Join us for the Texas Law Review Association Banquet on Saturday, April 7, 2018 Learn More

Addiction, Criminalization, and Character Evidence

Note - Volume 96 - Issue 3

A drug addict’s addiction is no defense to drug crimes. Criminal law rejects the disease model of addiction, at least insofar as the model would inform the Eighth Amendment, the voluntary act doctrine, or the insanity defense. This Note does not take issue with the criminalization of addiction, arguing merely that it should preclude evidence law’s treatment of addiction as something other than immoral.

The rule against character evidence precludes evidence of an immoral propensity when offered to prove action in conformity with that propensity. But in prosecutions for property crimes, courts routinely admit evidence of a defendant’s addiction on the theory that it proves a motive, not an immoral propensity. The law’s rejection of the disease model, though, teaches that an addict will only decide to acquire and use drugs if she succumbs not to an irresistible compulsion, but to a temptation to do wrong.

Criminal law’s treatment of addiction should have force in the law of evidence because, right or wrong, and among other reasons, criminalization teaches jurors that action in accordance with addiction is immoral. And the prejudice that arises from a perception of the defendant as prone to immorality is precisely the reason we have a rule against character evidence.


In 1985, Paul Goldstein theorized an “economic compulsive” model for the nexus between drug use and property crime.[1] His premise was simple: “drug users engage in economically oriented violent crime, e.g., robbery, in order to support costly drug use.”[2] Today, a substantial body of research questions the simplicity of the theory that drug use causes non-drug-related crime.[3] But some correlation between drug use and property crime remains undeniable. Compared to those who do not use drugs, drug users are several times more likely to commit crimes.[4] And as many as 30% of state prisoners convicted for property crimes acknowledge having committed their offenses to fund drug purchases.[5]

The economic-compulsive model makes intuitive sense. A drug user, particularly one who becomes addicted, has one more reason to steal than the average person has. And an addict’s additional motive increases the probability that she will commit property crimes.

Evidence of a defendant’s addiction is thus logically relevant in prosecutions for property crimes. Addiction evidence makes it at least somewhat more likely that the defendant, and not a non-addicted person, committed a given property crime.[6] In some instances (theft of narcotics, for instance), addiction evidence may be particularly logically relevant. But however relevant it may be, addiction evidence should be excluded under the rule against character evidence.

The rules of evidence prohibit proving a person’s “character” in order “to prove that on a particular occasion the person acted in accordance with” that character.[7] And while evidence of a defendant’s specific acts may be admissible for a non-character purpose, including “motive,” specific-acts evidence cannot be used to prove character.[8]

To avoid the difficulty of defining “character,” this Note adopts the widely accepted core of its definition: a propensity to do something moral or immoral.[9] The question then becomes whether addiction—and more specifically, acting in accordance with one’s addiction—is immoral.

Admittedly, the disease model of addiction has become more and more accepted. Both the medical and psychological communities have recognized addiction as a disease for decades.[10] And contemporary research indicates that some combination of genetics and environmental influence causes addiction.[11]

But notwithstanding the expert consensus to the contrary, courts do not adhere to the disease model. Under the law, either addiction is not a disease, or despite its disease-like characteristics, only the immoral are susceptible to its predominant symptoms: acquiring and using drugs. In 1968, the Supreme Court rejected the argument that “alcoholism is caused and maintained by something other than the moral fault of the alcoholic.”[12] And criminal law has refused to incorporate the disease model of addiction ever since. An addicted person cannot defend herself on the ground that the use of narcotics by an addicted person is involuntary.[13] Neither can she argue that addiction is a mental defect rendering her substantially incapable of conforming her conduct to the law.[14]

And yet, in the context of evidence, courts treat addiction as something other than an immoral propensity. Specifically, they treat it as non-character evidence admissible to prove the accused’s motive to commit crimes against property.[15] Courts routinely hold that addiction evidence is not character evidence because it proves a motive to commit property crimes, not a propensity to commit them.[16]

The logic of admitting addiction evidence as probative of a motive, and not a propensity, is fundamentally flawed. Linking addiction to property crimes depends on the intermediate inference that addicted persons will likely[17] succumb to their addictions. The accuracy of that inference notwithstanding, in the eyes of the law, someone who succumbs to an addiction succumbs to immorality. And the rule against character evidence purports to exclude evidence that depends for its logic on a person’s propensity for immorality.

This Note proceeds in four parts. Part I introduces character evidence, which is prohibited, and motive evidence, which is admissible when offered for a non-character purpose. And it defends the position that the character rule prohibits an immoral-propensity inference at any point in the chain of inferences that renders the evidence logically relevant.

Part II surveys courts’ treatment of addiction evidence, observing that those admitting addiction evidence omit any discussion of the necessary inference that addicts will conform to their addictions. Part II also observes that most courts excluding addiction evidence do so in light of its prejudicial effect, not because it fails the character rule. And those few courts that have excluded addiction evidence under the character rule have conflated weak evidence with character evidence.

Every court to have addressed addiction evidence, including those that have excluded it, has overlooked the most compelling reason that it should be excluded. Part III recounts the development of criminal law’s decisive stance that addicted persons exercise moral agency in deciding whether or not to use drugs. In the eyes of the law, when an addict acquires and uses narcotics, she acts immorally.

Part IV defends the proposition that criminal law’s addiction-related judgments should have force in the law of evidence. It then puts addiction evidence to the test and concludes that its use to establish a motive to commit property crimes violates the rule against character evidence.

Part I

This Part provides an introduction to character evidence, motive evidence, and the justifications for excluding the former while admitting the latter.

A. The Prohibition of “Character” Evidence

The rules of evidence prohibit using evidence of “character” to prove action in accordance with that character.[18] But the rules do not define “character,” and a definition of the term is notoriously elusive.[19]

In light of this difficulty, this Note adopts a definition of character safely within the limits of the term: evidence is character evidence if its relevance depends on a propensity for morality or immorality.[20] That is, not all propensity evidence is character evidence.[21] Evidence of habit, for instance, proves a person’s habitual propensity but is nevertheless readily admissible.[22] Habit establishes an involuntary propensity, independent of moral assessment, while character establishes a “general and morally tinged propensity.”[23] Similarly, evidence of a person’s intellect may prove a tendency to act in a certain way. But intellect is not a moral attribute, so it survives the character rule.[24]

The reasons we prohibit character evidence further reveal an emphasis on morality. Those reasons have nothing to do with the relevance of a defendant’s character, which is nearly always probative of guilt or innocence.[25] (Put simply, a defendant is more likely to have acted in a certain manner if that action accords with her character.)[26]

We exclude character evidence not because it lacks probative value, but because it implicates the defendant’s morality. While justifications for the exclusion may also include the risks of delay and confusion,[27] the character rule is predominantly concerned with the risk of unfair prejudice.[28] Such prejudice may take either or both of two forms: “inferential error prejudice,” where a jury assigns undue weight to character evidence, or “nullification prejudice,” where a jury convicts a defendant for being a bad person or having done bad acts other than the crime charged.[29]

In either case, the risk of prejudice arises from the moral overtones of character evidence.[30] Psychologists observe that we give greater weight to negative traits than analogous positive traits,[31] and so evidence of immorality bears a substantial risk of inferential-error prejudice. And with respect to nullification prejudice, we need not fear that juries will punish a defendant for a morally neutral character (a purely biological trait, for instance), as opposed to an immoral one.[32]

B. The Admissibility of “Motive” Evidence

While prohibiting evidence of specific acts offered to prove character,[33] the evidence rules endorse specific-acts evidence offered for “another purpose,” including proof of “motive.”[34] The admissibility of specific-acts evidence for purposes other than character is as prevalent and hotly contested as any issue in evidence law.[35] And a substantial portion of other-purposes cases concern evidence offered on a motive theory.[36]

