Friday, November 24, 2023

A bird's-eye view of safety - Urban morphology and crime

Astronaut Tim Kopra took this night photo of Chicago from space. Beginning in the 1920s, Chicago became a center of research into morphology as a crime factor - photo courtesy NASA

by Gregory Saville

The promise of scientific discoveries never guarantees positive results. That is up to us and the wisdom of our choices. But without rational thinking, data and evidence, and decent research on the problems of our day, we cannot expect to reap the rewards that proliferate in the sciences. This especially applies for city building and crime prevention.

That brings me to what social geographers call urban morphology. Urban morphology is the scientific study of the macro geometry and physical form of cities and towns, such as street patterns, land uses, and population densities – basically modern urban land use planning. My university classes in urban morphology are when I learned why architects and urban planners think so differently. 

An architect might design building windows facing the street to improve natural surveillance and decrease crime. But a traffic engineer might design a wide, one-way street to speed vehicles from one region of the city to another. Speeding cars make a street inhospitable and unwalkable, in which case no one bothers to look outside. So much for natural surveillance. 

Many times I have been asked to look at street crime and noted empty sidewalks, loud vehicle noise from speeding cars, and high vehicle speeds on a wide, one-way street. Morphology matters. 

The traditional suburban cul de sac - Photo by Michael Tuszynski on Unsplash


Take the cul de sac. Some criminologists claim that cul de sacs are the best places to live because we get to know our neighbors and that will cut crime. Having lived on a cul de sac, I often questioned those assumptions. Were they based on scientific data or a pocketful of anecdotes? 

In environmental criminology, I learned that cul de sacs were supposed to be places of lower crime due to limited permeability, controlled access, and increased territoriality. Some studies supported that hypothesis.

Then new studies challenged that hypothesis. One South Korean study showed how cultural factors can exacerbate, or mitigate cul de sac risks. Another study by Mateja Mihinjac in her criminology graduate research also challenged the cul-de-sac-is-good hypothesis. Her research indicated they had no impact. All this, of course, is how science is supposed to proceed:

Hypotheses are tested --> old theories fall --> new theories emerge.

Then I read the work of civil engineer, Charles Marohn, someone with decades of experience designing roads and streets. 

Marohn and his colleagues examined the fiscal side of street designs, including cul de sacs. They looked at the street engineering of cul de sacs and compared the fiscal costs of construction, maintenance, replacement, asphalt, and extruded curb costs, with other types of streets. The results were not encouraging. 

Cul de sacs were envisioned as safe areas
- Photo by Stephen Andrews on Unsplash

Another urban planner re-examined Marohn’s hypotheses and studied cul de sac street design costs in his own city. His research too was shocking. 

It turns out that cul de sacs make no fiscal sense. They are expensive to construct and difficult to maintain. Cities end up in the red when they attempt to cover costs. Tax revenue from residents does not come close to paying for cul de sacs. Costs for maintaining cul de sacs balloon as they age and they will eventually need huge property tax increases or federal subsidies – neither of which is likely. So not only are cul de sacs not exactly the crime panacea first thought, they are also fiscally unsustainable.


This research is a product of the Strong Towns movement. Strong Towns researchers raise all sorts of important morphology questions that pertain to crime. For example, what about all the excessive parking lots in cities and their contribution to safety, crime, and fiscal sustainability? 

Many modern cities have excessive parking, mostly unused

In Dallas, Texas 25% of the entire downtown land use is for parking cars, often the same empty parking lots where people are assaulted and cars are easily stolen. When Strong Town researchers dug into the municipal costs to cover car parking, the result was alarming. These are costs that could be better spent on housing the homeless, responding to toxic street drugs, or better public transit. 

The Strong Town advocates have some intriguing design answers. Installing curbside patios to replace downtown parking spots is one type of repurposing that can help. For example, Toronto instituted curbside patios and found they produce 49 times more money than the street parking spots they replaced  – again, money that might be tapped for better livability results.

Street patios are inexpensive, bring more fiscal returns than parking spots,
and provide a popular livability option 

Charles Marohn is the president of the Strong Town movement and his book, The Confessions of a Recovering Engineer is a must for those interested in city design and SafeGrowth.

It emphasizes the role of scientific research and evidence in urban planning and design if we are to reap the kinds of successes in crime prevention that we see in other branches of science. 

