For my last class on naturalism and free will I composed some readings on Sam Harris’ mistreatment of the concept of free will in American law. I already deal with the legal aspects of “free will” in some detail in Sense and Goodness without God (III.4.5, pp. 109-14), and really any discussion of the subject here must begin there (where I cite and explain key Supreme Court rulings as well as standard concepts like the criteria of guilt and the insanity defense, things the public often gets wrong because they think TV legal dramas accurately portray them). I also cover the whole free will debate generally (in the whole of section III.4, pp. 97-118), and explain the reasons compatibilism provides a more fruitful understanding of free will than any alternative. (I have also blogged on free will several times before.)

But to supplement all that, I’m here reproducing one of those course readings I composed, where I address cases not mentioned in SaG (United States v. Grayson – 438 U.S. 41 (1978) and Morissette v. United States – 342 U.S. 246 (1952)) as well as a legal concept also not mentioned there, “Battered Woman Syndrome” (as a legal defense), which supplements my discussion of the insanity defense in SaG. All of this was compiled in response to Sam Harris’ (IMO awful) book Free Will. I have a lot of problems with that book. But here I’ll be addressing only one claim in it. (For those who are curious, much better books recently on free will, though still flawed, are Gazzaniga’s Who’s in Charge? and Kane’s A Contemporary Introduction to Free Will.)

The Claim

In Free Will Sam Harris says:

The U.S. Supreme Court has called free will a “universal and persistent” foundation for our system of law, distinct from “a deterministic view of human conduct that is inconsistent with the underlying precepts of our criminal justice system” (United States v. Grayson, 1978). Any intellectual developments that threatened free will would [therefore] seem to put the ethics of punishing people for their bad behavior in question. [Harris, p. 48]

Harris here and throughout assumes the American legal system adopts a “libertarian” understanding of free will (meaning: a will free from all causation, not a will that votes Libertarian). In fact, as I demonstrate in SaG, the American legal system adopts a compatibilist understanding of free will (meaning: a will that can be both causally determined and relevantly free). To understand how Harris has misread the legal evidence, we must examine the actual court decisions he is quoting and citing. We’ll find his quotation in this case is misleading in several respects.

United States v. Grayson

In Grayson the dispute adjudicated was as follows: a man (Grayson) was convicted and imprisoned for drug possession, escaped, was caught, and when tried for the crime of unlawful escape, lied on the stand. The judge gave him a harsh sentence, partly due to the fact that he lied during the trial. The defense challenged that sentence on various grounds, among those grounds being that Grayson had not been duly convicted of perjury. A lower court ruled for the defense. The Supreme Court then overturned that decision. The higher court concluded as follows (citations now omitted and emphasis added):

A sentencing judge, in fixing the sentence of a defendant within statutory limits, may consider the defendant’s false testimony observed by the judge during the trial.

• (a) A defendant’s truthfulness or mendacity while testifying on his own behalf is probative of his attitudes toward society and prospects for rehabilitation, and is thus a relevant factor in the sentencing process.
• (b) Taking into account a defendant’s false testimony does not constitute punishment for the crime of perjury for which the defendant has not been indicted, tried, or convicted by due process; rather, it is an attempt rationally to exercise judicial discretion by evaluating the defendant’s personality and prospects for rehabilitation. To the extent that a sentencing judge is precluded from relying on relevant information concerning “every aspect of a defendant’s life,” the effort to appraise character degenerates into a game of chance.
• (c) Judicial consideration of the defendant’s conduct during trial does not impermissibly “chill” his constitutional right to testify in his own behalf, for the right guaranteed to a defendant is the right to testify truthfully in accordance with his oath. A sentencing judge, however, is not required automatically to enhance the sentence of a defendant who falsely testifies, but, rather, the judge is authorized, where he determines that the testimony is willfully and materially false, to assess the defendant’s rehabilitation prospects in light of that and all the other knowledge gained about the defendant. [Grayson, p. 41]

This is the context of the decision Harris quotes. In the course of providing the reasoning and basis for this ruling, the court said (here the quoted portion in Harris is presented in bold):

Only one Circuit has directly rejected the probative value of the defendant’s false testimony in his own defense. In Scott v. United States, 135 U.S. … (1969), the court argued that:

The peculiar pressures placed upon a defendant threatened with jail and the stigma of conviction make his willingness to deny the crime an unpromising test of his prospects for rehabilitation if guilty. It is indeed unlikely that many men who commit serious offenses would balk on principle from lying in their own defense. The guilty man may quite sincerely repent his crime, but yet, driven by the urge to remain free, may protest his innocence in a court of law.