Unlike character’s, the judicial definition of motive is relatively straightforward. A frequently cited definition is “the reason that nudges the will and prods the mind to indulge the criminal intent.”[37] Other definitions add that a motive may be emotional in nature, not merely cognitive.[38] In general, the meaning of motive in evidence law tracks the common usage of the term: “something (as a need or desire) that causes a person to act.”[39]

Why, though, admit motive evidence and not character evidence? Evidence of a defendant’s motive makes it more probable that the defendant is the perpetrator.[40] And the idea is that a person will act on a motive regardless of whether she has a good or bad character.[41] The jury can thus infer the defendant’s guilt without making an inference from propensity. It can simply find it more likely that the defendant is the perpetrator because (1) the perpetrator almost certainly had a motive to commit the crime charged; (2) not all, and perhaps not most, people had a motive to commit the crime charged; so (3) it is more likely that the defendant, compared to a person chosen randomly, is the perpetrator.[42]

To avoid a character-evidence problem, it must be true that a jury can find that the defendant had a motive without making an inference from the defendant’s character. This depends on the lack of moral judgment involved in motive evidence.[43] As such, motive evidence only survives the character rule if the existence of a motive does not arise out of the defendant’s moral fiber, or lack thereof.[44]

And whenever motive evidence is admissible for its non-character purpose, it remains subject to the probative value–prejudicial effect balance of Rule 403.[45] Particularly difficult is evidence relevant both for a prohibited character purpose and a permissible non-character purpose, such as motive. No “mechanical solution” can resolve this difficulty.[46] But where the risk of inferential-error or nullification prejudice is sufficiently high, the probative value of motive evidence must not be “substantially outweighed” by that risk of prejudice.[47]

C. The Chain of Inferences

The rules of evidence are commonly understood to preclude character inferences at any point in the chain of logic that leads from the evidence to the fact to be proved.[48] Some courts expressly require that the proponent “articulate a proper chain of inferences unconnected to character,”[49] lest they risk admitting “‘logical’ but forbidden inferences that disguise propensity and character as something else.”[50] The question is not merely whether the evidence is logically relevant for a non-character purpose, but precisely how the evidence is relevant to that purpose.[51]

Consider a prosecution of a felon who allegedly possessed a firearm. The accused defends himself by claiming that he was not aware that he possessed it. In rebuttal, the prosecution offers evidence of the accused’s two prior convictions for distribution of narcotics. The prosecution theorizes that the market for drugs is a dangerous one and, thus, that the accused had a motive to possess a firearm. A motive for possession would indicate that the accused knowingly possessed.[52] The chain of inferences, however, requires an intermediate character inference:

Evidence: The defendant sold narcotics twice in the past.

Inference 1: The defendant likely continued to involve himself in the illicit drug market through the date in question.

Inference 2: The defendant thus had a stronger motive to possess a firearm than an average person would.

Conclusion: The defendant is somewhat more likely than an average person to have knowingly possessed a firearm.

In this hypothetical, the motive theory’s logic depends on the initial inference, but that inference depends on the accused’s propensity to do something immoral. A court should thus exclude evidence of the two narcotics convictions.

This makes sense. The risk of inferential-error and nullification prejudice arises with equal force whether the character inference is intermediate or ultimate. If the fact finder assigns undue weight to the intermediate inference that the accused acted in conformity with her character, then the ultimate inference is assigned undue weight in turn. And a fact finder that would punish a defendant for his bad character or past bad acts will, presumably, do so regardless of what the bad acts or character are offered to prove.[53]

Admittedly, many courts fail to put each link in the inferential chain to an actual character-evidence test.[54] Regardless, this Note takes the rule for its plain meaning: evidence cannot be used to show motive if its logical relevance requires an inference “that on a particular occasion the person acted in accordance with [her] character.”[55] It should make no difference where in the chain of inferences a “particular occasion” occurs, so long as the initial character inference is essential to arrive at the subsequent motive inference.

Part II

Addiction-as-compulsion evidence is commonplace.[56] It falls into two categories: specific motive and general motive. In the former, addiction evidence proves that an addicted defendant had a motive to steal or otherwise illegally obtain the narcotic to which she is addicted. The theory of such specific-motive evidence is straightforward: most people do not want narcotics.[57] So proof that the defendant is amongst the small percentage of persons who would be motivated to acquire them is probative of the perpetrator’s identity.[58]

General-motive evidence is more attenuated. Courts admit it on the theory that addicts needs money to fund their addictions. Thus, they have an above-average likelihood of a motive to steal. From such a motive, the fact finder may draw an inference that the perpetrator’s identity and the defendant’s are one and the same (i.e., an inference of guilt).[59]

This Part provides a brief survey of the reasoning by which courts admit addiction evidence. Most do. It then discusses the reasons that other courts have excluded addiction evidence. A number have found addiction evidence inadmissible due to its minimal probative value and substantial prejudicial effect. But few have gone so far as to hold that addiction evidence, by its very nature, is character evidence. And those that have excluded addiction evidence under the character rule have done so unconvincingly.

A. Admitting Addiction Evidence

The theory for admitting addiction evidence as probative of a specific motive bears little explanation. The Seventh Circuit’s reasoning for allowing the admission of evidence of a nurse’s prior addiction to Demerol, in a prosecution for theft of Demerol, is representative: “Most people don’t want Demerol; being a Demerol addict gave [defendant] a motive to tamper with the Demerol-filled syringes that, so far as appears, none of the other nurses who had access to the cabinet in which the syringes were locked had.”[60] In such instances, the defendant’s addiction is proof of her identity as the perpetrator.[61]

The real prosecutorial value of addiction evidence lies in proving a general motive to commit any and all valuable crimes. Presumably, wealthy addicts do not have an economic-compulsive motive to commit property crimes, and neither would addicts with at least a steady source of income in excess of their regular narcotics expenditures. But many courts do not require an independent showing of financial distress, admitting naked evidence of drug addiction even in prosecutions for the most severe crimes against property.[62]

These courts have referred to addiction evidence as “extremely”[63] or “highly”[64] probative of a motive to commit property crimes. The Seventh Circuit, for instance, treated the probative value of addiction evidence as self-evident:

Admission of [co-conspirator’s] testimony that he and [defendant] used heroin together since their freshman year of high school does not rise to the level of plain error. Despite the fact that the robbery occurred approximately five and a half years after the two had been in the ninth grade, the district court found that the evidence of [defendant’s] drug addiction was relevant to establish [defendant’s] motive to commit the robbery (in all probability so as to finance his serious drug habit of some five years).[65]

This is not atypical. The First Circuit found the inference from use of cocaine to a motive to engage in heroin trafficking similarly self-explanatory: the defendant “used cocaine, an expensive substance, and . . . trafficking in heroin could provide the money to buy it. This is certainly evidence from which a jury could reasonably find ‘motive’ to commit the crimes charged.”[66] The Supreme Court of Ohio reasoned even more succinctly that “[a]ppellant’s drug addiction and use shows his need for money and, hence, his motive to steal and kill.”[67] And, in turn, these courts tend to consider the prejudicial effect of addiction evidence to be judicially manageable.[68]

Not all courts are so permissive. The Fourth Circuit in United States v. Madden,[69] rejected the admission of naked addiction evidence and required additional showings. It accepted the “obvious proposition” that addiction evidence proves “a logical motivation to commit bank robbery to generate the cash necessary to support the habit,”[70] but it required that the prosecution also show that the accused’s addiction was “significant” and that the accused “did not have the financial means to support” his addiction.[71]