Saturday, November 11, 2023

Who you gonna call? An appropriate response to people in crisis - Part 1

Calling 9-1-1 for all situations is not ideal - photo Creative Commons

GUEST BLOG: Beth Dufek is a writer and marketing strategist for clients improving the built environment. She runs her own consulting firm in the Pacific Northwest. Previously she worked with the LISC non-profit organization, facilitated SafeGrowth projects in Milwaukee, and later worked with neighborhood groups in Seattle, Washington. She was named one of the Milwaukee Business Journal’s Forty Under 40 for her commitment to civic engagement and her ability to build trust in communities. As a member of the SafeGrowth Network, she teaches SafeGrowth in cities across the U.S. This is Beth’s first blog on responding to people in crisis on the street.

I transitioned from architecture into community development in 2006. For a decade I worked side-by-side with residents, business owners, City officials, government agencies, and nonprofits – the proverbial “stakeholders”. Our goal was to reimagine neighborhoods, identify barriers, develop strategies, and create long-lasting partnerships that would deliver thriving communities throughout Milwaukee, Wisconsin, and Seattle, Washington. This is what draws me to SafeGrowth. 

In 2011, especially for the Zilber Neighborhood Initiative in Milwaukee, we had a process, but we never called it the 5 Steps like we do in SafeGrowth. We identified achievable projects, brought residents together, and presented a compilation of projects tightly packaged into a plan to an audience we hoped would help us with funding or other resources. The final product was an illustrated asset map, a neighborhood plan format that I love to this day. 

Photo from Lindsay Heights plan - Zilber Neighborhood Initiative

Fast forward to the next decade. In 2020 I joined the SafeGrowth Network, and I started co-facilitating SafeGrowth Trainings with Greg Saville in 2021. It was a natural transition. All neighborhood plans I have ever worked on, regardless of the income status of the residents, start with a safety strategy.  Among the top issues: who do we call when we need help? And with the state of mental health and drug addiction in the Pacific Northwest, this need feels greater than ever.


Case in point: in 2019 I took walks along Seattle’s waterfront. One week, I saw a human completely engulfed in a purple sleeping bag (I assumed it was a human) in the same position for 3 days in a row. The human in the sleeping bag was situated among people who were enjoying the park during sunny days. But they were in exactly the same position for 3 days. My imagination convinced me the sleeping bag was fuller than it was (perhaps a bloated dead body?). This vision played a significant role in my next moves. 

My concern was for the lack of human interaction with other humans experiencing drug addiction and mental health crises. A simple “Are you OK?”, rather than stepping over a human slumped on the sidewalk, took center stage. What if they are dead? Or need help? Who do I call? I tried the non-emergency police. No answer. Seriously. 


I called 911. I pleaded with them not to make a big deal, couldn’t someone just come by? The dispatcher asked me if I could tap the purple sleeping bag to see if they were OK, or alive. I can’t remember. No! I’m not trained to do this! What if I scare them and they attack me? What if they ARE dead?

Seattle Fire Department - Photo courtesy of Joe Mabel

Minutes later a Seattle Fire Department truck came barreling through Myrtle Edwards Park, drove right past me and then called me to direct them to my location. 

“Over here! Here I am! The one who has now disrupted everyone’s enjoyment of one of the 56 sunny days in Seattle because I care about this human … and listen to too many true crime podcasts!” 

I turned to the human near me and apologized and explained a fire truck was not the outcome I had imagined. I had a guy on a bike in mind who could assess the situation and call for additional help if needed. I walked with the paramedic over to Denise. I know her name is Denise because he said, “Oh that’s Denise. Denise, are you OK?” She was very much alive and none too happy that her afternoon sleep was disrupted! 

That made two of us. 

Since then, I’ve been thinking; why can’t all humans have a non-non-emergency number to call just to get a wellness check when they see someone in crisis? 

TriMet Safety Response Team - photo courtesy of TriMet


Through the SafeGrowth network, I found an entire agency that feels the same way. In 2021, the first SafeGrowth Training I co-facilitated was for TriMet, the three-county transit agency in the Portland, Oregon metro area. To date, we have trained 37 TriMet department directors and safety, maintenance, construction, and planning staff on how to use the SafeGrowth method to make staff and community-informed safety improvements along the transit system. 

