Neurolaw Revisited

As I noted a few weeks back, there was a recent article in The New York Times by Jeffrey Rosen concerning the growing new field of neuro-law.  The central issue in Rosen's piece is the role that cognitive neuroscience (CNS) can (and should) play with respect to the criminal law (and its underlying notion of criminal culpability).  In the article, Rosen mentions an interesting paper by O. Carter Snead entitled “Neuroimaging and the ‘Complexity’ of Capital Punishment.”  And while Snead mostly limits his attention to the role that CNS could play with respect to capital sentencing (and punishment) rather than punishment more generally, for present purposes I want to use his argument as a way of jump starting a discussion concerning the general relationship between CNS and criminal law.

For starters, I want to first provide a brief overview of what I am going to call the Cognitive Neurolaw Agenda (CNLA)—an agenda that Snead associates with the following two-fold goal:  (a) the short term goal of CNLA: get CNS findings into the courtroom as mitigating evidence, (b) the long term goal of CNLA: use findings from CNS both to debunk retributivism at the institutional level and to develop a forward looking two-pronged preventative and rehabilitative regime. 

According to Snead, the proponents of CNLA envision the long term goal as a focused attempt to adopt a more humane and compassionate response to the crime problem than the present retributivist regime. Snead criticizes CNLA on two fronts.  On the one hand, he tries to show that the short term goal and long term goal of CNLA are inconsistent.  By his lights, the present goal of using data from CNS to mitigate or exculpate blame would be undermined by the long-term goal of purging the criminal law of retributivist principles.  On the other hand, Snead claims that if the long-term goal of CNLA were to come to fruition, it would make the system of criminal law more rather than less brutalizing and inhumane. 

Jonathan Cohen and Joshua Greene are two individuals who advocate for something roughly like CNLA.  In a very fascinating paper entitled “For the Law, Neuroscience Changes Nothing and Everything,” they argue that as the public becomes better informed with respect to the gathering data from CNS, it will have a transformative effect on our beliefs about agency, autonomy, free will, and moral responsibility.  Moreover, according to Cohen and Greene, this paradigm shift in our self-conception will usher in a paradigm shift with respect to criminal law as well.  On their view, a forward looking and preventative and rehabilitative regime will eventually replace the backward looking retributive model we have now.  As Snead correctly points out, “Cohen and Greene do not merely believe that the overthrow of retribution by cognitive neuroscience will yield a better functioning system of criminal justice.  They believe that it will make it more humane for criminal defendants” (Snead: 45). 

Proponents of CNLA not only believe that retributivism is undermined by the gathering evidence from CNS, they also believe that purging the criminal law of retributivistic principles will have a humanizing effect on the law.  In this respect, the CNLA fits quite nicely with the progressive and reformative theories of punishment developed earlier by Becarria, Bentham, Wooten, and others.  What is envisioned is no less than an entire paradigm shift from a regime that focuses on the suffering of the defendants to one that focuses on both the defendant’s well-being as well as the well-being of society at large.  The benefit of CNS—according to proponents of CNLA—is that it may help soften (or perhaps even eliminate) the public’s retribuvistic impulses and beliefs—thereby paving the way for the transformation of the criminal law envisioned two centuries ago by reformers such as Bentham.

Snead’s main criticism of CNLA is that:

Simply put, the project, taken as a whole, is utterly at war with itself.  The short term aim relies on a particular theory of mitigation that is firmly grounded in the notion of just deserts—a principle that is explicitly rejected by the architects of the cognitive neuroscience project as destructive and unintelligible (Snead: 54).

On his view, “it is only by virtue of the doctrine of just deserts that neuroimaging evidence of the roots of criminal violence can be understood as mitigating…the long term aspiration seeks to undermine and destroy the very distributive principle of retributivist justice upon which that of the short term depends.” (Snead: 54)  Moreover, not only does Snead think that the long term success of CNLA will undermine the use of CNS for establishing mitigating factors, he also claims that the data from CNS will still be used to establishing aggravating factors and future dangerousness.  As he says:

In a further ironic twist, once retribution is snuffed out as a distributive principles of punishment is replaced with a regime that single-mindedly concerns itself with the prediction of crime and the incapacitation of criminals, the only possible significance of the body of neuro-imaging research on the roots of criminal violence for capital sentencing (on which the short term approach is erected) is as an aggravating factor of future dangerousness (Snead: 60).