… [This] rationale rests not only on the realism of the psychological pressures on a defendant in the dock—which we can grant—but also on a deterministic view of human conduct that is inconsistent with the underlying precepts of our criminal justice system. A “universal and persistent” foundation stone in our system of law, and particularly in our approach to punishment, sentencing, and incarceration, is the “belief in freedom of the human will and a consequent ability and duty of the normal individual to choose between good and evil” [Morissette v. United States, 342 U. S. 246, 342 U. S. 250 (1952). See also Blocker v. United States etc.]. Given that long-accepted view of the “ability and duty of the normal individual to choose,” we must conclude that the defendant’s readiness to lie under oath—especially when, as here, the trial court finds the lie to be flagrant—may be deemed probative of his prospects for rehabilitation. [Grayson, p. 51]

Notice that the context does not support Harris. He picks the single word “deterministic” and assumes what the court meant was the philosophical doctrine of determinism (wherein all outcomes are determined by prior causes), when in fact, from the context, it is clear that what the court meant was a doctrine that denies the entire ability of human beings to make choices—a view philosophers in fact call fatalism. Yet Harris agrees that fatalism is false (see Harris, p. 33).

Accordingly, the Supreme Court in the Grayson case Harris quotes is actually defending a compatibilist view of free will, as against a fatalist view of human nature, not a determinist view of human nature (the court can be excused for not knowing such specifics of esoteric philosophical terminology; Harris can perhaps be less excused for not paying better attention to the context and nuance of what he is quoting). The very paragraph Harris quotes in fact defines free will as the “ability and duty of the normal individual to choose between good and evil,” and not as the power to overcome all causes operating on a choice.

The Court does assume in this passage that a man in Grayson’s condition (i.e., a defendant legally considered to be in his right mind, etc.) can overcome one causal force operating on him: his fear of the consequences of telling the truth (namely, getting a sentence extension for his escape attempt, the consequence Grayson was attempting to avoid with his lies). But at no point does Harris ever demonstrate that anyone lacks that power. Quite obviously human beings have the power of overcoming such fears, and routinely do. No violation of causal determinism is necessary for that.

Indeed, the Court sensibly observed that if we allowed that no human beings can overcome the fear of what will happen from telling the truth, then no one could ever be expected to tell the truth or ever be held accountable for not telling to truth. Any social system that adhered to such a fatalistic notion of human nature would unravel into dysfunctional chaos. Harris would surely agree with the Court’s ruling; human behavior is not so robotic that it does not even involve human choice and (as a result) truthtelling in the face of negative consequences is impossible. Therefore, Harris cannot validly use this Court’s ruling as evidence of a concept of libertarian free will being employed in American law.

Instead, the Court’s ruling is regarding the “probative value” of Grayson’s lying with respect to his sentencing, in particular the way in which his lying informs the Court about his character (his “attitudes toward society”) and what that entails regarding his “prospects for rehabilitation.” Indeed, the Court specifically said that if a person’s character does not cause their behavior, then no Court (and by extension, no one at all) could ever evaluate the quality of anyone’s character by reference to their behavior. Behavior would then be completely disconnected from any predictable attributes of the person, and evaluating their character would become “a game of chance.” Again, it is such fatalistic absurdity that the Court is rejecting, and again Harris would be forced to agree with the Court on this.

To the contrary, the purpose of moral and legal judgment, this Court affirms, is to “appraise character” and “evaluate the defendant’s personality and prospects for rehabilitation,” and to do so anything the defendant does willfully (i.e. knowingly and intentionally) is evidence pertaining to the defendant’s character, personality, and prospects for rehabilitation. And it is evidence precisely because of causal determinism. The character of the defendant causes their choice. And it is the character of the defendant the court needs to judge. Not whether they are free of causality.

For if determinism were false, there would cease to be any reliable causal connection between what a defendant willfully does and his or her character, personality, and prospects for rehabilitation. Actions would no longer be evidence of intent. Thus, legal and moral judgment in matters of praise and blame require causal determinism, to ensure that what one does evinces what one thinks (about others, about society, about right and wrong, about telling the truth and telling lies, about behaving well or poorly, and so on).