With some variation in the tests themselves, several courts join the Fourth Circuit in requiring more than mere evidence of drug use.[72] Others, though less explicit in formally requiring evidence of financial need, note that it is an integral justification for the inference of economic compulsion.[73]

One might argue that proof of an economic motive is superfluous in prosecutions of property crimes—that is, no reasonable person would question why someone would rob a bank. But the purpose of motive evidence is not to prove the motive—its purpose is to prove the identity of the perpetrator.[74] In State v. Hughes,[75] the defendant argued that evidence of his drug addiction “was not relevant to establish a motive for the burglary and robbery because the motive is inherent in the commission of those crimes.”[76] In response, the Hughes court explained that “regardless of whether it was the primary motive—money—or the secondary motive—buying drugs,” proving a motive supports the “inference that [defendant] participated in the crime.”[77]

Importantly, for this Note’s purposes, these courts confine the character rule to prohibiting only evidence that proves a propensity to commit the specific crime charged, excluding motive evidence only where the charged crime “is motivated by a taste for engaging in that crime or a compulsion to engage in it,” not by “a desire for pecuniary gain.”[78] As one court reasoned, addiction evidence does not “create a danger that the jury would conclude that [the defendant] had a propensity to commit the home invasions, because drug use and home invasions involve completely different acts.”[79] So long as the addiction evidence is not admitted to establish that the defendant is prone to committing narcotics offenses, it is admissible as evidence of a motive to commit property crimes.[80]

Courts that admit addiction evidence make no mention of the necessary intermediate inference that an addicted person will probably purchase drugs. They reason that because addiction evidence shows a motive “to purchase more drugs,” it is not admitted for the “improper purpose of showing a propensity for criminal behavior.”[81] But the motive theory does not work without the defendant’s propensity to purchase narcotics, which is, of course, criminal. Put simply, unless an addicted person will commit narcotics offenses, she will not need money to commit narcotics offenses. The failure to address the intermediate inference from addiction evidence to the likelihood of drug purchases is the crux of this Note, as elaborated in Part IV.

B. Excluding Addiction Evidence Under Rule 403

Most courts that have excluded addiction evidence have done so under Rule 403.[82] Unsurprisingly, they find the prejudicial effect of addiction evidence substantial, particularly the risk of nullification prejudice. In the midst of the narcotics-fueled crime wave of the 1980s, for instance, the Supreme Court of California referred to the “impact of narcotics addiction evidence ‘upon a jury of laymen’” as “catastrophic.”[83] Of course, the nation’s drug epidemic continued (and continues). More than a decade later, the Fourth Circuit remarked on the risk of prejudice at the hands of “a jury in a big city ravaged by the deadly scourge of drugs and their attendant ills.”[84]

Several courts have focused less on the prejudicial effect of addiction evidence than on its minimal probative value. The Eighth Circuit described such reasoning persuasively:

We cannot say that the slight probative value of knowing one possible motive for Mr. Sutton to commit a robbery outweighs the likely prejudicial effect on the jury of being told that the defendant was a crack-cocaine user. In any event, it could hardly come as a surprise to the jury that Mr. Sutton was robbing a bank because he needed money for some reason. . . . This brings to mind the story of a more famous bank robber with the same surname. When asked why he robbed banks, Willie Sutton replied, “That’s where the money is.”[85]

Similarly, the Supreme Court of Washington reasoned that if addiction evidence proves a motive to commit property crimes, it does not sufficiently narrow the suspect pool to fairly prove identity.[86]

Here, it is worth briefly noting that courts excluding addiction evidence under Rule 403 are right to do so.[87] To put the matter plainly, the probative value of general-motive addiction evidence is minimal. It supports the inference that the defendant is the perpetrator, but that inference is weak—the suspect pool remains large.[88] As Professor Imwinkelried writes, “It is ideal if the defendant is the only person with such a motive. At the other extreme, if the motivation is almost universal . . . , proof of the motive has little or no probative value on the issue of identity.”[89]

If anything is almost universal, it is a desire for money.[90] This Note can do no better than to recite the observations of Edwin Sutherland, one of the preeminent criminologists of the twentieth century:

While criminal behavior is an expression of general needs and values, it is not explained by those general needs and values, since noncriminal behavior is an expression of the same needs and values. Thieves generally steal in order to secure money, but likewise honest laborers work in order to secure money.[91]

Sutherland continues by explaining that the “money motive” is amongst those theories that are “futile” with respect to explaining crime.[92] Such motives “are similar to respiration, which is necessary for any behavior but does not differentiate criminal from noncriminal behavior.”[93] And the risk that juries will punish defendants for their addictions (likely greater than courts tend to acknowledge[94]) outweighs such minimal probative value.

C. Excluding Addiction Evidence Under Rule 404

More interesting, for this Note’s purposes, are those rare courts that have excluded addiction evidence on the grounds that it is—by its nature—character evidence. Their argument is a functional one: proving a general motive to commit any and all property crimes, in effect, proves a propensity for committing crimes. A New Jersey court, in State v. Mazowski,[95] held that proof of a general motive was “indistinguishable from a claim that defendant has a ‘disposition,’ or general propensity to commit crimes . . . .”[96] That is, the admissibility of motive evidence does not permit proof of “a characteristic or condition (drug addiction) which makes defendant likely to commit crimes.”[97]

Similarly, the Supreme Court of New Hampshire prohibits the use of addiction evidence when offered to prove the perpetrator’s identity on a general-motive theory. In State v. Costello,[98] it explicitly cabined the admissibility of addiction evidence to instances in which there is ample other evidence of identity. In the case at bar, the addiction evidence could be used to bolster the prosecution’s theory of the case by “supply[ing] the jury with the defendant’s motive for an otherwise senseless crime,” but only because the prosecution had already sufficiently proved the defendant’s guilt.[99] Had the defendant’s addiction been offered to prove identity, the court would have excluded it.[100] Again, the court held proof of a general motive to be functionally, if not technically, character evidence:

The resulting inference imparted to the jury is: because the defendant is a drug addict he has the general intent to steal, and because drug addicts steal, it is safe to conclude that this particular drug addict is the unknown culprit in this case. Such reasoning allows the impermissible inference of propensity that the rules of evidence are designed to prevent.[101]

This Note agrees that addiction evidence should be excluded under the character rule. But the Mazowski and Costello courts have gone about it wrong. The position that addiction evidence functionally violates the character rule conflates the character rule with Rule 403. The character rule permits non-character motive inferences—and it does not distinguish strong motives from weak ones. Certainly, general-motive evidence proves identity to a lesser extent than does specific-motive evidence. But that goes to the Rule 403 inquiry, not to the facial prohibition of character evidence. That is, evidence does not become character evidence simply because it narrows the suspect pool less than other evidence might. It should not matter that general-motive evidence fails to provide a motive to acquire money in the particular way the charged crime was committed.

And restricting general-motive evidence to proof of something other than identity is even stranger. If addiction evidence cannot prove identity, it has no place in property crime cases.[102] If, say, a defendant admits to stealing property but disputes doing so intentionally, the defendant’s motive may prove intent.[103] But there does not appear to be a single property crime case in which the prosecution admitted addiction evidence in response to a lack-of-intent defense.

We need not, however, delve further into the prevailing approaches to addiction evidence. Each court that has excluded addiction evidence has missed the most compelling justification for doing so. The basic tenet of the economic-compulsive theory is that addicts will do what they must to acquire drugs.[104] And reliance on that tenet violates the character rule because it assumes that an addict is prone to acquiring drugs. The following Part considers criminal law’s treatment of addiction, and specifically its teaching that an addict makes an immoral choices when she acts on her addiction.

