Why Does Reform Fail?

I have spent my career developing and implementing criminal justice reforms, and I agree with Megan Stevenson’s intuition that criminal justice reformers too often make the mistaken assumption that discrete interventions can engineer social outcomes. At the same time, the lessons I’ve learned lead me to take issue with her analysis and recommendations.

I can’t think of a reform I’ve worked on where the process wasn’t as, or arguably even more, valuable than the outcomes it produced. For instance, when I led the state of Illinois’ public safety grantmaking and research agency from 2015-2019, my colleagues and I created a multiyear funding opportunity for medium-sized cities to implement evidence-based programs to reduce gun violence. The award also required grantees to meet regularly with my staff and subject matter experts to talk about their experiences. While the funding ended several years ago, I still hear from people who were part of the program. They talk not so much about the outcomes their work produced but about the relationships they built with experts they otherwise would never have met, what they learned about working with state grantmakers and researchers, and what they learned about better engaging their community.

This is the kind of value we miss when we focus too narrowly on an instrumental conception of “what works”: The very search for solutions and the faith that we can realize them enables people to build trusting relationships and learn from each other in ways that are only possible through collaboration. While these efforts may not always lead to quantifiably evaluable, replicable or durable change, they can arguably contribute to something greater: actions that cultivate shared values of mutual respect, trust and legitimacy that are essential for a safe, just, democratic society.

At the core of Stevenson’s conceptual critique is her assertion that criminal justice reform is based on the “engineer’s view,” the myth that researchers can develop and evaluate interventions that produce lasting and replicable changes by using scientific methods to map out and manipulate the social factors that cause people to engage in crime and violence.

To explain the inherent problems with the engineer’s view, Stevenson argues that the social world is charged with socioeconomic forces that she calls “stabilizers.” The stabilizers influencing individuals targeted by criminal justice reforms include poverty and the legacy of racism. Stabilizers have what Stevenson likens to a gravitational and tide-like influence on people, and their power will defeat most efforts to change them. As Stevenson puts it, “Any barriers to success that were readily moveable had already been moved — by people themselves and their communities.” Individual people and communities may change their own trajectories, but there is a fundamental equilibrium that concerted policy efforts cannot move. So, Stevenson argues that the only reliably certain policy responses are forms of direct interventions like cash assistance programs, because they provide “a straightforward and obvious way to ameliorate harm” by “simply giv[ing] people what they need.”

While this vision of social structure illustrates why most criminal justice reforms fail to change social outcomes, it does so by reinforcing a profoundly engineer-like and, I would argue, asocial understanding of the causes of human actions. Indeed, rather than “opening up doors to new ways of thinking” about the dynamics of social change, it reduces social change to people’s interactions with mechanistic forces that operate just like the natural forces of physics.

This conception fails to reckon adequately with the fundamental differences that distinguish objects of natural science from human beings who constitute the social world. Actions in the social world are always influenced by how we understand the meaning of what we’re doing, the values we hold, the ends we pursue, how we relate to a plurality of others (do we regard others as friends, enemies, trustworthy, incompetent, etc.) and vice versa. Unlike the mechanical processes that determine objects of natural science, this two-way, intersubjective dynamic shapes who we are and what we do.

Taking intersubjectivity seriously can help explain why criminal justice reform’s efforts to engineer social outcomes often struggle to achieve their intended goals — but also why we must keep trying. One common reason for failure might be that reforms tend to presume to know what people are thinking, rather than asking and meaningfully engaging them. What if the presumptions are wrong? What if people resist otherwise sound reforms because they feel disrespected by the people implementing them, even if they are acting with the best intentions and mean no disrespect?

Accounting for intersubjectivity should also lead us to be skeptical of whether even seemingly simple direct interventions like cash assistance will easily or necessarily work as intended. It’s tempting to assume that implementing this kind of program would be just like turning on a light switch. But that’s never the case with any efforts that involve multiple people or benefits that have social significance. For a direct assistance program to work, program staff need to think hard about how the way they deliver assistance might negatively affect the people they are trying to help. For instance, what should program staff do in cases where there’s evidence of domestic violence or other forms of harm or exploitation? Should the program be concerned about how benefits might impact people’s self-esteem or desire to find work or engage in other prosocial activities? These are hard and complex questions that are rooted in how we understand and interact with each other. And there’s no guarantee we’ll ever be able to get to them with an engineer-like precision.

More fundamentally, accounting for intersubjectivity should lead us to question whether even the most rigorous empirical research is capable of understanding and judging the value of criminal justice reform. Of course, outcomes matter, especially if there’s research that demonstrates a particular kind of reform causes harm. But beyond preventing bad outcomes, in my experience, research that narrowly focuses on “what works” tends to distort the unique intersubjective value of criminal justice reform.

At its best, the aspirations of criminal justice reform serve as an important reason to bring different stakeholders together to work for the common good. I would argue that Stevenson’s work taps into these same values. While I have critiqued parts of her argument, I have done so with respect for her insights and the courage she has shown challenging some of the most fundamental assumptions of her own academic discipline. If there’s value in the argument I have made, it has come from wrestling with the issues and questions that she has raised.