In addition to purportedly establishing that CNLA is internally inconsistent, Snead also claims that even if the long term goal of CNLA were to come to fruition, it would have a brutalizing rather than a humanizing effect on the criminal law.  In supporting this claim, he suggests that:

Many of the features of the criminal justice system that arte frequently criticized as draconian and inhumane, are in fact, motivated by the consequentialist crime-control rationale of predicting and controlling dangerous behavior.  Such measures include: laws that authorize life sentences for recidivists; laws that reduce the age at which offenders can be tried as adults; laws that punish gang membership; laws that require the registration of sex offenders; laws that increase sentences dramatically by virtue of past history; and, most paradigmatically, laws that provide for the involuntary commitment of predators who ‘show difficulty’ in controlling their behavior (Snead: 63).

Snead takes these examples as evidence that if the proponents of CNLA are successful in transforming both public opinion and the criminal law, this success will produce more rather than less draconian punishments.

I am presently working on a response to Snead’s paper—but it is still really rough around the edges.  For now, I just wanted to float two knee-jerk responses to Snead that may hopefully generate some discussion here on GFP.

Objection One:

Snead tries to show that if CNLA’s long term goal of purging the law of retributivism were successful, defense attorneys would no longer be able to use evidence from CNS as a mitigating factor.  According to Snead, the very notion of a mitigating circumstance only makes sense against the backdrop of retributivist justice.  After all, the whole point of trying to establish that there were mitigating circumstances surrounding a defendant’s actions is to show that he is less than fully blameworthy.  But if—as the proponents of CNLA suggest—no one is blameworthy, then the very idea of mitigating circumstances becomes otiose.  If all of us are equally blameless, there is no blame to mitigate!  Snead takes this to show that in some sense CNLA defeats itself.  But is this correct? 

The most obvious move for the proponent of CNLA to make is to simply deny that the underlying issues that presently arise in the courtroom when judges and juries consider mitigating and aggravating circumstances would have no roll to play in a purely forward looking regime.  Keep in mind that consequentialists such as Bentham insist that certain restrictions are to be put in place when it comes to punishment.  Minimally, in order for punishment to be justified in a given case, punishing that individual must not be groundless, inefficacious, unprofitable, or needless.  On this view, determining whether punishing an individual fails on one of these fronts requires judges and jurors to look very carefully at the specific circumstances surrounding a violation as well as the specific circumstances surrounding the mental states of the agent.  Finding out that someone had less than full control over their action does not make them less blameworthy—after all, in some sense, no one is blameworthy according to the proponents of CNLA—but it does make punishing them less sensible.

If Snead just means to claim that the long-term goal of CNLA would undermine the notion of a mitigating circumstance, then what he says is in some sense true by definition.  After all, if “mitigating” just means “blame reducing,” then clearly the long-term goals of CNLA threaten the role of mitigating circumstances in the criminal law.  But notice this would also undermine the notion of an aggravating circumstance as well.  But it does not follow than the very same issues that arise when we presently consider aggravating and mitigating factors won’t arise in a post-retributivistic world.  Just because we may no longer talk about a defendant’s blame, we may nevertheless be very interested in his upbringing, his beliefs, his mental and emotional capacities, his impulsiveness, his susceptibility to addiction, and the degree of remorse he expresses.  These are all factors that will surely continue to play a central role in our criminal proceedings—albeit for different reasons.  Rather than appealing to these factors under the banner of determining moral and legal blameworthiness, we will appeal to them under the banner of determining what needs to be done in order to rehabilitate and reintegrate criminals into society more effectively and efficiently. 

Moreover, by focusing on crime prevention and rehabilitation rather than general deterrence, the proponent of CNLA can further allay many of Snead’s worries since much of his fretting seems to be driven by the assumption that all forward-looking punishment theorists are going to focus on general deterrence more than prevention and rehabilitation (see, for instance, his discussion of the debate between Bentham and Blackstone concerning excuses and utilitarianism). 

Critics of forward-looking theories of punishment often conveniently forget that these theories almost always focus first and foremost on crime prevention rather than general deterrence.  The importance of this forgetfulness is a story for another day.  For now, I just want to point out that Snead makes a blanket assumption about forward-looking theories of punishment that cannot withstand close scrutiny.  Nearly all of the factors that retributivists think are relevant to determining how much punishment criminals deserve will be relevant to the consequentialist’s attempt to figure out how to reform and rehabilitate criminals.  Indeed, the only factor that will be irrelevant to consequentialist is how much suffering criminals deserve for their transgressions.