Part III

As it currently stands, criminal law rests on the determination that addicts can choose whether or not they will act in accordance with their addictions. The proper effect of that determination on the rules of evidence will be addressed in Part IV. This Part surveys the law of addiction outside the law of evidence.

A. Addiction at the Supreme Court

For a few years in the early 1960s, it appeared as though the law might shift in favor of treating drug use by an addicted person as an irresistible symptom of a disease. In Robinson v. California,[105] the Supreme Court invalidated, under the Eighth Amendment, a California statute making it a misdemeanor for a person “to be addicted to the use of narcotics.”[106] The Court reasoned that addiction is “an illness which may be contracted innocently or involuntarily.”[107] Thus, the California law might as well have imposed criminal liability for being “mentally ill, or a leper, or . . . afflicted with a venereal disease.”[108] And although the statute imposed a maximum of ninety days’ imprisonment, “[e]ven one day in prison would be a cruel and unusual punishment for the ‘crime’ of having a common cold.”[109]

Justice Stewart, writing for the Court in Robinson, distinguished the statute at issue from one imposing criminal liability for the “use of narcotics, for their purchase, sale or possession.”[110] But in a concurrence surveying and relying on the disease model of addiction, Justice Douglas indicated that any punishment of the symptoms of drug addiction (i.e. the use of narcotics) would not survive Eighth Amendment scrutiny.[111] He expressly rejected the argument that addicts are “in the category of those who could, if they would, forsake their evil ways.”[112] To the contrary, he concluded that an “addict is under compulsions not capable of management without outside help.”[113]

But six years later, the Court put to rest any fear that the law might embrace the disease model. In Powell v. Texas,[114] the Court upheld, as applied to an alcoholic, a Texas law imposing criminal liability for public drunkenness. In so holding, it rejected the argument that Robinson precludes punishing an addict “for being in a condition he is powerless to change.”[115] Its reasoning summarily rejected the addiction-as-compulsion reasoning of Robinson:

We are unable to conclude, on the state of this record or on the current state of medical knowledge, that chronic alcoholics in general, and [petitioner] in particular, suffer from such an irresistible compulsion to drink and to get drunk in public that they are utterly unable to control their performance of either or both of these acts and thus cannot be deterred at all from public intoxication.[116]

Although a four-Justice dissent emphasized the medical community’s belief that chronic alcoholics are “powerless to avoid drinking,”[117] Powell is understood to have rejected Robinson’s insertion of the disease model into the Eighth Amendment.[118] Combined, Robinson and Powell hold that the law may not punish for the status of addiction but that addicts have no “constitutional involuntariness defense.”[119]

And increased acceptance of the disease model of addiction notwithstanding, the Supreme Court has held firm in its refusal to incorporate the disease model into the law. In 1988, the Court rejected an argument that alcoholism is not always the product of “willful misconduct” and thus should not facially preclude the extension of veterans’ statutory time limit to file for benefits.[120] In so holding, it endorsed both the position that alcoholism is not a disease[121] and the position that even if “alcoholism [were] a ‘disease’ to which its victims are genetically predisposed, the consumption of alcohol is not . . . wholly involuntary.”[122] Most recently, the Court upheld a Montana law that prohibited introduction of voluntary-intoxication evidence in criminal trials.[123] Justice Scalia wrote for the majority that “the rule comports with and implements society’s moral perception that one who has voluntarily impaired his own faculties should be responsible for the consequences.”[124] The Court made no reference to whether intoxication is a choice for all persons, apparently continuing to operate under the assumption that it is.

B. Addiction as Voluntary Act

The constitutional question settled, challenges to the criminalization of drug use by drug addicts next arose in the context of the voluntary act doctrine. In most U.S. jurisdictions, criminal liability may not lie unless the criminal act is “a product of the effort or determination of the actor”—such effort or determination, however, may be “either conscious or habitual.”[125]

The leading case addressing whether use of drugs by a drug addict constitutes a voluntary act (United States v. Moore[126]) prompted six separate opinions. There, a defendant argued that his “long and intensive dependence on (addiction to) injected heroin” had resulted in “a loss of self-control over the use of heroin,” and that such a loss of will should constitute a defense to possession of heroin.[127] In multiple opinions concurring in the judgment, those judges voting in favor of upholding the conviction reasoned that, while the status of addiction may be involuntary, (1) the inception of addiction is not involuntary,[128] and (2) the choice to act on that addiction is difficult but not involuntary.[129] The court also expressed concern that, if addiction were a defense to possession, so too would it be a defense to property crimes committed to support an addiction.[130]

The Moore court rejected the dissent’s argument that “it can no longer seriously be questioned that for at least some addicts the ‘overpowering’ psychological and physiological need to possess and inject narcotics cannot be overcome by mere exercise of ‘free will.’”[131] Contemporary courts, applying similar reasoning, continue to reject the involuntary-act argument,[132] and the question is now well settled.[133]

C. Addiction and Insanity

In federal courts and those of twenty-eight states, addicted defendants have no grounds for an insanity defense. Those jurisdictions follow the rule in M’Naghten’s Case, requiring that the defendant’s “defect of reason” or “disease of the mind” render him incapable of understanding that “he was doing what was wrong.”[134] Put simply, addiction only (arguably) goes to one’s ability to avoid doing what is wrong, as opposed to being aware of its wrongfulness.

Most of the remainder follow the Model Penal Code, which reduces the requirement to a showing that the defendant’s “mental disease or defect” caused him to “lack[] substantial capacity” to “conform his conduct to the requirements of law.”[135] At first blush, this construction would seem to offer addicted defendants an opportunity to assert insanity defenses.

But no court applying the MPC insanity rule appears to have allowed addiction to suffice. Consider United States v. Lyons,[136] in which the Fifth Circuit (applying the MPC test)[137] put the matter bluntly: “[T]here is an element of reasoned choice when an addict knowingly acquires and uses drugs; he could instead have participated in an addiction treatment program. A person is not to be excused for offending ‘simply because he wanted to very, very badly.’”[138] In so holding, the Lyons court noted that the “great weight of legal authority clearly supports the view that evidence of mere narcotics addiction” cannot give rise to an insanity defense.[139]

Indeed, courts have rejected the very premise of an insanity defense for addiction by holding that addiction is not a “mental disease or defect.”[140] And some jurisdictions have gone so far as to preclude by statute the availability of an insanity defense for addicted defendants.[141]

The doctrine of “settled insanity” is instructive with respect to the rejection by courts of the theory that addiction is a disease giving rise to an irresistible compulsion. As the Lyons court noted, over time, drug addiction and alcoholism can “caus[e] physiological damage to [the defendant’s] brain,” which in turn may render the defendant incapable of conforming her conduct to the requirements of the law.[142] In some jurisdictions, “settled insanity” from substance abuse is a defense where “it can be demonstrated that substance use has triggered or exacerbated psychotic symptoms that become distinct and independent of acute intoxication.”[143] In California, for instance, “it is immaterial that voluntary intoxication may have caused the insanity,” provided that the insanity is “settled” and renders the defendant incapable of distinguishing right from wrong.[144] But settled insanity may not be “solely” a product of the typical effects of drug use, addiction included.[145]

Part IV

As demonstrated in the previous Part, criminal law has firmly determined that, however difficult it may be for addicts to avoid using narcotics, they can choose not to. The Supreme Court has rejected the notion that an addiction is an “irresistible compulsion.”[146] And in rejecting involuntary-act and insanity defenses, lower courts have determined firmly that an addict’s use of narcotics is a product of the exercise of her will,[147] an exercise in which she is capable of conforming her decisions to the requirements of the law.[148] The disease theory espoused in Robinson has been curtailed to limit only the criminalization of the status of addiction, not actions in conformity therewith.[149] Today, “the criminalization of low-level addictive behavior is routine.”[150]

But as this Part will argue, the admissibility of addiction evidence on an economic-compulsive theory of motive depends on the assumption that the addict does not exercise moral agency when acquiring drugs. Otherwise, it should be barred as character evidence. As introduced in Part I, evidence fails the character rule if its probative value depends on a propensity for immorality.[151] Courts cannot admit motive evidence if the existence of a motive turns on the defendant’s moral fiber, or lack thereof.[152] But the inference from evidence of addiction to the probability that an addict will choose to acquire drugs is only logical if the fact-finder assumes that the addict will once again succumb to immorality.

