DA Gascón and What “Studies Show”
Do “studies show” that longer prison sentences increase recidivism?
Los Angeles District Attorney George Gascón’s Special Directive 20-08 says:
While initial incarceration prevents crime through incapacitation, studies show that each additional sentence year causes a 4 to 7 percent increase in recidivism that actually outweighs the incapacitation benefit. 1
—————————
1 Mueller-Smith, Michael (2015) “The Criminal and Labor Market Impacts of Incarceration.”, available at https://sites.lsa.umich.edu/mgms/wp-content/uploads/sites/283/2015/09/incar.pdf.
Note the plural, “studies show.” Despite the plural, Gascón has never, to my knowledge, cited anything other than this one unpublished manuscript. In his opposition to the ADDA writ petition, he merely referred back to the Special Directive, incorporating its citation to the same article. Prof. Mueller-Smith lists this article on his U. Mich. page as “Revise and Resubmit at the American Economic Review.” Apparently the editors of that journal had some problems with it.
The claim “studies show” tends to imply that a survey of the literature in the field supports the proposition stated. That implication is not correct. The Mueller-Smith paper is an outlier. I am presently reviewing the literature and have yet to find a single published study that provides solid support for the proposition that longer incarceration actually causes an overall, substantial increase in recidivism.
First, everyone should be aware that social studies are not really “sciences,” despite their renaming a few decades back. Nothing can really be proved in the same way that we prove things in physics or biology: through controlled, reproducible experiments. The difference was vividly illustrated last year when early, uncontrolled, observational studies (the only kind that exist for criminal justice questions) indicated certain treatments for Covid-19 were effective, only to be disproved when the definitive controlled studies came in.
That said, decisions need to be made, and it is better to make decisions illuminated by whatever light empirical studies may shine rather than shooting in the dark. We must always keep in mind though, the limitations of the studies. We need to know what we do not know.
Why does Mr. Gascón rely on an unpublished manuscript rather than a published article of similar vintage such as Mears, Cochran, Bales & Bhati, Recidivism and Time Served in Prison, Journal of Criminal Law and Criminology, vol. 106, no. 1, pp. 83-124? JCLC some time back dropped any semblance of neutrality and has become a megaphone for one side. It would be a good place to look for support of Mr. Gascón’s thesis if any existed. Yet this article fails to support the proposition that Mr. Gascón wants supported:
Prison stays may be criminogenic, they may reduce offending, or they may vary in their effect, depending on the amount of time served and the balance of criminogenic and deterrent or rehabilitative experiences incurred during a prison stay. The bulk of work to date suggests that time served in prison may exert mixed effects on recidivism, though more recent work suggests that the effect may be minimal. As reviews and studies have highlighted, it remains the case that few rigorous empirical assessments of the time served and recidivism relationship exist, and fewer still have assessed whether the relationship is curvilinear.
In other words, the answer to the question “do longer prison sentences increase, decrease, or have no effect on recidivism?” is “it’s complicated, and we don’t know a lot.” That is the “wrong” answer to support Mr. Gascón’s agenda.
Few, if any, of those of us on the tough-on-crime side have suggested that reducing the recidivism of the particular criminal is a reason for a long prison sentence. Our primary reasons have been incapacitation and retribution. That is (1) locking a criminal up greatly reduces his ability to commit crimes, and executing him eliminates it altogether, and (2) the violent thug deserves it.
Mr. Gascón cannot deny the protective effect of incapacitation. To maintain the myth that soft policies actually promote public safety, he must concoct a countervailing effect. He must assert that not only do long prison sentences increase recidivism, the effect is so strong that it outweighs the protective effect of incapacitation. That is exactly what he asserts in the Special Directive quoted above.
To support this counter-intuitive proposition, he comes up with a grand total of one unpublished manuscript.
No, the studies on the whole do not show an effect of the direction and magnitude that Mr. Gascón claims. Not even close. A full review of the literature is in progress. Stay tuned.
SUBSCRIBE TO OUR FEED
The “study” is nonsense on its face. First, over the last generation, when incarceration increased considerably from what it was in the generation before that, crime massively DEcreased. It’s just very hard to think that that would be true if incarceration INcreased recidivism (i.e., repeat crime). Second, did it ever occur to Gascon that the people incarcerated for the longest sentences are the most hardened criminals — that is, those who are committed to living a life of crime no matter what? If they’re out on the street, they’re committing crime. If we let them out earlier, they’ll commit more. This has been true for decades.
