Judge the Alter, Punish the Whole?

Introduction

Dissociative Identity Disorder (DID) (formerly known as Multiple Personality Disorder) is a severe mental illness in which an individual switches between more than one distinct personality. While the condition poses an interesting metaphysical conundrum, the problem becomes practical in the courtroom: if a DID sufferer commits a crime while in one of their distinct personality states (alters), how should we gauge the criminal responsibility of the whole person? 

In what follows, I will explain and defend my objection to a fundamental point in Sinnott-Armstrong and Behnke’s 2002 argument, one that underpins their “test of responsibility”. After honing in on where I think the authors went wrong, I will demonstrate that, if my objection stands, it could significantly compromise the soundness of their thesis.  

Context 

To evaluate Sinnott-Armstrong and Behnke’s argument, it’s helpful to understand their principal aim. They were trying to overcome medical uncertainty and legal inconsistency by contributing to a debate on whether there were one or more metaphysical persons in the body of an individual with DID. 

So Sinnott-Armstrong and Behnke set out to determine, firstly, whether a DID sufferer was in fact constituted by one, or many persons, and then to propose a practical solution to determine criminal responsibility. Their intention was worthy - to develop a test that could be applied across all cases and would consistently result in fair outcomes. Whether they were on the right track may turn on your conception of justice. 

The Thesis Arises

Sinnott-Armstrong and Behnke begin by weighing up the main arguments for the multiplicity of persons thesis and come to the conclusion that they are insufficient - a DID sufferer is indeed one person. From here, they infer their central thesis, or ‘test for responsibility’: that a DID sufferer should be held criminally responsible if the alter in control at the time of the crime does not meet the “applicable standards of criminal sanity.” 

An example serves to illustrate this claim: 

Imagine a DID sufferer called Emma. Emma has two distinct personalities, one is timid, polite and shy, and would never dream of breaking the law (let’s call her Nice Emma), while the other is aggressive and impulsive, with a drinking problem (let’s call her Angry Emma). If Emma commits a crime while Angry Emma is in control, then (according to Sinnott-Armstrong and Behnke), what the judge should take into account is the following: 

Was Angry Emma legally insane at the time she committed the crime? Was Angry Emma experiencing delusions such that she did not know what she was doing? If not - if Angry Emma was just being ‘Angry Emma’, with no delusions, and she committed a crime - then the whole of Emma should be judged criminally responsible.

The Supporting Point

This central thesis - the ‘test of responsibility’, relies upon one of the authors’ previous claims, that: “alters in cases of MPD are analogous to temporal parts of a normal person.” This is the point I am directly refuting. This notion is introduced when the authors state: “I cannot control what I did yesterday, since it is past. I might not be aware of it, if I do not now remember doing it. But that does not mean that I am not the person who did it, or that I cannot be punished for doing it.”

Shortly after this, they claim: “Alters in a case of MPD seem analogous in relevant respects to different parts of a single person’s consciousness at different times.” They assert this theme again, when they note: “We argued above that alters in cases of MPD are analogous to temporal parts of a normal person. This suggests that the agent at the time of the crime is the alter in control at that time.”

The authors then proceed directly from this statement to their thesis: “The crucial legal question then (my emphasis) must be whether that alter at that time meets the relevant standards of insanity.” It’s important to note that this last statement is written as though it is the natural conclusion of the one preceding it, thus their thesis appears to rely on DID alters and temporal parts of sane individuals being analogous. 

Are They Really Analogous? 

Let’s compare two examples that, by the logic of the above claim, should be analogous:

Example A:

The difference between the mentally healthy temporal you that had avocado on toast for breakfast a month ago, and the current temporal you that can’t remember having it, due to the normal memory loss that occurs as a natural function of a healthy brain sorting through which memories are significant enough to store in long term memory, and which are not. 

Example B (let’s use Emma again):

The difference between the part of Emma (Angry Emma), that murdered someone a month ago in a bar fight, and the current Emma (Nice Emma), who can’t remember committing the crime, not due to healthy-brain function memory loss, but because, as a DID sufferer, she (statistically probably) experienced serious trauma at a young age, and, as a result, dissociated and developed a DSM recognised, severe mental disorder which involves the presence of another personality which presents spontaneously, over which she has no control in the moment, and the (serious, life-changing), actions of whom she may have no memory of the very next day, or ever.

It seems rather obvious to me that Example A and Example B are not analogous. 

I think that the healthy continuity of memories is essential to the development of moral character. I believe that a person needs to learn moral lessons and receive and integrate moral feedback continuously over time to develop moral agency. Therefore, I don’t think DID sufferers have the level of psychological continuity over time (specifically in terms of memories), that allow for enough agency and control to be criminally responsible. In anticipation of an objection, I do think we would be justified in requiring dangerous DID offenders to undergo treatment within a psychiatric unit, based on a right to protect ourselves and others, without need for blame. 

Conclusion

As I have demonstrated, Sinnott-Armstrong and Behnke’s claim that different temporal selves in a sane individual are analogous to different alters in a DID sufferer is essential to their central thesis. If the analogy fails, which I think I have shown that it does, then it would seem that the entire ‘test of responsibility’ is weakened to a point I believe renders it misguided and unreliable, if justice is our aim. The question I am left with is this: is there not a contradiction in arriving at a single person thesis, and then endorsing judgement of the alter, yet punishment of the whole? 

Bibliography:

Brand, B., Vissia, E., Chalavi, S., Nijenhuis, E., Webermann, A., Drajier N & Reinders, A. DID is trauma based: Further evidence supporting the trauma model of DID. Acta Psychiatrica Scandinavica, 134(6), 560-563. 2016

Kennett, J & Matthews S. Identity, control and responsibility: The case of Dissociative Identity Disorder. Philosophical Psychology, 15:4, 509-526, 510-511. 2002

Kennett, J & Matthews S. Delusion, Dissociation and Identity, Philosophical Explorations. 6:1, 31-49, DOI: 10.1080/10002002108538738. 2003

Sinnott-Armstrong, W & Behnke, S. Responsibility in Cases of Multiple Personality Disorder Nous, vol.34 no. 14, 2000.