A. Why Should Criminal Law Matter to Evidence Law?

This Note’s argument depends on the premise that, where criminal law speaks to the morality of a given action, its determination should have force in the law of evidence. Admittedly, lawmakers are well within their rights to establish different rules for criminal law than they do for other spheres of law. Several constitutional protections, for instance, apply only in the realm of criminal procedure.[153] So too is criminal law’s burden of proof not required in civil cases.[154]

But with respect to definitional legal judgments, relating to whether conduct is culpable, other areas of law adhere to criminal law. More specifically, where criminal law deems conduct wrongful, that determination sets the floor for minimally acceptable behavior, and the rest of the law adopts that floor. The evidence rules themselves adopt, for purposes of impeaching a witness with prior convictions, criminal law’s definition of a felony.[155] And evidence law looks to the defined elements of criminal charges to determine whether a crime falls within the category of crimen falsi.[156] Convictions may be admissible for impeachment purposes even where criminal law has not deemed them “dishonest,” but the determination by criminal law that a conviction was for a dishonest act concludes the inquiry.[157]

The influence of criminal law exists outside of the rules of evidence as well. In tort law, an actor who violates a criminal statute without an excuse is negligent per se.[158] Violations of criminal law conclusively prove negligence “because the criminal statute reflects a ‘legislative judgment that acts in violation of the statute constitute unreasonable conduct.’”[159] And contract law considers a threat “improper,” for purposes of establishing that a contract was executed under duress, when, inter alia, the threatened action constitutes a crime.[160] Commercial law, too, accepts the premise that violation of a criminal law in pursuit of a trade secret constitutes one of several “improper means” of appropriation.[161]

Moreover, regardless of whether the validity of criminal punishment ought to require the necessary precondition of morally condemnable conduct, it does impose moral condemnation. Although legal positivists deny that “moral culpability” is a necessary condition of criminal liability,[162] that goes to whether criminal law may only punish immoral actors, not whether criminal liability in fact imposes moral condemnation. Contemporary scholars of the expressive nature of the law argue that criminal liability serves the purpose of communicating society’s condemnation of immoral conduct.[163] And opposition to the expressive theory takes issue with condemnation as a precondition for punishment, not with the foundational premise that criminal punishment in fact condemns.[164]

The expressivists have it right, at least with respect to the premise that criminal liability communicates to society that criminalized conduct is immoral. We tend to avoid illegal conduct not only for fear of legal punishment but also in light of moral commitments and the threat of social disapproval.[165] And we feel justified in holding persons criminally liable only when we can hold them morally culpable.[166] But the law takes upon itself the obligation to draw bright lines between the moral and the immoral, lines that do not exist (at least not so brightly) in reality.[167] As such, writing in the specific context of the criminalization of addiction, Professor Boldt concludes that “we have seen that its effect, in the ordinary case, is to reinforce notions of individual autonomy and free choice, while simultaneously obscuring the causal roots of criminal behavior.”[168]

And when courts determine that drug use by an addicted person is punishable, they expressly adopt the judgment that moral failing causes action in accordance with addiction. The Supreme Court in Powell relied in part on the Texas statute’s recognizing that intoxication “offends the moral and esthetic sensibilities of a large segment of the community.”[169] It rejected Justice Fortas’s position that “alcoholism is caused and maintained by something other than the moral fault of the alcoholic.”[170] Montana v. Egelhoff relied on “society’s moral perception” that one who becomes intoxicated “should be responsible for the consequences.”[171] Judge Wilkey, rejecting the involuntary-act defense in United States v. Moore, summarized the addict’s decision to obtain drugs as one in which “the addict’s moral standards are overcome by his physical craving for the drug . . . .”[172] He concluded that “[d]rug addiction of varying degrees may or may not result in loss of self-control, depending on the strength of character opposed to the drug craving.”[173] The Moore court rejected Judge Wright’s argument that “[n]o outer moral compulsion” can free an addict from “the spiraling web of [her] addiction.”[174] And the Lyons court rejected an insanity defense based in part on its characterization of an addict as someone who merely “wanted to [use] very, very badly.”[175]

B. Applying the Criminalization of Addiction to Evidence of Addiction

To date, the relevant scholarship has not taken issue with addiction evidence offered on a motive theory. Instead, scholars have argued that the question should turn on considerations outside the law, including scientific consensus[176] and localized codes of morality.[177]

But, as this Note has argued, there is no cause to look outside the law—it has already answered the question. And its answer is, emphatically, that addicts who act in accordance with their addictions by acquiring drugs have acted immorally. This subpart applies criminal law’s judgment in the context of addiction evidence offered on a compulsive theory of motive.

1. Specific Motive.—Addiction evidence, when offered to prove that a defendant unlawfully acquired narcotics, depends for its relevance on a finding that the defendant had more reason than the average person to acquire the drug in question. But that finding depends on an intermediate inference that the defendant would likely decide[178] to act on her addiction by acquiring narcotics. Indeed, the law of addiction instructs that this intermediate inference be treated as distinct from a showing that the defendant was addicted. In rejecting the disease model of addiction in favor of a moral-choice model, criminal law teaches that addictions do not mandate that addicts acquire drugs.

Courts admit addiction evidence on a specific-motive theory according to the following logical chain:

Evidence: Defendant was addicted to the stolen drug.

Inference: Defendant thus had a greater motive to steal the drug than an average, non-addicted person.

Conclusion: Defendant is more likely to be the perpetrator than is an average, non-addicted person.

It should be clear at this point that the chain is missing a necessary inference. The law of addiction teaches us that addicts will not necessarily acquire and use drugs. They will only do so in the event that they succumb to their addictions, an eventuality that turns on the individual addict’s strength of character. The chain of inferences in fact looks like this:

Evidence: Defendant was addicted to the stolen drug.

Inference 1: Defendant likely made the decision to act on that addiction by acquiring the drug.

Inference 2: Defendant thus had a greater motive to steal the drug than an average, non-addicted person.

Conclusion: Defendant is more likely to be the perpetrator than is an average, non-addicted person.

The chain’s logic should be prohibited. It requires a judgment that a person in a given state (addicted) will probably decide to act on that state. Because criminal law has determined that such a decision is an immoral one, the inference should be prohibited as a character inference.

Consider United States v. Cunningham.[179] There, the Seventh Circuit admitted evidence that the defendant was addicted to Demerol four years prior to her alleged theft of Demerol.[180] The necessary chain of inferences should look like this:

Evidence: Cunningham was addicted to Demerol.

Inference 1: Cunningham likely decided to act on her Demerol addiction by acquiring Demerol.

Inference 2: Cunningham thus had a greater motive to steal Demerol than an average, non-addicted person.

Conclusion: Cunningham, compared to someone who is not addicted to Demerol, is more likely to have stolen the Demerol.