Objection Two:

Snead claims that forward-looking theories of punishment are ironically to blame for what he takes to be the recent dehumanizing of the criminal law (e.g., mandatory minimum sentences, trying children as adults, and the like).  My response is two-fold:  First, his claim appears to be false on the facts.  The criminology literature over the past twenty years reveals that the pernicious and counter-productive punishments Snead identifies have been driven not by consequentialism but by the rebirth of retributivism that we witnessed in this country beginning in the late 1970’s.  Moreover, not only does the criminology literature show that Snead is wrong about the source of these new forms of “hard treatment,” it also reveals that most criminologists are painfully aware of just how unsuccessful many of these programs are in the grand scheme of things when it comes to crime prevention.  Indeed, I suspect that one could show that each and every one of the punishments Snead mentioned would fail one of Bentham’s aforementioned four conditions for justified punishment.

Second, even if we grant for the sake of argument that abandoning retributivism would produce longer sentences for less serious violations, Snead’s worries seem to sneak retributivism in through the back door.  Keep in mind that one reason we are particularly keen on making sure we don’t send innocent people to jail is that prisons are brutal, dehumanizing, and depressing places that are designed to punish criminals rather than reform or rehabilitate them.  It is precisely because of the prevalence of hard treatment that makes punishment so brutal.  Minimally, the brutality of our current regime makes the worry about punishing the innocent (or punishing people more than they deserve) all the more pressing.  But if we viewed prisoners as human beings who understandably need our assistance in being part of society rather than apt targets of our retributivistic impulses, “prisons” would look less like dungeons in the dark ages and more like treatment facilities.  For now, we need not worry about what these facilities might look like.  It should suffice to point out that they would not look anything like they do now.  By my lights, Snead’s argument only appears successful because he forgets to take into account just how radically different the criminal law would look if the proponents of CNLA have their way.  His worries about dehumanization are fueled by a failure to take into consideration the myriad of ways that CNLA would change the way society views criminality. 

Once we let go of the idea that people deserve to suffer for their anti-social sins, we will quickly see that what criminals need is encouragement, a sense of self-worth and accomplishment, the ability to read and write, the fortitude and determination to stay off drugs, anger management, treatment for mental illness, job skills, and the like.  And when our best attempts to build up the criminal fail to stave off his future criminality, we may regrettably need to quarantine him.  But the place where he will be housed will not be a prison of ill-repute but a more hospitable place that is not designed to make the individual suffer any more than he already has as a result of his inability to play by the social rules.  By my lights, retributivists such as Snead can only scare us away from forward-looking theories of punishment that focus on prevention and rehabilitation by keeping a lot of the backdrop against which we are asked to judge these theories decidedly retributivistic.  This kind of shortsightedness makes retributivism appear to be part of the solution rather than the ultimate source of the bulk of the problems!

Comments

[Tamler Sommers]

Great post, Thomas. As it happens, I just broke down Greene and Cohen's argument for my class. Here's how I see it working:

1. The law is compatibilist in that it permits retributive punishment when certain compatibilist conditions are met.

2. But the legitimacy of the law, as written, depends on whether it adequately reflects certain moral intuitions and societal commitments.

3. Our moral intuitions are libertarian: we believe (a) we have libertarian free will (LFW), and (b) that libertarian free will is required to justify desert and retributive punishment.

4. Our belief in LFW grounded in our belief in a dualist conception of ourselves as agents.

5. Neuroscience will undermine the dualist conception of ourselves as agents.

6. Therefore neuroscience will undermine our belief that we have LFW.

7. Therefore neuroscience will undermine our belief that punishment based on just-deserts can be justified.

8. Therefore neuroscience will undermine the legitimacy of the (compatibilistic) law, because it will no longer reflect our moral commitments.

Snead's mistake in Objection One is to think that Greene and Cohen believe that neuroscientific evidence itself is what exonerates the agents. But G and C explicitly claim over and over that this is not the case. Long before fMRI machines came along we knew that the brain caused behavior. Long before the discipline even existed we had enough scientific information to know that our notions of blame and praise were not justifiable. According to G and C, what neuroscience will do is make us better appreciate old philosophical arguments against free will by undermining our natural dualistic conception of ourselves as agents.

Obviously compatibilists are going to go after premise 3--in fact, I used Nahmias et al in PPR to challenge it in my class. But there is nothing incoherent about the project. All G and C are saying is that neuroscience will make us UNDERSTAND what we already should have known--that there is no basis for punishing people on retributivist grounds.

That said, you're right that neuroscientific results will likely have important implications for any utilitarian system of punishment once we have already accepted that this is the way to go. I do think, however, that utilitarians like G and C are too quick to claim that punishing the innocent or extra-severe punishment could never be the utilitarian result. That's an empirical question and we don't have enough evidence in my view to state with certainty that utilitarianism would not support these measures in certain cases. I also wonder why we're so resistant to punishment of the innocent in rare cases. Why should this be such a problem, one that every utilitarian feels compelled to address? (After all, if skeptics are right, we're all innocent in one sense: nobody deserves more punishment than anyone else.) Is it really because we think punishing the innocent would never lead to greater happiness? Or is it that we're at least partly still (negative) retributivists at heart? We think it's less FAIR to punish people who did not commit a crime than people who did.

