« Advice for the US Sentencing Commission from former USDJ Nancy Gertner | Main | How common are DVD submissions as mitigation evidence as part of federal sentencing? »

September 30, 2013

"Does Liberal Procedure Cause Punitive Substance? Preliminary Evidence from Some Natural Experiments"

The title of this post is the title of this very interesting new paper on SSRN authored by the always interesting Donald Dripps. Here is the abstract:

The late, and justly celebrated, William Stuntz made many contributions to the literature on criminal procedure. Among these is the arrestingly counter-intuitive thesis that the Warren Court’s pro-defense procedural rulings made a causal contribution to the “punitive turn” in the substantive criminal law.  This article, contributed to a symposium on Criminal Justice at the Crossroads held at USC on June 7, 2013, and forthcoming in the Southern California Law Review, aims to test this thesis empirically.

Before the Warren Court, criminal procedure was not uniform across the states. Some were more liberal and some more conservative.  The article argues that these differences set up natural experiments.  We would expect the Warren Court’s decisions to provoke more powerful reactions in jurisdictions where local practice was more radically transformed.  We can assess whether conservative jurisdictions increased the severity of the substantive law faster than counterpart jurisdictions with more liberal baseline procedures.

The article measures punitiveness according to an index of prisoners per homicide. It codes eight US jurisdictions as liberal or conservative in their pre-Warren Court criminal procedure.  Generally similar jurisdictions with marked differences in their criminal procedure are then compared: liberal California with conservative New York, liberal Illinois with conservative Ohio, liberal Kentucky with conservative Maryland, and liberal DC with conservative Virginia.  The data in general do not support Professor Stuntz’s claim that liberal procedural rulings encouraged more punitive substance.

Further study is warranted.  The available evidence, however, does not suggest the existence of a general substance-procedure feedback loop that should cause judges, legislators, or law enforcement officials to hesitate to adopt otherwise justified reforms.

September 30, 2013 at 04:49 PM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference "Does Liberal Procedure Cause Punitive Substance? Preliminary Evidence from Some Natural Experiments":


Inmates per homicide seems like a very odd measure of how punitive the criminal law is.

I would have expected something like inmates per conviction or inmate-years per conviction (or perhaps inmate-years per discrete act, though I can imagine that last being difficult enough to calculate as to make the exercise impossible).

Posted by: Soronel Haetir | Sep 30, 2013 5:44:38 PM

Post a comment

In the body of your email, please indicate if you are a professor, student, prosecutor, defense attorney, etc. so I can gain a sense of who is reading my blog. Thank you, DAB