The unhappy truth is that, by the time a person is in his twenties, it’s already pretty much determined how he’s going to live. Rehab mostly doesn’t work. For those like me on the law-and-order side, the other unhappy news is that deterrence mostly doesn’t work either, because criminals are convinced (with pretty good reason) that they won’t be caught or punished even if they are. The only thing we’re sure works is incapacitation.
I look forward to seeing your review of the literature, Kent, and I know the US Sentencing Commission has a recent study from April 2020 on “Length of Incarceration and Recidivism”: https://www.ussc.gov/research/research-reports/length-incarceration-and-recidivism. I am pretty sure that report found less recidivism after longer incarceration terms, though I know a number of thoughtful folks have been quite critical of the USSC’s methodology.
On this general topic, I have always wondered and worried that crimes committed by persons while IN prison are systematically undercounted and/or disregarded when incapacitation/recidivism effects are measured. A released former prisoner who gets caught with drugs or who assaults his neighbor will get arrested and be deemed a recidivist, whereas a current prisoner who gets caught with drugs or assaults someone on his cell block may only be subject to internal discipline and many not be “coded” as a recidivist. Some (perhaps most?) persons in prison are likely only really partially incapacitated — though the nature of their crimes and their victims may be significantly (and quite valuably) altered.
Also, I suspect the nature of confinement likely has to have some impact. Time spent in brutal conditions or solitary certainly seems more likely to be more criminogenic than time spent in “club fed” or aided by effective prison education/programming. Not sure if any of your literature review will get to this element of the story, but I am sure we still have a lot to learn about the relarionships — past, present and possible — between incarceration and subsequent offending.
Doug, thank you for that link. The study is interesting, though one must always be careful about generalizing from federal to state offenders or vice versa. The two populations are not the same. The references do include state-prisoner studies, which will be helpful in the literature review.
I agree that crimes within prison are a serious problem, but I’m not so sure they should be lumped together with the recidivism problem.
Conditions in prison may well matter, but it seems quite difficult to disentangle conditions from predisposition. The prisoners who go to Supermax are the hard-core criminals whom the prison authorities have decided can’t be safely confined any other way. Of course they are going to have higher recidivism rates.
Doug — There is much to say here, but I want to start with one basic question first. It used to be said that the criminal owes a debt to society. Do you agree with that, or do you think it’s more nearly the other way around — that society owes a debt to the criminal?
Bill, I do not think “society” — whatever that means — owes anything to anyone, though I think it can often be quite beneficial for governments to be involved in helping to provide services (e.g., schooling) and goods (e.g., parks) to advance and improve the general welfare. Figuring out the most effective and efficient ways to provide these services and goods, and whether past criminal activity ought to move one back or forward in the line, present hard questions in a lot of settings — e.g., providing “free” access to colllege coursework and/or vocational training in prison has the potential to reduce recidivism, but many people without a criminal record go deep into debt to access these services; lots of prisoners are getting the “good” of the COVID vaccine before me, but I do think that is proper because I am at liberty to social distance while they are not.
As for the criminal “owing a debt to society,” I am sincerely at a loss for what this means. I understand the notion that criminal activity that harms a tangible victim may create a kind of moral debt between the wrongdoer and his victim. But if, say, a neighbor smashes into my mailbox while drunk driving and then pays to have it replaced with a nice newer mailbox, I will surely feel that any “debt” to me is “paid” and I struggle to see what additional “debt” my neighbor owes to the broader society. This weekend, we can expect over 200,000 Americans will drive under the influence and at least as many teenagers will consume alcohol and/or marijuana. Do you think all these weekend criminals owe “a debt to society”? What kind of debt? How can it be paid off? Do people who speed also incur these debts? How about those who try to corruptly alter the resuls of a lawful election?
If and when you can further explain this concept of “the criminal [owing] a debt to society,” I would be especially eager to hear if you think a criminal paying fines and/or doing service in society can and should be considered a sound way to pay off this “debt” rather than sitting in a cage. I strongly believe requiring criminals, especially rich criminals unlikley to pose much risk of re-offense, ought to have to pay large economic penalties and/or do community service because that would be much more productive for society than having them locked in a cage. But I am drawn to these types of punishments mostly for utilitarian reasons (e.g., they seem cheaper and more productive). But I could imagine someone with a robust vision of criminals having a “debt to society” might see particular value in punishments that more directly involve a contribution back to society.