[Rob Sica]

If, as William Ian Miller’s work (e.g. “Eye for an Eye”) powerfully and entertainingly suggests, Nietzsche (e.g. GM 2) is right that punitive retributionism is an aspect of -- and reinforced by -- a considerably broader and deeper talionic structuring of our thought in non-punitive domains, then not only is the long-term goal of purging the law of retributivism quixotic, but Snead’s concerns about the realization of such a goal are misplaced. One must wonder how some of the empirical claims about human nature and the way in which retributionism is conceptualized in this debate would be affected by consideration of the actual human moral past. Bernard Williams, help!

[Kip Werking]

I agree that it was a great post (of course Tamler and I are going to say this!).

I think Thomas' discussion of the first objection is right on. Before I read it, I didn't understand Snead at all. Reading Thomas' response, I understood both Snead's point and how, I think, it fails.

I'm less sure that Thomas is being fair to Snead about the second objection. Here I want to echo what Tamler said: if we are going to be utilitarians (or something like utilitarians) it might require us to be extremely firm in certain situations---like killing innocent bystanders.

Snead cites several examples of "draconion" principles that are consistent with, and allegedly motivated by, forward-looking considerations. Thomas responds by arguing that these principles are (i) motivated by retributivism and (ii) not effective. It is not clear to me that either is the case; I would like to see the relevant data. If, for example, a seventeen year old serial killer is tried as an adult, the motivation for this might very well be deterrence---with or without considerations of retribution.

But, to make my point, I don't need to quibble with these examples. If Thomas is right about these examples, we can just as well imagine others: terrorists are threatening to blow up a major city unless we take ten innocent people and throw them off of a skyscraper. Should we do it? If you're more concerned about prevention and deterrence than just desert, then you might say yes. I would say definitely yes. The example is far fetched but less far fetched examples could be conjured up. Yet throwing innocent people off of a skyscraper seems quite draconian. That is a Snead's point, and I think it is not without some merit.

[tnadelhoffer]

Kip and Tamler,

I won't have time to post anything substantive until later on this evening. But for now I just wanted to say that I believe that one could show that if you adopt Bentham's four conditions for justified punishment, you could try to show--at least if we limit ourselves to the world we inhabit--that punishment is never justified. Indeed, I am in the process of writing a paper where I argue that we may be able to use data from criminology to show that punishing someone--if by punishment we mean hard treatment--is always either groundless, inefficacious, unprofitable, or needless.

As far as the Problem of Punishing the Innocent (PPI) is concerned, that is the topic of yet another paper I am currently writing. This time I try to argue that the PPI only arises for deterrence theories of punishment. After all, in order to motivate PPI, you have to do so by showing that the general deterrent effect justifies punishing the innocent. If you instead adopt a forward-looking preventative and rehabiliative approach to crime reduction, the problem of punishing the innocent simply does not arise. On the one hand, this approach is not a punitive approach to crime. On the other hand, the two-fold goal of this approach is to (a) adopt social policies and programs that will minimize the pernicious effects of social inequality and poverty, and (b) when people violate the rules, the sole focus will be on trying to give them the interpersonal and intrapersonal skills to reintegrate. On this approach, there is simply no conceptual space for the PPI to gain any traction.

[Neu]

"On this approach, there is simply no conceptual space for the PPI to gain any traction."

Isn't this just Quinton all over again?

Calling "punishment" something else - say "prevention" - does not solve the core problem of this line of thinking, which is that it does not preclude the idea of identifying "at risk" in advance and preventing future crimes simply by locking them up until they get "better." Of course, if we "rethink everything," then in the long run, we would be able to be far more humane and wiser than simply "locking them up" and imposing "hard treatment" on these at risk individuals. But in the long run we are all dead.

[Tom Clark]

Thomas,

Without yet having read Snead, thanks for your good replies; I'm with you 100% and look forward to your paper. I'm wondering if, besides the Center for Naturalism, there's any organized group/movement/project promoting CNLA, or more broadly a naturalistic critique of retributivism in order to humanize criminal justice policy, along the lines of Greene and Cohen. I've put together a modest proposal for such a thing at http://www.centerfornaturalism.org/proposals.htm#CCC .

