Why Can’t Reformers Speak the Same Language?

Print More
reform protest

Protesters rally in Philadelphia, June 2020. Photo by Joe Piette via Flickr

At some point around the 40-year mark in a career spent representing indigent defendants in urban courts, I started to receive invitations to meetings about Criminal Justice Reform.  This might be explained by the fact that I had received a Visiting Fellowship at the National Institute of Justice, and my new hosts may have (mistakenly) believed that in some obscure way I was connected with awarding grant money.

I never reached the point of feeling comfortable at these gatherings.  Everyone seemed to know everyone else.  My new companions were considerably better educated and better groomed than myself and my usual crowd.

But I was always impressed.  The talent, seriousness and dedication of the attendees were striking. The organizers were uniformly scrupulous about mustering diverse sets of participants and including community voices.   And everyone spoke a common language of Reform.

Or at least everyone used the same words.  After a while, I started to wonder whether everyone meant the same things by those words.

It would be better if we did—if the same words captured the same ideas every time anyone used them.  That isn’t impossible.

Here, at any rate, are some meanings I think we might try to share.


The first word to nail down is “safety.”

For more than half a century, the language of criminal justice in the United States has been focused on control.  Most public debates take place on the spectrum captured by Herbert Packer’s famous Crime Control Model (control the people, control the streets) versus Due Process Model (control the officials) dichotomy.

But the language of contemporary reform reflects a recognition that control is at best merely a means to an end, and that a mistaken fascination with control seen as an end in itself sets in motion an inexorable drift in the direction of mass incarceration and, ultimately, divided communities.

Reform discussions now consistently recognize that the real goal of the justice system should be not simply “control,” but  “safety.”

This can be a crucial shift, but its impact will be muted if we don’t make it clear that every time we talk about safety we are talking about everyone’s safety.  We mean that we want innocent defendants like Ronald Cotton to be safe from wrongful convictions, but we also want potential victims of future crimes to be protected too.

We want George Floyd and Eric Garner and Breonna Taylor to be safe from police violence, but we also mean that we want patrol officers to be safe from armed and inadequately treated psychotics attempting “suicide-by-cop.”

Traffic stops are dangerous to citizens, and particularly dangerous to minority citizens, but the data indicates that they are dangerous for cops too. In talking about how we distribute them (or don’t), the safety lens allows us recognize that a stop implicates everyone’s safety.

Once you do that, then working to prevent recurrences can happen on common ground.

When we say “safety,” we can also be clear that our version of  “safety” includes safety from the whole range of harms.   This range includes “iatrogenic” (caused by the treatment) harms like unnecessary fees and fines and promiscuously distributed criminal records.

It includes safety from the pain and grief suffered by victims’ families and from the vicarious trauma suffered, for example, by Black citizens distant—in either place or time, or both—from the immediate scene of fatal police shootings. Safety in our usage should always include safety from the chronic and cumulative harms inflicted by “high frequency/low impact” practices such as the profligate use of stop and frisk tactics and pretrial detentions.

“Safety” describes an environment where communities can flourish.

But most importantly, our references to “safety” will always incorporate the extensive work on the concept done by other high impact enterprises like aviation and medicine. The reform view of safety does not see it simply as the absence of known tragedies; it realizes that the first tragedy could come tomorrow.

As in modern aviation and medicine, “safety” here acknowledges that we are operating within a complex socio-technical system.  We don’t confront a world of dangerous single causes that produce inevitable results; we have to deal with a constant, dynamic swirl of conditions and influences—racism, inequality, budget cuts, media hysteria, caseloads, political profiteering—that bend the probabilities.

Achieving this version of safety means that everyone has to accept his or her individual responsibility for a just collective outcome.


Accountability means that rule-breakers must face sanctions, but it should always mean more than that.

One result of accepting the modern definition of safety is that it forces you to recognize that while the punishment of the system’s bad actors will be necessary, inflicting punishment on a lone “bad apple” is always a terrible place to stop.

Ask who is “accountable” for a bad outcome, and the answer will always be everyone involved to one degree or another, if not by making a mistake, then by failing to anticipate and intercept someone else’s—often by just doing nothing.