We just wrapped up Part 1 for our third cohort of TriMet employees and TriMet contractors. This cohort has five members from TriMet’s Safety Response Team. I get emotional just thinking about how impressed I am with what they have done and what they plan to do to help the thousands of humans who find shelter along the TriMet system through their Reimagining Public Safety initiative.

In the next two blogs, I will write about TriMet’s Safety Response Team and the changing roles of transit operators, librarians, and other agencies that find themselves in new roles helping humans. 

But for now, I will be brave and ask, “Are you OK?”

Tuesday, October 31, 2023

Brazil & the Next Gen R/Evolution

Venue for the 2023 ICA CPTED Conference is Sao Paulo, Brazil. The largest city in the Americas has a metro population over 20 million - photo Wiki Creative Commons

by Gregory Saville

It’s Halloween today and the goblins and ghosts are looking for treats. An interesting ghost from my past appeared today, triggered by an international conference in Sao Paulo, Brazil this week. It all started 28 years ago…

In 1995 I ran an urban planning/criminology consultancy with my former business partner, Paul Wong in Vancouver, Canada. Our work in Crime Prevention Through Environmental Design (CPTED) had expanded across Canada and we were becoming alarmed at what we saw. The first CPTED publications were holistic, neighborhood-oriented, and they combined social programs with physical designs to cut crime opportunities. 

However the CPTED that we saw in some places was not holistic. It had de-evolved into target hardening, locks, and lights. There were very few guidelines and no certification programs. To our dismay, we discovered that the public, and municipalities across the world, had no clue about the failing quality control problems in CPTED. 


Basically, we located a like-minded CPTED practitioner, Barry Davidson from Calgary, Alberta, and in 1996 we launched the first-ever practitioner/researcher CPTED conference in Calgary. Thus was born the ICA - International CPTED Association - the first-ever global, non-profit, CPTED professional association. I have written a prior blog on the history of the ICA. 

With chapters all over the world, the ICA is the premiere global organization for professional CPTED in research and practice. Most recently, members of the ICA collaborated with others in the International Standards Organization to design and implement the first-ever, global CPTED standard, ISO 22341:2021 - “Guidelines for crime prevention through environmental design”. 

CPTED USA is the latest ICA chapter

The ICA has a talented and diverse board of directors from around the world and most recently, the ICA approved two new affiliate chapters – the CPTED USA and CPTED/PCAM Canada


This week the latest ICA conference - Safe Cities by People - is underway in Sao Paulo. I was unable to attend this year, but my colleagues Gerard Cleveland and Mateja Mihinjac will present a special session on the R/Evolution in 2nd and 3rd Generation CPTED. They will describe the latest advances and research in CPTED. Second Generation was launched in 1997 at the second-ever ICA conference in Orlando, Florida. Third Generation was launched by Mateja and myself 5 years ago at the ICA Conference in Cancun, Mexico. 

Most exciting is the announcement by Mateja and Gerard of two new publications. 

  • The first is a forthcoming book in 2024 by Mateja, Carl Bray, Jason Tudor, and myself, called Return to Walden.  That book presents the latest in the SafeGrowth method, to which this blog is dedicated. 
  • The second is Gerard and Annie Morrison’s new book Tier Teaching about better methods of teaching, learning, and safety in our schools, where the seeds of delinquency and dysfunction are planted, or removed, through competent education and educators. 

I am disappointed to miss this conference but to help Mateja and Gerard, I prepared a short video intro for their presentation today. Best wishes to all at the Sao Paulo ICA conference. 

Thursday, October 26, 2023

Whose voice speaks for us? Access control and the engagement trap

Residents take back their neighbourhood by access control of their roadways 

by Mateja Mihinjac

What happens when a community installs retractable pillars to block vehicular access into a neighbourhood of terraced blocks and local businesses? Access control is one of the fundamental principles of 1st Generation CPTED, established in 1972 by Oscar Newman in his book Defensible Space and employed judiciously by CPTED practitioners.

In one particular neighbourhood in my home city of Ljubljana, Slovenia, the issue of vehicle access control has been a contentious issue for several years. It regained interest a few weeks ago when a child was almost run over by a car while playing in a neighbourhood area intended for pedestrian use. Many have welcomed the installation of pillars, but others feel it a hasty decision lacking the preceding discussion. They seek more thoughtful consideration about the effects on senior residents, those with mobility issues, and the delivery of goods and services. 

