O'Toole's Response to "Correcting the 'Crimogenic' Crowd"
By Randal O'Toole
February 21, 2005

Laurence Aurbach takes a long time to say a few things, most of which are wrong. I don't have time to respond to all if his post, but I can respond to a couple of statements.

Stephen Town and I "flagrantly twisted and misread Newman's findings." Sorry, but it is Aurbach who is twisting or misreading those findings. Shortly before he died last April, Newman emailed the following to Town:

I am not very impressed with the work of the New Urbanists. It is nostalgia--a throwback to the past, with little thought about what made those environments work then (long term occupancy by an identical economic class and ethnic group), and unworkable today (occupancy by itinerants, mixed incomes, and ethnic groups).  "Natural surveillance" is not automatically created by high density environments. This is the same battle I had with Jane Jacobs a few decades ago. Unless the grounds around a dwelling are assigned to specific families (or small groups of families) and are understood as belonging to them, those families will not take care of these areas or screen them for activities these families determine as acceptable. There are  many incidents of dozens of people simultaneously witnessing a crime (robbery, rape, assault) in a "public" area adjacent to there dwelling without even calling the police. Assign that same area to specific families and they will guard it as their own and control the activity within it.

Aurbach also says that Newman favored a form of "natural surveillance" that was equal to Jacobs' eyes on the street.

Here is what Newman wrote: 

Jane Jacobs started as a newspaper reporter in New York and then wrote a book called: The Death and Life of Great American Cities (or some such). Her theory (if you can all it that) is that if you design a housing environment which is dense and has many people looking down on the street, the street will be safe because of the many eyes on it. Her theory was disproven when we had a major incident in New York--the Kitty Genovese murder--in which over a hundred people witnessed the beating to death of a woman in the street by a stranger and not only did not intervene, they didn't even call the police. The reason: they did not identify the street as theirs.  Territorial definition of environments (through the subdivision and assignment of spaces) 

In short, the key is "territorial definition of environments," not eyes on the street. Private ownership is one way of doing this. Newman specifically stated that "the residential environments they (the New Urbanists) are creating are very vulnerable to criminal behavior, unless, of course, these environments are exclusively occupied by high income groups--which, by their own definition, they are not."

For some reason, Aurbach diverges from the main point of the article -- crime -- to address some other issues. Yet he does no better on these issues.  For example, Aurbach claims that the CNU web site does not say "All development should be in the form of compact, walkable neighborhoods." This was on a page called "New Urban basics." While CNU may have deleted it from its web site, it is on several other sites with attribution to CNU, including http://www.ub.es/escult/docus2/NEW_URBANISM_%20BASICS.doc

Aurbach makes the conventional New Urban argument that New Urbanism reduces congestion. He confuses reducing per capita driving with congestion, which is not the same thing. He claims that "even Wendell Cox" says that increased density reduces per capita driving by 40 percent, but Cox will tell you that the increases in density needed to reduce per capita driving are huge -- 10 times the average density of most U.S. urban areas. 

Aurbach also claims that "credible research is now appearing that indicates that suburban developments may not be the health panacea they have been sold as." I have read all the research that claims to find this result and it is anything but credible. It is based on extremely weak correlations between questionable sets of data. Even if the correlations were strong and the data were reliable, correlation does not equal causation. 

In short, Aurbach fails to contradict the key points of our article: 

* Research by police in England has shown that New Urban design significantly increases crime. 

* While there is no equivalent study of New Urbanism in the U.S., research by Newman and other criminologists on specific aspects of New Urbanism, such as mixed uses and alleys, shows that they make residences more vulnerable to crime.   

Randal O'Toole
Thoreau Institute


1. Correcting the Crimogenic Crowd
2. O'Toole's Response to Aurbach
3. A Reply or Two to O'Toole
4. Related Materials on Safety and Neighborhoods