Since we want to promote telling the truth and suppress lying, one who cannot overcome their fear of the consequences of telling the truth when under oath is rightly judged a person of unreliable character with poorer prospects of rehabilitation and a disrespect of the minimal requirements of living in a society. And that is exactly what this Court ruled. Note that that ruling at no point requires rejecting causal determinism. To the contrary, it embraces causal determinism.

Morissette v. United States

Note that where the Grayson decision defines free will (the very part Harris conspicuously doesn’t quote) it quotes a previous decision, that of Morissette v. United States from 1952. So we need to look at the full context of that remark. There, the matter being adjudicated was the claim, upheld by a lower court, that “intent” should be disregarded in deciding a defendant’s guilt when intent is not specifically mentioned in the statute. That ruling was overturned by the higher court, on grounds that directly lead to the quoted remark in the Grayson case.

In the Morissette case, a man (Morissette) was convicted of “stealing” scrap metal abandoned on federal land, but without the prosecution having proven intent. The defense argued that the prosecution should have been required to prove intent (in this case, that meant proving that Morissette knew he was stealing when he collected the scrap metal, rather than simply salvaging abandoned material). A lower court ruled for the prosecution. But the Supreme Court ruled for the defense. The summary of its decision was as follows (citations now omitted):

(1) A criminal intent is an essential element of an offense under 18 U.S.C. § 641 [the law Morissette was convicted of breaking], which provides that “whoever embezzles, steals, purloins, or knowingly converts” property of the United States is punishable by fine and imprisonment.

• (a) Mere omission from § 641 of any mention of intent is not to be construed as eliminating that element from the crimes defined.
• (b) The history and purposes of § 641 afford no ground for inferring any affirmative instruction from Congress to eliminate intent from the offense of “knowingly converting” or stealing government property.

(2) Where intent of the accused is an ingredient of the crime charged, its existence is a question of fact which must be submitted to the jury for determination in the light of all relevant evidence, and the trial court may not withdraw or prejudge the issue by instructing the jury that the law raises a presumption of intent from a single act. [Morissette, p. 246]

The decision does discuss “statutory” offenses and “strict liability” where statutes specifically rule out considerations of intent, but where such specific statements are omitted, intent is always to be considered an essential component of guilt for any crime. In short, in such cases, if someone didn’t know that they were committing a crime, they cannot be held responsible for doing so, because there was no intent (there is no mens rea, to use the law school parlance). Thus, the purpose of most criminal statutes (and by extension, I would argue, most acts of moral judgment even outside the legal system) is to regulate bad intentions, not bad actions per se. It is thus intending to do harm that you are held accountable for, not merely doing harm (except in certain specific kinds of cases, which the Court also discusses).

The context in which free will comes to be defined is as follows (citations now omitted and emphasis added):

In those cases, this Court did construe mere omission from a criminal enactment of any mention of criminal intent as dispensing with it. If they be deemed precedents for principles of construction generally applicable to federal penal statutes, they [would] authorize this conviction. [However], such adoption of the literal reasoning announced in those cases would do this and more—it would sweep out of all federal crimes, except when expressly preserved, the ancient requirement of a culpable state of mind. We think a resume of [the] historical background [of those other kinds of laws] is convincing that an effect has been ascribed to them more comprehensive than was contemplated and one inconsistent with our philosophy of criminal law.

The contention that an injury can amount to a crime only when inflicted by intention is no provincial or transient notion. It is as universal and persistent in mature systems of law as belief in freedom of the human will and a consequent ability and duty of the normal individual to choose between good and evil. [Here in Footnote 4 is quoted a history of English law that declares, “Historically, our substantive criminal law is based upon a theory of punishing the vicious will. It postulates a free agent confronted with a choice between doing right and doing wrong and choosing freely to do wrong.”]

A relation between some mental element and punishment for a harmful act is almost as instinctive as the child’s familiar exculpatory “But I didn’t mean to,” and has afforded the rational basis for a tardy and unfinished substitution of deterrence and reformation in place of retaliation and vengeance as the motivation for public prosecution. [Here in Footnote 5 is quoted a prior Supreme Court decision that declared, “Retribution is no longer the dominant objective of the criminal law. Reformation and rehabilitation of offenders have become important goals of criminal jurisprudence” and “. . . a prevalent modern philosophy of penology [is now] that the punishment should fit the offender and not merely the crime.”]