“As for the criminal ‘owing a debt to society,’ I am sincerely at a loss for what this means. I understand the notion that criminal activity that harms a tangible victim may create a kind of moral debt between the wrongdoer and his victim. But if, say, a neighbor smashes into my mailbox while drunk driving and then pays to have it replaced with a nice newer mailbox, I will surely feel that any “debt” to me is “paid” and I struggle to see what additional “debt” my neighbor owes to the broader society. ”
As a professor of criminal law, you are certainly aware of the difference between tort liability and criminal liability. The former addresses the harm bad behavior does to an individual; the latter address the harm it does to society.
Do you think that when someone cracks your head open with a tire iron and takes your wallet, that everything is OK — that society should have nothing to say — if your attacker pays your medical bills, pays you for pain and suffering, and returns your money with interest? That should end the matter??
If so, criminal law is not your field. Its whole premise is that the wrongdoing to a member of society harms the society as well, which is then entitled (some might say obligated) to respond with punishment, deterrence, (sometimes) incapacitation and (sometimes) rehabilitation.
Still, your remark here does give me greater insight into why you so stoutly resist criminal punishment and so uniformly want lower sentences. If there is no such thing as harm done to society (by aggravated robbery, or murder or rape or carjacking, etc., etc.) then you’re right — society should butt out. But I don’t really believe you think that, since it’s nuts and you’re not nuts.
So let me rephrase my question. Do you think it’s mostly up to the convicted criminal to change his behavior towards others, or is it mostly up to the society (or government if you wish) that has convicted him to change ITS behavior (to become more accommodating or compassionate or whatever)?
Bill, I see an important role for criminal law in seeking to keep people safe from various harms, and ideally it will work efficiently and effectively in harmony with tort and regulatory laws (which have various limitations in doing so). I readily understand and generally support consequentialist rationales and justifications for all sorts of criminal liability and punishment.
But I have always struggled to understand and give concrete meaning to retributivist thinking framed in terms of “debt to society.” In particular, I remain very eager to hear your account of (A) what kinds of “debt to society” you think is owed by someone who commits a crime lacking tangible victims, and (B) whether and how that “debt” can and should be paid . If you cannot explain what is owed by, say, somone who uses drugs illegally, then I do not think you really mean a “debt to society,” but rather may be conflating debt to a victim and consequentialist concerns. (I focus on drugs since we have more arrests each year for that crime than for the seriuos violent crimes you mention.) I am not trying to troll you with this question — which I will keep asking beause it goes to the heart of the reach and rationale of criminal liability and punishment — as I quite sincerely just want to better understand just what you mean by asserting that a “criminal owes a debt to society.”
The fact that you ask about a criminal “chang[ing] his behavior toward others” suggests to me further that you are really just thinking about criminal activity that directly harms others rather than illegal drug use or other so-called “victimless” crimes. I see critical differences between what criminal law can achieve and how it should be used to protect “others” and how it might be used for other purposes. I believe you had a role in federal drug enforcement, so I hope you can explain to me think concept of “debt to society” with that type of crime since it plays such a large role in the operation of our crimina justice systems And I really cannot effectively move forward with this line of discussion unless and until I know if you are ONLY really eager to focus on crimes and criminals with tangible victims or are talking about all crimes and criminals like marijuana users and college drinkers .
Can you understand what this is a critical issue to address? You are asking me to answer questions and I am looking for much more explanation and precision concerning what you are asking about. (And, again, statistics show that one of the most common “crimes” and “criminals” in American society involve illegal drug (and alcohol) users.) So, in short: do you think all criminals owe a “debt to society” or ony some? If only some, which ones?
Oh yeah, Bill, I would also be eager to hear about whether and how your notion of criminals owing a “debt to society” intersects with legal adjudication. I sense a large number of rioters — in cities this past summer and/or on Jan. 6 at the Capitol — will not be subject to formal prosecution. Do they still have a “debt to society” because they acted criminally or do only those subject to prosecution actual incur this “debt”?
Has Derek Chauvin’s “debt to society” already been established or does it turn on whether he gets convicted by a criminal jury of murder or manslaughter or is acquitted? (I think of OJ Simpson as a classic example of a wrongful acquittal, but maybe you think that acquittal morally (of functionally) eliminated his “debt to society.”) Could Chauvin or another alleged criminal reduce his “debt to society” by striking a plea deal to a lesser charge or does the highest plausible charge define the “debt to society” that you have in mind? Does the fact Chauvin was a police officer influence whatever “debt” he may have for his behavior?
I sincerely want to hear a lot more about what you mean when you say the “criminal owes a debt to society” if it serves as the very raison d’etre for a criminal justice system (and a key calibrator of punishment ) in your view.
Doug — In response to your last two comments, I’m going to try to make things easy.