Note that Cunningham would have no affirmative defense (at least none rooted in her addiction) to the charge of unlawful possession of Demerol. The law has determined conclusively that her decision to acquire Demerol and her action taken to acquire Demerol were immoral acts. But the “motive” evidence against her depends on the determination that, because of her addiction, she was prone to acquiring the drug. Because the motive theory depends on an inference of a propensity for immorality, the character rules should prohibit it.

General Motive: In the general-motive context, too, the relevance of addiction evidence requires an initial inference that the defendant would act in accordance with her addiction. Put simply, addiction itself is free of charge. Without an initial showing of a decision to act on her addiction, an addicted person has no more need of money than a non-addicted person. Consider the chain of inferences:

Evidence: Defendant was addicted.

Inference 1: Defendant likely made the decision to act on her addiction by purchasing narcotics.

Inference 2: Defendant thus had a greater motive to steal than an average, non-addicted person.

Conclusion: Defendant, compared to a non-addicted person, is somewhat more likely to be the perpetrator of a property crime.

As with specific-motive addiction evidence, the initial inference violates the rule against character evidence. Note that it should make no difference whether a court requires additional evidence of the extent of the defendant’s addiction or the defendant’s legitimate financial means.[181] In any chain that requires the inference that an addicted person is more likely to obtain narcotics than an average person, at least one link in the chain depends on a character inference.

Even testimony that a defendant was addicted contemporaneously with the crime should be excluded. That contemporaneous addiction is more probative of motive than past addiction makes no difference. Contemporaneous addiction’s relevance, too, requires the intermediate inference that an addict is likely to once again use drugs, and thus that the accused was likely to take action to get the funds necessary to purchase them. Take, for example, evidence that “in October 1989 [the defendant] had a $20 to $30 a day heroin habit.”[182] Although the bank robberies the defendant is accused of committing occurred in the same time frame,”[183] the relevance of his contemporaneous addiction depends on the inference that, on the days of the bank robberies, he woke up addicted and acted on his addiction. And that inference depends on his propensity to do just that.[184] This is not to say that the prosecution should be prohibited from admitting a defendant’s statement, made at or near the time of the crime, that he needed money to purchase drugs.[185] But where a defendant admits only that she had needed money to purchase drugs at some other time, the logical relevance of that admission requires an improper, intermediate character inference.[186]

2. A Propensity to “Act” Immorally.—This Part has thus far portrayed the character-evidence problem as arising when the jury infers from addiction evidence that the defendant decided to act in accordance with her addiction. Admittedly, the rule against character evidence appears to limit its prohibition to inferences that a person “acted” in accordance with her character.[187] And in the eyes of the law, a decision is not an act.[188]

But if we restrict the character rule to prohibiting only a showing of a propensity to act in the technical sense, then we elevate form at the expense of substance.[189] Motive has no independent relevance to property crimes. Prosecutors of property crimes use addiction evidence to prove that the defendant did the crime, not to prove a mental state.[190] Addiction evidence is relevant to property crimes strictly because it proves (to some degree) that the motivated defendant had a propensity to take action in accordance with her addiction.

Recall too that the character rule protects against inferential-error prejudice and nullification prejudice.[191] Juries are no more likely to give undue weight to a direct propensity to do immoral acts than to a propensity to make immoral decisions that subsequently cause immoral action. Indeed, the risk of inferential-error prejudice arises from the unpredictability of mental processes and decision making.[192] Neither is nullification prejudice any less likely with respect to the defendant’s propensity to make immoral decisions. Juries are prone to punishing for a “criminal mind” at least to the extent that they are prone to punishing for specific criminal acts.[193]

And compared to adopting criminal law’s definition of immorality, there is less reason for evidence law to adopt criminal law’s definition of action. Evidence law in general, and the character rule in particular, concern themselves with the influence of evidence on juries.[194] Because criminalization has the power to (and in fact does) communicate to juries what is immoral and what is not, criminal law’s definitions should have force in defining immorality.[195] In contrast, there is no reason to suspect that criminal law’s definition of action influences jurors’ reactions to or interpretations of evidence.

Moreover, even if the character rule ought to import criminal law’s definition of action, the difficulty of “decision” versus “act” is less a product of addiction evidence itself than of the manner in which this Note has portrayed its logic. Regardless of how we frame each link in the chain of inferences, it should be clear at this point that the chain’s logic depends on an inference of immoral propensity. That is, the theory only works if the fact finder understands addiction as a propensity to use drugs, which requires that they be acquired. And a propensity to acquire is a propensity to do an act. Consider an alternative characterization of the chain of inferences:

Evidence: Defendant was addicted, i.e., she was prone to using drugs.

Inference 1: Because using drugs requires that they be acquired, the defendant had a motive to acquire the funds necessary to purchase narcotics.

Inference 2: It is somewhat more likely that the defendant, compared to a non-addicted person, did what was necessary to acquire funds to purchase narcotics.

Conclusion: Because property crimes are one method of acquiring funds, it is somewhat more likely that defendant committed the property crime.

Understood this way, the initial motive inference allows the second inference of immoral action.[196] Indeed, on the theory that an addict committed a property crime to acquire funds to purchase drugs, the crime itself is an “act in accordance” with the character of addiction. And the criminal law teaches us that acquiring drugs is immoral, whether the acquirer is addicted or not.


This Note has argued that prohibited character inferences are those that constitute inferences of a propensity for immorality; that criminal law has determined that addicted persons exercise immorality when they take or acquire narcotics; that criminal law’s determination should have force in the law of evidence; and thus that addiction evidence should be excluded in prosecutions for crimes against property. As yet, courts have failed to recognize the essentially character-driven nature of addiction evidence.

The extent (if any) to which morality influences the decision to act on one’s addiction remains a matter of genuine debate. Scholars continue to argue that contemporary understandings of addiction mandate that, at least in some cases, addiction should constitute a complete or partial excuse.[197] Others have advocated not for an excuse defense but instead for removing addicted persons from the criminal system into the medical system.[198] But as Professor Morse summarizes, “Current Anglo-American law concerning addiction is most consistent with the choice model of addictive behavior,” and “the no-choice model has made few inroads despite the enormous advances in the psychological, genetic and neuroscientific understanding of addiction.”[199]

So, this Note takes the law as it stands, arguing simply that it cannot have it both ways.

Michael Davis[200]