Unqualified acceptance of this doctrine by English common law in the Eighteenth Century was indicated by Blackstone’s sweeping statement that to constitute any crime there must first be a “vicious will.” [Citation in their Footnote 6] Common law commentators of the Nineteenth Century early pronounced the same principle, [citations in their Footnote 7] although a few exceptions not relevant to our present problem came to be recognized. [Those exceptions are mentioned in their Footnote 8]

Crime, as a compound concept, generally constituted only from concurrence of an evil-meaning mind with an evil-doing hand, was congenial to an intense individualism and took deep and early root in American soil. [Here in Footnote 9 they quote a legal textbook making the pithy remark, “Even a dog distinguishes between being stumbled over and being kicked.”] As the state codified the common law of crimes, even if their enactments were silent on the subject, their courts assumed that the omission did not signify disapproval of the principle, but merely recognized that intent was so inherent in the idea of the offense that it required no statutory affirmation. Courts, with little hesitation or division, found an implication of the requirement as to offenses that were taken over from the common law. [Example given in their Footnote 10]

The unanimity with which [courts] adhered to the central thought that wrongdoing must be conscious to be criminal is emphasized by the variety, disparity and confusion of their definitions of the requisite but elusive mental element. However, courts of various jurisdictions, and for the purposes of different offenses, have devised working formulae, if not scientific ones, for the instruction of juries around such terms as “felonious intent,” “criminal intent,” “malice aforethought,” “guilty knowledge,” “fraudulent intent,” “willfulness,” “scienter,” to denote guilty knowledge, or “mens rea,” to signify an evil purpose or mental culpability. By use or combination of these various tokens, they have sought to protect those who were not blameworthy in mind from conviction of infamous common law crimes. [Morissette, p. 250-52]

So here their definition of free will is very carefully and explicitly framed in terms such that denying the relevance of free will in establishing guilt “would sweep out of all federal crimes…the ancient requirement of a culpable state of mind.” They are therefore, yet again, arguing not against causal determinism as Harris assumes, but a fatalism in which intent is no longer the cause of criminal behavior nor what we seek to regulate with laws. Another conclusion Harris must surely agree with.

Here, “freely” choosing does not mean free of causal determinism, but free of exonerating circumstances (such as duress, necessity, involuntary action, ignorance, or insanity). As the court in fact goes on to say in a footnote:

Absence of intent also involves such considerations as lack of understanding because of insanity, subnormal mentality, or infancy, lack of volition due to some actual compulsion, or that inferred from doctrines of coverture [i.e. being in a state of legal guardianship]. Most extensive inroads upon the requirement of intention, however, are offenses of negligence, such as involuntary manslaughter or criminal negligence and the whole range of crimes arising from omission of duty. [n. 8]

Yet even in those latter cases, one must prove that the accused reasonably should have known the duty they are neglecting stands upon them. Thus intent remains even there. True exceptions, the same note says, are statutory offenses, such as statutory rape where the accused could have no reasonable knowledge of the age of a sexual partner yet is convicted anyway (one might question the moral legitimacy of that, but it wasn’t the issue decided by the Court in this case).

Accordingly, in this case the Supreme Court concluded the following regarding the prosecution’s claim that intent was not necessary for guilt:

The Government asks us by a feat of construction radically to change the weights and balances in the scales of justice. The purpose and obvious effect of doing away with the requirement of a guilty intent is to ease the prosecution’s path to conviction, to strip the defendant of such benefit as he derived at common law from innocence of evil purpose, and to circumscribe the freedom heretofore allowed juries. Such a manifest impairment of the immunities of the individual should not be extended to common law crimes on judicial initiative.

[…And for that and various other reasons] we hold that mere omission from [a statute] of any mention of intent will not be construed as eliminating that element from the crimes denounced. [And accordingly…] the law under some circumstances recognizes good faith or blameless intent as a defense, partial defense, or as an element to be considered in mitigation of punishment. [Even] treason—the one crime deemed grave enough for definition in our Constitution itself—requires not only the duly witnessed overt act of aid and comfort to the enemy but also the mental element of disloyalty or adherence to the enemy. [Morissette, p. 263]

The conclusion is therefore quite clear. The Supreme Court rejected (and even deemed well-settled as almost universally rejected) such contra-causal/libertarian-free-will notions as retributive and retaliative justice (i.e. vengeance) as purposes of assigning guilt and instead upheld the distinctly compatibilist purposes for assigning guilt: the deterrence and reform of those of vicious will.