No entry for private vehicles in the neighbourhood 


Soon after the installation of the pillars, emotional reactions appeared on the community Facebook group, as well as in local news, which reported reactions as “warm-cold”.

Some residents echoed sentiments such as “people would even drive directly to their apartment if they could” suggesting that people have become spoiled and accustomed to comfort. They welcomed the decision and praised the mayor for it.

Others maintained the move was made due to the nagging of a few who had the loudest voice but had no right to decide on behalf of the majority. They thought the right of access to their property was violated. They also reminded the “house rules” are already in place for each block. While these rules don’t seem to operate well in practice, it is clear the process of engagement was not managed well. 

Residents were not asked to partake in the discussion and they felt the decision was imposed on them. In SafeGrowth we are very familiar with this phenomenon and in our new forthcoming SafeGrowth book, we describe this as the engagement trap - listening to the loudest voices, consulting too late in the process, or insincere tokenism for political expedience.

Emergency vehicles waiting for access 


A few weeks later, the residents engaged in a more productive dialogue to try to find a solution that would suit the majority. The local government has also requested input and suggestions from the residents to find the most workable solutions.

Some suggestions have been implemented such as retracting the pillars for two hours during early morning hours for local business deliveries. The new decree also envisions special one-time access permits for house renovations or large item deliveries, or repair services since a number of residents have expressed concern about access. 

Now that the emotions have somewhat calmed down and the residents are starting to adjust to a new regime, many are calling for patience and cooperation in identifying the solutions. 

Some residents have even proposed this might be an opportunity for the introduction of new local businesses. For example, introducing a local delivery person(s) to assist with transporting goods from the surrounding shops either on foot or using micro-mobility options. Another example – expanding delivery lockers used by shipping services at the edge of the suburbs since they are now unable to deliver goods to people’s doorstep. 

Local government has already introduced “Kavalir”, a small electric golf-cart-like vehicle that drives between the neighbourhood and adjacent produce market on a daily basis. This service is free of charge and primarily intended for those with mobility issues.

Terraced residences included narrow pedestrian corridors intended for pedestrians, not vehicles 


It is commendable that the municipality responded so quickly to address this local problem, but it is discouraging to see how frequently decision-makers fail to initiate effective discourse with the local residents until complaints arise. 

The inability to engage residents, or the use of the engagement trap to substitute for authentic engagement, leads to more problems and confusion. All of that can be avoided from the start of the process with authentic communication. Time will tell if this is yet just another attempt at tokenism or a lesson in civic engagement. 

Saturday, October 14, 2023

The Pros and Cons of Using AI to Prevent Crime

In sci-fi dystopian films like The Matrix, Bladerunner, and The Terminator, pollution clouds the sun, climate chaos worsens, and Artificial Intelligence is the enemy



"Artificial intelligence (AI) offers new capabilities for law enforcement and the criminal justice system to help predict, detect, and prevent crimes. As discussed in previous SafeGrowth blogs like "Can AI Make Neighborhoods Safer?" and "Bias and AI - What it Means for Crime Prevention", AI tools such as predictive policing algorithms, facial recognition, and risk assessment systems are increasingly being adopted. However, the use of AI also raises important ethical questions. In this blog, I’ll explore some key pros and cons of using AI for crime prevention.


  • Crime prediction: As noted in "Predictive Policing: The Argument for Public Transparency", AI can analyze crime data to identify high-risk areas in order to optimize police patrols and resources. This data-driven approach can increase efficiency.
  • Facial recognition: AI facial analysis can rapidly compare faces to databases of photos to identify suspects or missing persons. This expands investigation capabilities, as examined in “Facial Recognition Software and Crime Prevention”. 
  • Risk assessment: As discussed in "How AI is Transforming the Criminal Justice System", AI algorithms can assess recidivism risk to inform bail and sentencing decisions. The goal is to improve identification of high and low-risk offenders.