It used to be a commonplace (rather than the deep mystery you say you see), that the criminal owes a debt to society. But to clear away the underbrush, I’ll be more precise: Do you think a person who is and has been found guilty of a serious crime owes a debt to society? Or do you think it’s more nearly true that society owes a debt to him?
If you want to give separate answer for drug crimes vs. non-drug crimes, please feel free.
P.S. I’ve noticed that one of your debate techniques — one that’s frequent among defense lawyers but you do it with more persistence and skill — is, instead of answering a question, to ask for this refinement and that refinement off into them thar’ hills until everyone just wanders away in frustration. What I have in mind is not just your parry here, but your response in an early thread about defense bar ethics.
My position was (and is) that it’s dishonest for a defense lawyer to contest a charge he knows or has strong reason to believe is true. I asked if you agreed. Your response was that you weren’t sure what I mean by the word “contest,” asking me to explain.
At that point, I saw no great point in continuing. You’re a bright and experienced lawyer and law professor, and there is just no chance you don’t know what it means to “contest” a charge. I strongly suspect you’ve known it since eighth grade if not before.
Words have meanings, and not everything is or needs to become a thicket. My experience in litigation, which is lots, tells me that the side that wants to make like it’s a thicket is the side that either doesn’t have a straight answer or doesn’t want to give it.
But let me try again nonetheless. Do you think a person who is and has been found guilty of a serious crime owes a debt to society?
Bill, you tend to ask very broad questions and make broad statements — e.g., does a “criminal owe a debt to society?” and “it’s dishonest for a defense lawyer to contest a charge he knows or has strong reason to believe is true.” Hoping our conversations can be more precise, I follow up by asking questions seeking to better understand your meaning — e.g., “do you think a college kid drinking alcohol at a party owes a debt to society?” and “do you think lawyers contesting charges against the Jan 6 rioters caught on video are dishonest?” Notably, you often dodge my follow-up inquiries, suggesting you are the one who “doesn’t have a straight answer or doesn’t want to give it.”
You now ask, “Do you think a person who is and has been found guilty of a serious crime owes a debt to society?”, and again I need to better understand what you mean by “debt to society” and now also “serious crime.” According to FBI data, the highest number of arrests were for crimes that involve drug abuse violations, driving under the influence, liquor-related offenses, larceny-thefts, assaults and frauds. Depending on the circumstances, all of these kinds of crimes can be quite “serious” but they also can be relatively minor. (E.g., there are 25 drunk driving deaths on average each day, but many thousands drink and drive “safely” every weekend. Illegal drugs in the US are now contributing to up to 2000 overdose deaths weekly, but millions are “safely” and illegally using marijuana as I write this.)
When you say “serious” crimes, I am inclined to assume you mean crimes with clear victims. But I doubt by “serious” you mean minor shoplifting or small bar fights without lasting injuries, even though these crimes have clear victims. I also would be eager to know if you think all (or just some) of the Jan 6 rioters committed “a serious crime.”
Whatever your meaning of “serious,” as I explained before, I can understand the idea that a wrongdoer has a “debt” to a victim and I also have a sense for how that “victim debt” can be in some sense “repaid” through our criminal justice systems. But I remain eager to have you help me understand (A) what kinds of “society debt” you think follows from a “serious crime” lacking tangible victims, and (B) how criminal processes and punishments should account for this “society debt.” (Recall, I explained clearly that I do not think “society” owes anything to anyone, so maybe I just need a clearer understanding of what you mean by “debt” and “society” — e.g., does a person who commits a “serious crime” in Columbus, Ohio owe a “debt” to all of Ohio? is the same debt also owed to all of the US? all of the world? As a member of “society” do I have any standing in “serious” criminal cases if I fear this “debt” is not being addressed properly? )
I am not trying to be obtuse, but I have always seen the language and “practices” of retributivist criminal philosophy to be vague (and often misguided) to the point of frustration. I am digging here because I find that the harder one thinks about retributivist aphorisms like “debt to society” the harder it is to see them making any real sense (or at least being of actionable significance). Notably, you have now reframed your inquiry in terms of only convicted criminals found guilty of “serious crimes,” but I really do not think that adds any much clarity to the myserious concept of “debt to society.”
Finally, a group of modern criminal law thinkers focused on visions of criminals repaying a “debt” are advocating for restorative justice models to replace our traditional adversarial models of criminal law. These folks often talk about the importance of being much more focused on “restoring” the community and victims (and offenders) through much different procedures than used in traditional criminal courts. I am assuming your “debt ot society” ideas are not designed to start us down the restorative justice path — but the proponents of these models do often claim they are much more “truth-centric” so maybe that is where you want to be headed.