Key to this construction, in fact, is a compatibilist notion of free will: not only do deterrence and reform imply an assumption of causal determinism (because in this scheme judging the guilty must be believed to have causal effects on people: to cause them to refrain from crime, or to cause them to change their ways), but more clearly than that, by explicitly linking the concepts of “free will” and “freedom of the will” to the requirement of establishing intent to prove guilt, the U.S. Supreme Court (and all its cited legal experts and precedents) is making clear that a will is free precisely when it results in a decision that can be causally construed as evidence of intent, and is not free precisely when it results in a decision that cannot be construed as evidence of intent.

This is why a claim of duress, for example, exonerates the guilty: even though someone who has a gun to their head can certainly choose to be killed rather than do what they are being commanded at gunpoint to do, courts universally recognize that such actions cannot be evidence of a defendant’s vicious will, since they can only be evidence of a defendant’s wish not to be killed by the one holding the gun—the vicious will in that case is that of the one holding the gun, and they are held accountable for what results. Their will is in that case being substituted for the defendant’s, therefore the defendant’s actions are not evidence of the defendant’s criminal intent, but the criminal intent of the one coercing them.

Similarly, insanity and other defenses that exonerate the guilty all have as their defining component a presumption that, if that condition did indeed obtain, then what the defendant did cannot be considered evidence that they have a vicious will (mens rea) and therefore cannot be used as grounds to punish them for having a vicious will [thus even diagnosed psychopathy never produces a valid insanity defense, and schizophrenia does only if, for example, the psychotic hallucinations it produces tricked someone into doing something bad without knowing it was wrong, a key point to understand in the next section].

The fact that the Supreme Court declares the punishing of a vicious will (and not the mere doing of harm) as the function of the law entails embracing a compatibilist notion of free will, whereby what makes a will free is not a will being free of causation, but a will being free of interference, such that what a person chooses to do can be held as evidence of what their true will was.

And I would argue this is likewise the function of most moral judgments generally. This is what we mean by accountability and responsibility, and this is the purpose of praise and blame: to cause people’s intentions to align with what is necessary for a civil and functional society, through the causal mechanisms of either deterrence, persuasion, restraint, or reform—and to communicate to others evidence of a person’s character.

Laws and morality, therefore, serve the dual function of identifying and evaluating people’s characters and intentions, and in turn regulating them (to whatever end is deemed the proper end of law and morality, which is a separate debate beyond the scope of the present point). And in law we use such terms as “free will” and “freedom of the will” and “free choice” and “freedom of choice” etc. as tools for determining when one’s will is and is not evidence of their character and intentions.

Battered Woman Syndrome as a Legal Defense

Illustrating these points is the use of BWS as a defense at law. For a handy introduction to the idea, see “Criminal Prosecution of Battered Native Women for Failure to Protect” (a 2005 Joint Report of the Southwest Center for Law and Policy and Office on Violence Against Women, U.S. Department of Justice), because it provides a good working case for discussing boundaries, it covers various kinds of affirmative defense, including “duress,” and it illustrates the kind of reasoning we engage in when connecting freedom of the will to identifying injustice in the legal system. Another resource on this issue, which discusses the creation of a “Battered Woman’s Syndrome” defense in law and some of the legal battles and principles relating to its implementation, is Scott Baker, “Deaf Justice? Battered Women Unjustly Imprisoned Prior to the Enactment of Evidence Code Section 1107,” Golden Gate University Law Review 24.1 (1994), pp. 99-130 [full text here as a somewhat clunky PDF].