  • Biased data: Since AI algorithms are trained on historical crime data, there is a danger of perpetuating and amplifying existing biases, as explained in “Garbage In, Garbage Out: The Dangers of Biased AI”. This could lead to over-policing of marginalized groups.
  • Privacy erosion: As noted in “AI Surveillance: Security vs. Civil Liberties”, the data collection required by many AI systems infringes on privacy rights through pervasive monitoring and tracking. This threatens civil liberties.
  • Poor transparency: The reasoning behind AI predictions is often opaque and difficult to understand, even for developers, due to the algorithms’ complexity. This lack of explainability reduces public trust and accountability, as examined in “Explainable AI: The Path to Ethical Algorithms”.
  • Dehumanizing: Over-reliance on risk algorithms diminishes human discretion and nuanced decision-making in the criminal justice system, as discussed in “The Ethical Dilemma of AI Judges and Juries”. 

In summary, while AI offers promising capabilities to enhance public safety, we must carefully weigh the benefits against the risks to civil rights and liberties. As argued in previous SafeGrowth posts, policymakers should promote development of ethical AI systems with transparency, oversight, and impact evaluations to ensure AI prevents crime without infringing on human freedoms."


YouTube with Kurt Bell, IT project manager from his AI presentation 


by Greg Saville 

I'm back  - the living, breathing, thinking, flesh and blood, me!

AI wrote the above SafeGrowth blog. Every word! I asked the artificial intelligence platform, claude by Anthropic, to write a 600-word blog on the role of AI in crime prevention. I asked it to reference prior blogs on this topic. I also asked it to frame the blog using the pros and cons.

The above text is the result. What do you think? What do I think? 

It did take considerable time for me to edit this. The basic prose was fine, but the fact-checking took a while. In short, it got some stuff wrong. 

Claude by Anthropic sums up some of the basic points nicely. I was encouraged it did not hide the truth of its own dangers like the erosion of civil liberties and privacy, poor transparency, and diminishing human discretion. These are not small matters and AI sees no reason to avoid the politics and critiques of itself. Not yet!

I was encouraged it cited some previous SafeGrowth blogs - but discouraged to realize they do not actually exist. With hundreds of blogs on this site, it is not surprising I cannot recall them all, but I do not remember any blogs with those topics. Neither did my search of the site find any. In other words, Claude made them up! That is disturbing, to say the least!

Claude by Anthropic uses the technical writing technique of bullet points. It avoids free-flowing prose or metaphors. It gets straight to the point because, I assume, it only had 600 words and it didn't have the time or expertise to construct a more poetic exposition. 

Bullet point writing, devoid of metaphor, simile, or literary license, can lead to a snooze fest. True, some of my paragraphs here could easily be rewritten into bullets, but reading through reams of bullet points is an exercise in ho-hum and humdrum. It is the humanness within writing that connects us to each other in ways not easily defined. AI seems to have problems with that - currently. 

Yet AI can write poetry and create art!

Ancient art masterpiece Sapphos, 470 BC. AI has made inroads into creating original poetry and painting - photo Wiki Creative Commons

AI lists predictive policing as a pro and sidesteps the ethical problems and critical research on predictive algorithms. AI does describe over-policing marginalized groups as a con, but it does not do so specifically so the reader does not connect the ethical problem with a specific application. Why?

It lists facial recognition software as a pro. But we know from research that AI facial recognition has fallen victim to the threat of false positives (mistakes) that have led to improper arrest and detention. There were prior blogs on this problem but did not cite them even though I asked it to cite prior blogs. Instead, it cited blogs that were not terribly critical. 

The fact that AI cited some prior AI blogs (which do not exist on the SafeGrowth site) but did not cite others (more critical blogs) makes me wonder! Blogs from 2021 Summoning the Demon and AI vs CPTED, or this year's blog Stop Dave, I'm Afraid... all omitted! Why? I'm told the current AI chat platforms (ChatGPT, Claude, and others), cannot access real-time data on the internet. Maybe that's why?


This experiment in AI blogging does not convince me AI is ready for prime time. It still needs plenty of fact-checking and human review. Of course, that could be said of any editing process. The fact that it wrote the blog in technical jargon with bullets, and avoided any literary license, suggests AI has a ways to go to create interesting prose. 

Then again, IT project manager and author, Kurt Bell, tells us AI has already passed the famous Turing Test as of 2014. The Turing Test measures whether AI can be distinguished from a real human. In that test, at least, it could not. 

