In a 2016 dissent, Justice Sonya Sotomayor described Americans becoming mere “subjects of a carceral state” as a result of repeated Supreme Court decisions broadening the power of the police to stop and to arrest people under the Fourth Amendment. In a powerful new empirical article, The Great Decoupling, Social scientist Vesla Weaver and coauthors have developed stunning empirical evidence of what it means exactly to be what they similarly call a “custodial citizen,” and who is likely to become one. In earlier work with Amy Lerman, Vesla Weaver defined the difference between being defined as a criminal offender by the state’s response to your “behavior” and simply being “‘defined by [your] relationship to the state;’ a relationship predicated more on who one is than what one has done.” In most criminal law and criminology work we presume that the institutions of justice are concerned with “criminal offenders” (a term I normally avoid myself) but Weaver and colleagues strongly suggest it is as “custodial citizens” that the police, and perhaps other justice agencies, look at young people, and especially young Black people.
The authors leverage a unique data source (the National Longitudinal Survey of Youth) that captures self-reported criminal behavior as well as criminal justice contact among advanced juveniles (generally considered by criminologists the most crime prone and most likely to be targeted by police) to examine a question shockingly understudied in both criminology and criminal law: what is the relationship between criminal behavior and criminal justice contact. With a few noteworthy exceptions (Elizabeth Hinton’s recent book being one), many, even critics of mass incarceration (like this one), have tended to assume that what varies in periods of punitive expansion (like the forty year long one that may have ended in the past decade) is the state’s response to criminal conduct (or at least suspect criminal conduct). Regardless of whether one supports aggressive enforcement and punishment policies, almost everyone assumes that there is at least a close relationship between crime and contact. Indeed, in a decently functioning justice system as the authors put it “contact should follow conduct” regardless of how aggressively so. But it turns out it ain’t so, and especially if you are Black. (Their data did not yield meaningful comparison of White-Latinx participants in the surveys.) Even more interestingly the authors leverage two waves of this study that capture similarly aged people in years that reflect the beginning (1979) and the peak of the great punitive turn of the late 20th century.
It turns out the world in 1979 kind of approximated the reasonable norm of contact following conduct. Only 18 percent of respondents who reported being arrested in 1979 reported no criminal activity. By 1997 that had grown to fully 70 percent. Your chances of getting arrested committing no crime in 1997 were higher than your chances of getting arrested while committing crimes in 1979.
Because of the detailed nature of the self-reported data, the authors are able to look at this relationship for varying levels of crime involvement and it holds up pretty much across the level of that involvement. In 1979 your odds of getting arrested only crossed .2 after three to four crimes. In 1997 they reached .2 at 0 and hit .4 at 1 crime. In 1979 that relationship increased at a nearly linear diagonal. To peak at about .8 after 9 or 10 crimes. In 1997 the curve flattened at above .08 after 5.
As Justice Sotomayor recognized, the reality of being defined as a subject to arrest regardless of your criminal behavior is far more common for people of color and in this study in particular, for Black people. Remarkably (and I’m not sure I believe it), the relationship between arrest odds and crime activity was virtually the same for Black and White participants in the survey in 1979. For example, in 1979 both Black and White participants who were at the midpoint of criminal activity both had a .25 chance of being arrested. By 1997 it had exploded to .8 for Black participants (and .6 for Whites). In short, young people of both races are dramatically more exposed to arrest in 1997 they were a generation earlier, but significantly more so for Black youth.
In this Jot, I cannot do justice to the careful efforts made by the authors to assess the statistical and substantive validity of their data, or their use of logistic regression to assess the relationship between race and generation in this relationship. The one inferential issue I would like to have seen discussed is how the crime decline of the 1990s might play into this great decoupling. Afterall, youth in 1997 were less likely to be involved in crime than at any time in the previous two decades including 1979 (which was near the top of the crime wave). Police in 1997 were possibly less likely to find a real crime going on behind their racial profiling based stop and arrest than in 1979 (in both periods they could use petty offenses to back up their arrest in any event).
The very important point of this article however is that the drop in criminal behavior across the country since the 1990s did little to protect youth, especially Black youth, from the disruptive and destructive effects of arrest because arrest is no longer closely tied to actual criminal conduct. The wave of policies we unleashed to increase punishment and police contacts in the 1980s and 1990s effectively decoupled crime and punishment. We have long recognized wrongful conviction of those imprisoned or facing the death penalty as a problem of enormous moral significance and not nearly rare enough (if rare overall). Here at the other end of the system, which is now beginning to be studied more intensively, this important research suggests that the punishment of the innocent is routine. Moreover, the remarkably high odds Black youth in 1997 faced of being arrested without having been involved in a crime strongly suggests that the conflation of crime with Blackness that began in the Eugenic era has actually increased in its grip on American criminal justice.