my view, shuffling the "where-would-you-like-to-live" cards, the deck is NOT stacked in favor of Las Vegas, at least anywhere near the Strip. Las Vegas cannot say it is safe when it reports that for every 100,000 people there were 364 robberies, 842 stolen autos, over 1,000 burglaries, and 8 murders per capita in 2008. That is 120 actual murders in the Metro area.

Not much to brag about. It's true that in 2009 the Las Vegas Sun reported that crime was down. That is small solace for the thousands of victims.

Add that to reports that Las Vegas may be one of the worst foreclosure ridden cities from the current Recession, and the picture is pretty dismal.

Policing strategies, numbers of cops, and crime reduction?

That's a well-worn road I won't tread on again. Suffice to say my blog last year on Zimring's book The Great American Crime Decline, puts those "theories" to rest.

In my view police serve an important role responding to emergencies like car crashes, people in trouble, disorder, and crime. But those responses are usually after the fact - and none of that can be said to prevent crime when it has already happened!

I'm afraid the simple "theories" like Broken Windows and Routine Activities won't do. We still have to go back to finding and testing more robust theories of crime in our prevention work.

Luckily, with the emerging work on restorative justice, capacity building, community development, and SafeGrowth, I really think we're finally back on track.

Severin Sorensen, CPP said...


Thanks for your blog contribution -- you are doing good work here.

Your article on LV and casino crime opportunity generators caused me to think about another splinter issue that impacts criminal justice model results -- deliberate under-reporting of crime by the stakeholders, whether by law enforcement, commercial business owners, or politicians.

All too often, empirical crime reports, such as the UCR Part 1 and 2 do not capture the all the crime, nor the nuance of crime.

In the case of the casinos and the local police, they have a vested interest not to report crime -- it is bad for business. I suggest that many crime situations are handled "off-the-books" so as to promulgate the perception of it being a safe place. Follow-me on this for a moment.

In the 1990s, I managed a large CPTED program for HUD. On day, I received a call from the national crime office director from HUD to accompany her to Atlantic City, NJ (a smaller sister city of LV gambling on the east USA coast). We were responding to a spate of gang violence shootings and murders that occurred within a few weeks of each other that had occurred in Atlantic City between rival gangs, that happened to have ties (they were living on and around) public housing in the city.
Most interestingly, news of the gang related homicides were not widely reported, nor were the efforts of police to work on the streets to suppress crime and deal with this situation through direct gang communications and negotiations, to stop future homicides from occurring and to keep the situation and news bottled up -- as it would be bad for business and reputation of the city. The methods used to suppress crime were clearly "off-the-books" and any scholar trying to make an assessment of the empirically reported data in UCR to what actually occurred would not have sufficient information to make conclusions; something asymmetric occurred, and the reporting systems were not designed to catch these irregularities.

In such situations, actual crime results reported would not reflect the behind the scenes work to suppress not only crime, but reported crime, as it was bad for business.
Consider now Las Vegas Casinos. Many casinos today are owned and operated by public entities where their stock is traded daily world wide. A simple negative report about crime or violence on these properties can impact public perceptions, business, and negatively impact the share price, and public perception of these properties. Is it any wonder therefore that the anticipated model results differ from the actual data when there is such self-interest to change the outcome?

Another example, in the 1990s, I was doing some security consulting on drug and gang related issues in the Detroit metro area, at a little town called River Rouge. I was culling through the data and found a peculiar set of entries -- "floating bodies." There is no UCR data capture method for floating bodies. I asked local PD what this was, and I was told something to the effect, these are dead bodies that were pulled out of the river -- somebody else's problem happening up river; they did not occur here, so they were not reported as homicides on our books. Consequently, no homicide was reported, and UCR data underreported homicide in this jurisdiction.

In my view, all to often, there is an inherent conflict of interest in law enforcement, business leaders, government leaders, in reporting crime that is increasing. Only when situations are catastrophic or there is money to be gained from grants, do the data seem to come out showing that crime is rising rapidly. There is just too much on the line to keep the reporting factually honest.

So add to your crime analysis, the problem of identifying objective measures from 3rd parties disinterested in the results of the intervention or theory; for where there is interest, there is opportunity for influence and asymmetric action not detailed or captured in the data.


Severin Sorensen, CPP
Sikyur LLC

GSaville said...

Severin, you very appropriately penetrate to the heart of the issue: deliberate underreporting.

In other words, corruption!

When I said it is possible that casino security displaces crime to venues just outside the Strip, I had no idea of the deliberate depth to which you intimate this occurs. A sad affair indeed (though not unexpected).

Perhaps that should be the basis of a real theory of casino crime locations? Wouldn't it be interesting to map the rates of deliberate underreporting and municipal corruption! It would siphon out the decent casino's playing by the rules from those not.

Data collection difficulties aside, I have yet to see a crime mapper with the gumption to take that one on! I guess burglary rates are an easier (and more simplistic) bet!

I love the gauntlet you toss down for crime mappers and analysts: find objective measures to capture that difficult data. As Kennedy said, in all honorable adventures, we don't go because it is easy. We go because it is difficult.

Thanks for your important insights.

Joshua said...

The issue of crime and place vs dealing with the root causes of crime is a serious one at our agency. We are in the throws of getting past the technical obstacles in mapping crime, but what little information we can accumulate at this point isn't telling us anything about who is involved, why crime is occurring or what to do about it. This means that officers are reduced to driving around 'hotspots' in a 'directed' manner, but with no understanding of the problem there is literally nothing for the officer to do. Additionally, calls for intelligence information are going unanswered within the agency. If we don't know the neighborhoods we patrol, how to work collabortively (inside and outside of policing)or how to conduct basic problem solving what are the maps really going to tell us - what we already know - crime is bad in certain parts of town. It feels as though everyone is waiting for the information systems to evolve when in reality the way in which we engage problem areas needs to evolve. Maybe a start is to stop rewarding competitive and individualistic policing efforts which tend to contaminate cooperative efforts.

Josh K.

GSaville said...

Thanks for those comments Josh. Those points are excellent and right on the mark. I too appreciate the power of crime mapping, but like you I also have great reservations about the quality, quantity and appropriateness of most of the crime data. CompStat program in most police agencies not only DON'T solve these issues, I think there are cases where they have made things worse by mis-identifying the nature of crime problems.

The issue gets even worse when you look at some of the "academic" mapping solutions that attempt to use "science" to solve these problems. They slice and dice the mapping data and map configurations with this statistical procedure and that. But in the end they are doing little more than using fancier and shaper types of knives to slice and dice that data...they ignore the fact that they are still slicing stale bread to go into a minestrone soup without the ingredients.

And as we all know, soup without ingredients is just a lot of hot water! That's the metaphor for the contemporary crime mapping world in my view.

Unless we start to develop better causal theories to drive our mapping activities - beyond the symbolic descriptions of routine activity "theory", broken windows "theory", "environmental criminology" theory - we will never move beyond a lot of hot water.

Thanks again for your insightful comments Josh.