The term of course is outdated. Gender neutral versions of the concept (e.g., “Battered Person Syndrome”) have since begun entering the law, in some cases by statute, in others by judicial ruling; for recent and more thorough discussion, see the kindle book by Brenda Russell, Battered Woman Syndrome as a Legal Defense: History, Effectiveness and Implications (2010). But here I shall quote for analysis the relevant section of Baker (citations now omitted and some emphasis added):

“Learned helplessness” occurs when the battered woman cannot predict her own safety because regardless of her conduct she is faced with the batterer’s random and unpredictable abusive behavior. Although a battered woman does not necessarily learn to be helpless, she often discovers that she is unable to predict the effect her behavior will have. “Learned helplessness” causes a person to choose “behavioral responses which will have the highest predictability of an effect within the known, or familiar, situation; they avoid responses—like escape, for instance—that launch them into the unknown.” [p. 104]

…

Under the California Penal Code, the classification for a homicide committed in self-defense is a justifiable homicide, as opposed to an excusable homicide. Excuses for homicide include accident, misfortune, sufficient provocation, insanity, and provocation, and a genuine, but unreasonable, belief in the need for self-defense. Justification declares the allegedly criminal act legal, while excuse admits the act’s criminality but declares the allegedly criminal actor not to blame.

Two types of self-defense exist in California: perfect self-defense and imperfect self-defense. Perfect self-defense requires both subjective honesty and objective reasonableness, and completely exonerates the accused. Subjective honesty evaluates the defendant’s state of mind. Objective reasonableness questions whether a reasonable person would have believed and acted as the defendant did. Imperfect self-defense requires only subjective honesty and negates malice aforethought, reducing the homicide to voluntary manslaughter. Both perfect and imperfect self-defense require an honest belief that the killer is in imminent danger of death or great bodily injury from the victim.

BWS [i.e. Battered Woman’s Syndrome] expert testimony is useful in the defense of battered women accused of killing their abusive spouses because it can explain why a particular battered woman subjectively believed that her self-defense was necessary. Because BWS can illustrate why a particular battered woman had an honest belief of imminent danger, it should be used in the objective reasonableness test to show why a battered woman believed and acted as she did.

In [the] Aris [case], the defendant, a battered woman, appealed a second degree murder conviction because the trial court excluded BWS expert testimony which would have revealed that BWS affected her mental condition at the time of the killing. On the night of the killing, Brenda Aris’ abusive husband beat her and threatened that he did not think he would let her live until the next morning. When her husband fell asleep she went next door for some ice for her injuries. She found a gun and took it for protection. She thought that when she returned home she would again be beaten. Aris went to the bedroom, sat on the bed, and shot her sleeping husband five times in the back.

In Aris, the court held that expert BWS testimony about a defendant’s state of mind was not relevant to the objective reasonableness of the defendant’s actions in self-defense. The court found that the defendant presented no substantial evidence that a reasonable person under the same circumstances would have perceived an imminent danger and a need to kill in self-defense. The court believed no reasonable jury could conclude that a sleeping victim presents an imminent danger of great bodily harm, especially when the defendant left the bedroom and subsequently returned to shoot the victim. Nevertheless, the court also held that it was error not to permit an expert to testify about the defendant’s particular experiences as a battered woman and its effect on her perceptions of danger, its imminence, and what actions were necessary to protect herself. The court found that the error was harmless because Aris’ own testimony showed that there was nothing in the victim’s behavior indicating the existence of imminent danger.

Relying on the expert testimony of Dr. Lenore E. Walker, the court stated that a woman who has been battered and then threatened with more abuse is more likely to perceive the danger involved faster than one who has not been abused. The court found that a battered woman accurately senses when an abusive episode is not yet over. However, the court cautioned that the jury may misuse BWS evidence to establish the reasonableness requirement for perfect self-defense. Upon request by the prosecution, the judge should instruct the jury that self-defense testimony is relevant only to prove the honest belief requirement for both perfect and imperfect self-defense, not to prove the reasonableness requirement for perfect self-defense.

Two years after Aris was decided, another court held that failure to present evidence of BWS constituted ineffective assistance of counsel. In People v. Day … the prosecution used many of the stereotypes and commonly held misconceptions about battered women, such as the stereotype and misconception that battered women can easily leave their batterers. The prosecution used this basic misconception to challenge the defendant’s credibility before the jury by asking the question, “why did she not simply leave?” In fact, the defense counsel characterized the defendant’s relationship with her batterer as “mutual combat.” The misconception about battered women is that they all behave in the same way and if a woman dares to fight back she removes herself from the category of women for whom BWS is available as a defense.