  1. .Economic compulsion was one of Goldstein’s three explanations for the nexus, in addition to “psychopharmacological” and “systemic” models. Paul J. Goldstein, The Drugs/Violence Nexus: A Tripartite Conceptual Framework, 15 J. Drug Issues 493, 494 (1985).
  2. .Id. at 496.
  3. .See Richard Hammersley et al., The Relationship Between Crime and Opioid Use, 84 Brit. J. Addiction 1029, 1040 (1989) (summarizing research as demonstrating that “[d]ay-to-day, crime was a better explanation of drug use than drug use was of crime”); Scott Menard et al., Drugs and Crime Revisited, 18 Just. Q. 269, 269 (2001) (finding, consistent with past research, that “the ‘drug use causes crime’ hypothesis is untenable because crime typically is initiated before substance use”); Benjamin R. Nordstrom & Charles A. Dackis, Drugs and Crime, 39 J. Psychol. & L. 663, 683 (2011) (concluding that “drug use and criminal activity feed off of each other”); Toby Seddon, Explaining the Drug–Crime Link: Theoretical, Policy and Research Issues, 29 J. Soc. Pol’y 95, 103 (2000) (asserting that it is time to “rethink the whole concept of ‘drug-related crime’” because after an analysis of the literature, “acquisitive crime causally related (in a deterministic way) to drug use . . . does not fit in with research findings”); Mark Simpson, The Relationship Between Drug Use and Crime: A Puzzle Inside an Enigma, 14 Int’l J. Drug Pol’y 307, 318 (2003) (“Research must now move beyond theories based upon singular causality and instead examine further the multiple and complex ways that drug use and crime interact with each other in people’s lives.”).
  4. .Trevor Bennett et al., The Statistical Association Between Drug Misuse and Crime: A Meta-Analysis, 13 Aggression & Violent Behav. 107, 117 (2008).
  5. .Drugs and Crime Facts: Drug Use and Crime, Bureau Just. Stat., [].
  6. .See Fed. R. Evid. 401 (defining evidence as logically relevant if it makes the existence of a material fact “more or less probable than it would be without the evidence”).
  7. .Fed. R. Evid. 404(a). Although this Note will make frequent reference to the Federal Rules, nearly all state rules of evidence mirror their federal counterparts in prohibiting character evidence and admitting evidence of specific acts for “other purposes,” including motive. For an early summary of the quick and widespread adoption of the Federal Rules by the states, see L. Kinvin Wroth, The Federal Rules of Evidence in the States: A Ten-Year Perspective, 30 Vill. L. Rev. 1315, 1329–30 (1985); see also, e.g., Cal. Evid. Code § 1101(a) (2017); Tex. R. Evid. 404; People v. Denson, 42 N.E.3d 676, 681 (N.Y. 2015).
  8. .Fed. R. Evid. 404(b).
  9. .See infra notes 20–24 and accompanying text.
  10. .See, e.g., Am. Psychiatric Ass’n, Diagnostic and Statistical Manual of Mental Disorders 541 (5th ed. 2013) (“Opioid Use Disorder”); World Health Org., The ICD-10 Classification of Mental and Behavioural Disorders 57 (1993), http://apps.who
    .int/iris/bitstream/10665/37108/1/9241544554.pdf [] (“Dependence Syndrome”).
  11. .See Richard C. Boldt, The Construction of Responsibility in the Criminal Law, 140 U. Pa. L. Rev. 2245, 2298 (1992) (“[R]esearch has revealed that the disease is produced by a shifting confluence of genetic/biochemical, environmental, and sociocultural factors.”); David P. Leonard, Character and Motive in Evidence Law, 34 Loy. L.A. L. Rev. 439, 531 (2001) (referring to the “clear trend” toward treating addiction as a “disease of the brain, leading to uncontrollable and compulsive behavior”); Joan Ellen Zweben & J. Thomas Payte, Methadone Maintenance in the Treatment of Opioid Dependence—A Current Perspective, 152 W. J. Med. 588, 589 (1990) (observing that, in the medical community, it is “commonly agreed” that addiction is a “complex phenomena” founded in “the interaction of biologic, psychosocial, and cultural variables”); see also National Alcohol and Drug Addiction Recovery Month, 2016, Proclamation No. 9479, 81 Fed. Reg. 61973 (Aug. 31, 2016), [] (“Substance use disorder, commonly known as addiction, is a disease of the brain, and many misconceptions surrounding it have contributed to harmful stigmas that can prevent individuals from seeking the treatment they need.”). See generally John C. Crabbe, Genetic Contributions to Addiction, 53 Ann. Rev. Psychol. 435, 451–52 (2002); Alan I. Leshner, Addiction Is a Brain Disease, and It Matters, 278 Sci. 45 (1997). But see Gene M. Heyman, Addiction: A Disorder of Choice (2009).
  12. .Powell v. Texas, 392 U.S. 514, 561 (1968) (Fortas, J., dissenting); see infra subpart III(A).
  13. .See infra subpart III(B).
  14. .See infra subpart III(C).
  15. .See infra subpart II(A). See generally Debra T. Landis, Annotation, Admissibility of Evidence of Accused’s Drug Addiction or Use to Show Motive for Theft of Property Other Than Drugs, 2 A.L.R. 4th 1298 (originally published in 1980) (collecting cases).
  16. .See infra notes 79–83.
  17. .Throughout this Note, terms like “likely” and “probably” are used as short-hand for a judgment that evidence surpasses the logical-relevance threshold. Logical relevance requires only that evidence have “any tendency” to make a material fact somewhat “more probable.” Fed. R. Evid. 401. So when this Note uses terms like “likely,” it means “somewhat more likely,” not “more likely than not.”
  18. .Fed. R. Evid. 404(a)(1), (b)(1).
  19. .See State v. Williams, 874 P.2d 12, 25 (N.M. 1994) (Montgomery, C.J., concurring) (“I am unable to do what all the text-writers and other legal authorities have failed to do. I am unable to outline the contours of the term ‘character’ . . . .”); Leonard, supra note 11, at 450 (“Unfortunately, there is no general agreement about the precise meaning of the term [character].”); Richard B. Kuhns, The Propensity to Misunderstand the Character of Specific Acts Evidence, 66 Iowa L. Rev. 777, 799 (1981) (referring to the “nature of the term ‘character’” as “elusive”).
  20. .See Fed. R. Evid. 405 advisory committee’s note on proposed rules (“Traditionally, character has been regarded primarily in moral overtones of good and bad . . . .”); 22B Charles Alan Wright & Kenneth W. Graham, Jr., Federal Practice and Procedure § 5233 (2017) (defining “character” evidence as involving both a claim that a person “has a pattern of repetitive behavior” and a claim that “the behavior is morally praiseworthy or condemnable”); Leonard, supra note 11, at 451 (“[I]t is best to conceive of character as a subset of propensity, embracing only moral aspects of a person.”); Barrett J. Anderson, Note, Recognizing Character: A New Perspective on Character Evidence, 121 Yale L.J. 1912, 1921 (2012) (describing morality as “the key to understanding the difference between character evidence and non-character propensity proof”).
  21. .Kuhns, supra note 19, at 794.
  22. .Fed. R. Evid. 406; see also Fed. R. Evid. 406 advisory committee’s note on proposed rules (distinguishing character, which speaks only to a “tendency,” from habit, which “is the person’s regular practice of meeting a particular kind of situation with a specific type of conduct”) (quoting Charles T. McCormick, McCormick’s Handbook of the Law of Evidence § 162, at 340 (1st ed. 1954)).
  23. .Paul F. Rothstein, Intellectual Coherence in an Evidence Code, 28 Loy. L.A. L. Rev. 1259, 1265 (1995). Note that a propensity for acquiring and using drugs does not rise to the level of specificity and regularity to constitute a habit for purposes of evidence law. See Fed. R. Evid. 406 advisory committee’s note on proposed rules (contrasting character traits like “temperance” with regular and specific habits like “going down a particular stairway two stairs at a time”) (quoting McCormick, supra note 22, at 340).
  24. .See United States v. West, 670 F.2d 675, 682 (7th Cir. 1982) (affirming admission of evidence of lack of intelligence); see also United States v. Cortez, 935 F.2d 135, 138 n.3 (8th Cir. 1991) (rejecting the argument that the character rule “encompasses slowness to answer, forgetfulness, or poor ability to express oneself”). The fault lines of the non-character–character dichotomy are most visible in the question of whether a personality trait, which makes conduct by that personality more probable, necessarily constitutes character evidence. See, e.g., State v. Ferguson, 803 P.2d 676, 685 (N.M. 1990) (concluding that evidence that a person is “suspicious” or “paranoid” may survive the character rule because such attributes do not “bear strong moral connotations”). This Note avoids that difficulty.