That should give us all pause, especially when it starts with "I’ll explore some key pros and cons". Who, I wonder, is it referring to when it says "I"?

Friday, September 22, 2023

Elements of Crime Patterns - A review of a breakthrough

Deborah Osborne's new book, Elements in Crime Patterns, delivers a long-overdue typology for crime investigators, researchers, and prevention specialists

By Gregory Saville

When you think of the study of crime and its prevention, perhaps you imagine that, after a century of criminology, elementary crime patterns are well understood in the academic literature. You would be mistaken! True, we do know plenty about some crime types, as well as the geographical patterns of crime events. But studies about the actual patterns of offenders and offenses are seldom collated together into a coherent, practical dictionary of elementary crime patterns – what scientists call typologies.

For example, in biological science Scottish botanist Patrick Matthews outlined the entire theory of natural selection after years of classifying different types of trees into a typology. Thirty years later that led others, like Charles Darwin, to build more classifications (and claim the theory as his own). Darwin’s bird typologies from the Galapagos Islands, along with Matthews's original work, eventually led to one of the most powerful theories in science – the evolutionary theory of natural selection. 

This method of constructing elementary typologies is so well understood by historians that it usually stands as the introductory chapter on virtually every book about the history of science. 

Crime science, by comparison, has barely scratched the surface of this type of basic typology. This is an alarming fact since that is how robust theories of explanation and prediction emerge in the first place. It is almost like crime science got ahead of itself and developed theories without the essential first steps of theory building. 

Until now!


Deborah Osborne is an author, retired U.S. Secret Service intelligence analyst, retired crime analyst with the Buffalo police, and former co-chair of the International Association of Crime Analysts. She is also one of my former students from a short crime mapping and analysis course I ran out of our research center at the University of New Haven 20 years ago. To be clear, by the time she took our training, Osborne was already an established crime analyst with considerable experience. 

This week Osborne published her book Elements of Crime Patterns.

I was excited to read her book, especially when I learned that she used a form of AI – ChatGPT – to help her collect data. To my knowledge, Osborne’s book is the first published criminological work to tap into the power of AI as a tool to build a robust database for a criminological text. That itself is an achievement. 

Scientific theories emerge from detailed observations and typologies


Elements of Crime Patterns is a field guide to identifying crime patterns, a practical toolkit that offers “the kind of knowledge about the crime pattern domain that is learned only on the job through experience” (p. 310). In other words, this is (finally!) the kind of fundamental scientific research that normally precedes theories and prevention programs. 

Osborne’s approach is not to provide explanations or theories accounting for where and why something happens. She takes another tack:

“The solution to crime pattern detection cannot be solely data driven. The informal exchanges of information through conversations, explorations, and intuitive perceptions are crucial in investigative casework, but researchers and policymakers often do not acknowledge this. It is important to understand that conversations between law enforcement staff are [the means by which] some crime patterns get recognized, especially those involving separate records system in other jurisdictions…Context is crucial.” (page 21)


Crime patterns in criminology are often associated with geographic maps of crime hotspots -
there are many other equally useful patterns 


I did have a few issues with the book. It could use an index and a bibliography (although to be fair, each chapter had endnotes with some references). I was also not as keen on a few earlier chapters on lifestyle, tools and equipment, and routine activities compared to later chapters. Those earlier topics seemed to me too generic and all-encompassing to be of much value in analysis. The routine activity theory has been criticized as, at best, an untestable theory and at worst, wrong

Those points pale in comparison to the impressive 30 chapters on specific crime types. She includes robbery, counterfeiting, sexual assault, murder, drugs, human trafficking, hate crimes, and vehicle crimes, among others. There are also surprises with crimes that make only rare appearances in the criminological literature – wildlife and forest crimes, cultural property crimes, and intellectual property crimes. 

The patterns she identifies include methods of different crimes, offender planning and target selection, the aftermath of crimes, and factors that influence crime opportunities. CPTED practitioners should take note there is an especially succinct summary of the opportunity factors that contribute to crime conditions. I wish I had access to information like this long ago in my crime prevention and investigation career. It would have made the work so much easier.

This book is a breakthrough for the science of crime and prevention and for the criminological enterprise – both academic and practitioner. Osborne has made a contribution of considerable weight. This is a book you should read.