On appeal, a psychologist and authority on BWS submitted an affidavit supporting appellant’s motion for a new trial. The affidavit dispelled many myths associated with BWS which were used by the prosecution to question defendant’s credibility during the trial. The affidavit established that battered women do employ active self-defense as a strategy. The court held that evidence explaining BWS informs the jury that how they think the average reasonable person would behave and how the jury thinks they personally would behave are not necessarily the same way that people who have been battered in fact behave.

The court understood that when deciding a case where a battered woman kills her batterer, the jury will ask “what would I do in that situation?” Most people who are unaware of BWS would naturally answer this question, “I would just leave.” BWS testimony helps explain why the battered woman does not leave and helps a jury understand the psychological effects battering has on a battered woman. BWS theory explains how a battered woman can accurately sense that danger still exists and why her subjectively honest belief is that she must act to save her life.

Here two distinctions are being made relevant: (1) it is not argued that just any causal factor that influences a battered woman’s decision to kill her batterer is exonerating (as if she were to be excused simply because she was causally determined to act), but rather that a certain sustained and systematic environment of abuse substantially alters a battered woman’s state of mind in such a way that it no longer reasons as ordinary minds would (hence the jury is to be instructed not to judge by the standard of what they would do in the circumstance, but by the standard of what a woman forced into that particular mental state can reasonably be expected to do in the circumstance) and (2) this is exonerating solely because it removes the causal conditions that would otherwise entail a “vicious will.”

In other words, a woman in such conditions is not acting with mens rea, but in reasonable fear of her life, “reasonable” in this case being measured by the standards of her altered mental state, an altered mental state only persons who have been victimized by sustained abuse will find themselves in. Accordingly, this is a defense limited to a specific set of causal circumstances. Why is this set of causes exonerating and not others? Because what exonerates a will from blame under the law is evidence that the otherwise-criminal act (actus reus) was not produced by a criminal intent (mens rea) but by an intent recognized as morally or legally or socially acceptable.

Therefore, the issue in cases like this is not that causal determinism exonerates agents from blame, but that decisions that do not evince a criminal mind exonerate agents from blame. Free will in this case is a will that exhibits a sound connection between the action and its intent, as opposed to a will that does not. For example, a coerced will is where the coerced’s will is being replaced by the coercer’s will; the action the coerced takes therefore does not indicate a criminal mind in the coerced, only in the coercer. Or an insane will, where, for example, the insane’s will is being replaced by an altered will that is making decisions based on a delusion that was not itself willfully chosen, such that within the context of that unavoidable delusion, the actor’s decision makes sense and is not criminal. As, for example, killing someone in a delusional state whereby you genuinely believed they were trying to kill you: such an act does not evince a vicious will and therefore is not found guilty. Causation is not the issue here. The link between a choice made and the chooser’s intent (and hence their character, personality, “attitude toward society,” and so on) is the issue. This is what makes American law intrinsically compatibilist in respect to free will.

In the case of BWS, we are in a circumstance similar to an insanity defense, whereby a battered woman delusionally believes her life is in more immanent danger than it objectively is (e.g. by believing that fleeing would ultimately result in her death and staying would ultimately result in her death, leaving no perceived alternative). Thus, BWS does not provide what California law calls a “perfect” self-defense (which requires objective reality to match subjective belief; objectively, a battered woman can just leave). But it does provide an “imperfect” self-defense, by establishing the act was not the product of criminal intent (or to be more precise, under American law, by establishing that the act cannot be proved beyond a reasonable doubt to be the product of criminal intent).

There are other aspects of BWS and its relation to free will (legally and morally), even beyond those discussed in the Joint Report. But in all cases it reduces to the same set of analyses. Overall, BWS is similar to being in a state of slavery as defined in United States v. Kozminski (the case I discussed in Sense and Goodness without God, pp. 111-13): chains and bars are not necessary, all that is necessary is the legitimate cognitive fear that escape is impossible or too dangerous to attempt (regardless of whether it objectively is). And altered mental states can create such a legitimate cognitive fear. Thus, we recognize that fact in law, and generally in moral judgment as well.

In this and many other ways we can see that the American legal system predominately adopts a compatibilist understanding of free will, and even helps us understand just what a “free will” really and meaningfully is, both legally and morally. And it is not “being free of all possible causation.” Sam Harris, along with Christians obsessed with contra-causal free will, are simply wrong about that.