Concerns about punishing the innocent of course reflect the deontological intuition that persons shouldn't be sacrificed instrumentally for the interests of others - they are ends in themselves. This principle properly limits the pursuit of deterrence since without it any innocent, at any time, could be arrested on trumped up charges and sacrificed for the good of society, on grounds of deterrence. No one, I take it, would want to live in such a society, whatever the deterrent value of such sacrifice.

The same principle limits the pre-emptive detention of those “at risk” for crime. The value of liberty sometimes conflicts with the value of security and must be given its due, without which we don’t have a culture worth having. Of course this conflict doesn’t have any ideal normative solution. When we see that someone really and truly presents a grave risk to society, we might decide to quarantine (we do this – clumsily – with the mentally ill and some sexual offenders).

But the way to avoid these conflicts and maximize both liberty *and* security is, as Thomas suggests, to invest in policies and programs that minimize the need for deterrence and pre-emptive detention in the first place. I suggest this in “Maximizing Liberty: Retribution, Responsibility, and the Mentor State” at http://www.naturalism.org/maximizing_liberty.htm .

[Kip Werking]

Thomas:

It's not clear to me from your post:

In my hypothetical (with the terrorists and innocent people), does your view recommend killing the innocent bystanders or letting the terrorists blow up a major city?

Note: you don't have to call killing the innocent people "punishment"; you could call it whatever you like (e.g. tellishment).

It seems that your view is very similar to that of Slobogin, who (in The Civilization of the Criminal Law) argues for a similar position (including your position on general deterrence). Since our views are similar too, I'd be very interested to know what other sources and authors you are reading.

[Tamler Sommers]

Tom, you write:

"Concerns about punishing the innocent of course reflect the deontological intuition that persons shouldn't be sacrificed instrumentally for the interests of others - they are ends in themselves. This principle properly limits the pursuit of deterrence since without it any innocent, at any time, could be arrested on trumped up charges and sacrificed for the good of society, on grounds of deterrence. No one, I take it, would want to live in such a society, whatever the deterrent value of such sacrifice."

Here's what I don't understand about this position (one Pereboom seems to defend as well): If no one can deserves blame for commiting a crime how is punishing the innocent a more flagrant violation of the Kantian principle than punishing the guilty? In both cases, the person is being punished as a mere means to achieve some further end, and the guilty person deserves this no more than the innocent person.

Now if your position is ultimately utilitarian in nature (you do say "no one, I take it, would want to live in such a society, whatever the deterrent value of such sacrifice") then it seems punishing the innocent is still a possible option as long as it's done rarely and in secret. True, we wouldn't want to live in this society, but what we don't know won't hurt us except in very rare cases. And those rare cases might be justifiable if they bring further benefits. I don't see how the utilitarian value of doing this can be ruled out a priori. In fact, if I were convinced that it was the correct utilitarian result (which I'm not as of yet), I WOULD want to live in such a society. It's like scuba diving: there's a small risk, but it's worth it.

By the way, I just received a copy of Tom's book and I recommend it highly to everyone!

[Tom Clark]

Tamler,

That the guilty and innocent might be equally non-deserving of blame from some ultimate perspective doesn’t of course obviate the functional role of punishing only the guilty as a way of guiding goodness. Unless sanctions are made contingent on the commission of intentional harms, then they’re properly perceived as unfair since the innocent don’t need guiding (they’ve shown no propensity for wrong-doing), and those doing the punishing properly lose legitimacy since they’re inflicting unnecessary burdens.

One of the primary goods that I as a consequentialist want to secure is that each individual’s end-in-herself personal autonomy not be sacrificed to promote the interests of others. Before agreeing to punish the innocent, I’d want to know what sort of good the sacrifice of their autonomy achieves, how great the good is, how the sacrifice reliably achieves it, and that no other means exist of achieving it. If this rationale is forthcoming (a very difficult burden to meet in the real world), then I might reluctantly (because of the moral conflict involved) agree that such a regime should be instituted. But only if it’s democratically arrived at as a matter of public policy, since another good I want to maximize in my utopia is equal participation in open debate about means and ends.

Thanks for the kind words about the book!

[David]

Hello,
My name is David and I am the owner of Own Defense. I am involved with Non Lethal Self Defense. Like most people I am not an experts in karate or some type of fighting technique. You may not own a gun or simply don't like guns. So for the most part we would like to have a way to protect ourself and love ones from harm. That is where Non lethal defense products come into play, such as pepper spray with a UV dye that could be used in a police lineup. Stun gun that would make the attacker muscles go wild with no lasting effects. Do you travel a lot. The personal travel alarm are a very popular items that hang from the door know and sounds an alarm when the knob is being turn. If you would like to find out more about these items and much more check out my business and see the different ways of protecting yourself from harm.