In this context, “everyone” includes people far from the scene of the event who set the budgets, created the pressures, developed the legal architecture—people who built the environment that shaped the “bad apple’s” decision to zig when he should have zagged.

Even when there is a clear villain involved—a trigger-happy cop, or a prosecutor who hides evidence—someone (or many “someones”) hired him, assigned him, supervised him, or failed to constrain him.  Someone devised his incentives.

Certainly, he or she could have been influenced by the unlikelihood of discipline, but that influence was one influence among many.  Every failure is a system failure.

The “account” in this sense is a debt that must be paid for a damaging action, but it is also a story that must be told: a story about a full system crash and everyone’s role in it.

Without that story we will let it happen again.

We have to be careful that in our usage “accountability” always encompasses this “forward-looking” perspective. The impunity enjoyed by violent cops and crooked prosecutors is infuriating, but we can’t let that blind us to the fact that they acted—or failed to act—in response to conditions and influences that others created and that we can change.

We will have to hold ourselves (and each other) accountable for continuous learning.


During the course of a methodical (and just slightly gleeful) demolition of Herbert Packer, John Griffiths pointed out that Packer’s Crime Control and Due Process models were not really two models at all; they were simply variants on a single Battle Model of the criminal process—that the only difference between them was who won.

Take a look at the swelling commentary on “warrior cops” confronting the “warrior defenders” of “Gideon’s army” and you may agree that Griffiths was onto something.

The Battle Model that dominates our thinking is built on the vision of an irreconcilable conflict between the state and an individual:  a distinct and particular kind of individual—a “criminal” who is not just different in degree from the rest of us, but different in kind.

The issue is the excision and exile of these organisms: one side wins this struggle, the other side loses.

When you see things that way, “reform” inevitably becomes a matter of adjusting where the balance of advantage lies in the zero sum battle.

This is not what we should mean by “reform” today.

Developing and then evaluating new rules, strategies, and policies for other people to execute is congenial work for Reformers, but for real reform, we need to renovate not just strategies, but  a whole culture.  Culture, as the business school cliché has it, eats strategy for lunch.

Griffiths showed an example of this when he illuminated the Battle Model’s ideology by contrasting it with another institution in society which regularly administers punishment: the family.  The family is built on the understanding that the disciplinarian parent and the wayward child will reconcile: “A parent and child have more to do with each other than obedience, deterrence, and punishment, and any process between them will reflect the full range of their relationship.”

The effort by the early 20th century juvenile justice movement to impose the Family Model orientation on the system’s Battle Model culture was a debacle.

So “reform” now is not about new rules and more draconian punishments for violations; it is about implanting and nourishing a resilient  “culture of safety” on the streets, in the courtrooms, in corrections, and in collaborative efforts with the educational and public health systems with which the criminal system constantly interacts and “co-produces” safety.

This reform is one that will happen on the frontlines, one located in the lives the practitioners and communities lead, not in the programs Reformers devise.

Can frontline cops, or defenders, or community members be empowered to “stop the line” when they see something going wrong?  Can we build a platform for the routine non-blaming, all-stakeholders, forward-looking event reviews we need to learn?

And when using the word “reform,” our Reformers will have to agree to actively reject the idea that the term refers to something that we, Reformers, do to you practitioners.

Above all, somehow the word should convey a little humility: a sense that the Reformers recognize that “work-as-imagined” in their sessions will never be exactly the same as “work-as-experienced” on the frontlines, and that no “fix” they devise is permanent, but will be under immediate assault.

james doyle

James Doyle

“Reform” means collaborating to make things better. Like safety and accountability, reform happens locally (not exclusively in Washington and New York), continuously (not just when Reformers convene or publish).  And it happens—if it happens at all—with communities and their practitioners, not to them.

There are other terms we could talk about. “Restorative,” “transparency” and “predictive” come to mind.

But these three will get us started.

James M. Doyle is a Boston defense lawyer and author, and a regular columnist for The Crime Report. He welcomes readers’ comments.

Leave a Reply

Your email address will not be published. Required fields are marked *