  25. .Michelson v. United States, 335 U.S. 469, 475–76 (1948) (acknowledging the “admitted probative value” of character evidence); David P. Leonard, In Defense of the Character Evidence Prohibition: Foundations of the Rule Against Trial by Character, 73 Ind. L.J. 1161, 1182 (1998) (observing that “courts rarely exclude character evidence on the ground that it is irrelevant in determining conduct” because “it has long been believed that evidence of character satisfies the lenient test of logical relevance”).
  26. .See Fed. R. Evid. 401 (defining relevant evidence as “evidence having any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence”).
  27. .See Michelson, 335 U.S. at 476 (including “confusion of issues” and “unfair surprise” as justifications for the disallowance of character evidence). Note that, with respect to evidence of a defendant’s specific acts, the risk of unfair surprise is today alleviated by a notice requirement. See Fed. R. Evid. 404(b)(2).
  28. .Leonard, supra note 11, at 450.
  29. .Leonard, supra note 25, at 1184; see also Michelson, 335 U.S. at 475–76 (expressing concern that character evidence may “weigh too much with the jury and . . . so overpersuade them as to prejudge one with a bad general record”); United States v. Carrillo, 981 F.2d 772, 774 (5th Cir. 1993) (“Character evidence is not excluded because it has no probative value, but because it sometimes may lead a jury to convict the accused on the ground of bad character deserving punishment regardless of guilt.”); Edward J. Imwinkelried, Uncharged Misconduct Evidence § 1.3 (rev. ed. 1999) (describing these two “probative dangers” in the specific context of past-wrongs evidence).
  30. .Kuhns, supra note 19, at 779 (limiting specific acts that evince “character” to those that “have some moral overtone” because “ascribing such a meaning to it for the purposes of the specific acts prohibition is consistent with the concern over the potentially prejudicial impact of specific acts evidence”); Anderson, supra note 20, at 1944 (“Prejudice would not exist unless jurors could use the proffered evidence in an inappropriate manner, and courts have rightly noted that morally neutral traits do not engender the types of gut-level reactions from jurors that would cause prejudice.”); see also 1 Kenneth S. Broun et al., McCormick on Evidence § 195, at 1080 n.3 (7th ed. 2013) (suggesting that consideration of whether traits are prejudicial is integral to determining whether they “qualify as traits of character”).
  31. .See Miguel A. Méndez, Character Evidence Reconsidered: “People Do Not Seem to Be Predictable Characters”, 49 Hastings L.J. 871, 881 n.38 (1998) (collecting studies).
  32. .See Kuhns, supra note 19, at 796 (“The degree of prejudice associated with any specific act evidence is a function of how the factfinder is likely to respond to the badness of the act.”).
  33. .Fed. R. Evid. 404(b)(1).
  34. .Fed. R. Evid. 404(b)(2). Much ink has been spilled in arguing whether so-called “other-purposes” evidence opens the floodgates to character evidence in disguise. See, e.g., Andrew J. Morris, Federal Rule of Evidence 404(b): The Fictitious Ban on Character Reasoning from Other Crime Evidence, 17 Rev. Litig. 181 (1998); Edward J. Imwinkelried, The Use of Evidence of an Accused’s Uncharged Misconduct to Prove Mens Rea: The Doctrines Which Threaten to Engulf the Character Evidence Prohibition, 51 Oh. St. L.J. 575 (1990); Glen Weissenberger, Making Sense of Extrinsic Act Evidence: Federal Rule of Evidence 404(b), 70 Iowa L. Rev. 579, 579 (1985) (observing the “frequently asserted and winkingly cynical statement that an inventive prosecutor will almost always succeed in devising a theory that will support the admissibility of the accused’s extrinsic antisocial act”). This Note will not join that debate.
  35. .Fed. R. Evid. 404 advisory committee’s note to 1991 amendment (“Rule 404(b) has emerged as one of the most cited Rules in the Rules of Evidence.”); 22B Charles Alan Wright & Kenneth W. Graham, Jr., Federal Practice and Procedure § 5239, at 97–99 (2014) (“While the general rule of exclusion is often applauded—and occasionally enforced—it is the exceptions that are of most practical significance.”); Imwinkelried, supra note 34, at 576 (referring to the admissibility of specific-bad-acts evidence as “the single most important issue in contemporary criminal evidence law”).
  36. .Leonard, supra note 11, at 441 & n.8.
  37. .United States v. Beechum, 582 F.2d 898, 912 n.15 (5th Cir. 1978) (quoting M.C. Slough & J. William Knightley, Other Vices, Other Crimes, 41 Iowa L. Rev. 325, 328 (1956)).
  38. .See United States v. Day, 591 F.2d 861, 874 (D.C. Cir. 1978) (“Motive is a state of mind that is shown by proving the emotion that brings it into being.”); Wright & Graham, supra note 35, § 5240, at 144 (defining motive in the “generally accepted sense” to include either “an emotion or state of mind” that “incentiv[izes] . . . certain volitional activity”).
  39. .Merriam-Webster’s Collegiate Dictionary 810 (11th ed. 2006).
  40. .State v. Pullens, 800 N.W.2d 202, 242 (Neb. 2011) (“Motive is normally used as an intermediate inference to prove identity.”); Thomas J. Reed, Admitting the Accused’s Criminal History: The Trouble with Rule 404(b), 78 Temp. L. Rev. 201, 217 (“Motive is rarely a stated element of a criminal offense, but the courts have traditionally found that proof of the accused’s motive for committing a crime is relevant to the issue of guilt.”); see also, e.g., United States v. Benton, 637 F.2d 1052, 1056–57 (5th Cir. 1981) (admitting motive evidence as “evidence of identity” where defendant was aware that the victim might implicate him in other homicides). Motive evidence may also be relevant toward proving state of mind, Reed, supra, at 217; Imwinkelried, supra note 34, at 595, 597, or to support the prosecution’s theory of the case, United States v. Pedroza, 750 F.2d 187, 200–01 (2d Cir. 1984), or to prove that the crime occurred at all, Leonard, supra note 11, at 489–90. But because addiction-as-motive evidence is relevant only to identity, this Note does not address the other uses of motive evidence.
  41. .Although the relevance of a motive “always involves a type of propensity inference, . . . the validity of that inference does not depend on an assumption about the person’s character.” Leonard, supra note 11, at 488–89.
  42. .Imwinkelried, supra note 29, § 3:15; Leonard, supra note 11, at 469.
  43. .See supra notes 20–24 and accompanying text.
  44. .Leonard, supra note 11, at 452, 458.
  45. .Fed. R. Evid. 403.
  46. .Fed. R. Evid. 404 advisory committee’s note to proposed rules.
  47. .See, e.g., United States v. Madden, 38 F.3d 747, 751 (4th Cir. 1994) (acknowledging a “difficult balancing question between admissibility under Rule 404(b) and prejudice under Rule 403” where a prosecutor attempts to prove motive via financial need arising from “some other” illegal act).
  48. .See Wright & Graham, supra note 35, § 5239, at 125 (requiring that the “inference to conduct” from Rule 404(b) evidence “be made without the need to infer the person’s character as a step in the reasoning”); Leonard, supra note 11, at 442 (stating that if any inference in the “chain of inferences” attempts to show that the defendant acted in conformity with moral propensity, it is inadmissible as character evidence); see also Imwinkelried, supra note 34, at 581–84 (analyzing each “inferential step” in a sequence of logical reasoning and rejecting a logical chain depending on “an intermediate assumption about the accused’s character”).
  49. .United States v. Givan, 320 F.3d 452, 471–72 (3d Cir. 2003).
  50. .Id.; see also, e.g., United States v. Gomez, 763 F.3d 845, 856 (7th Cir. 2014) (“[T]he rule allows the use of other-act evidence only when its admission is supported by some propensity-free chain of reasoning.”); United States v. Smith, 725 F.3d 340, 345 (3d Cir. 2013) (“[T]he proponents of Rule 404(b) evidence must do more than conjure up a proper purpose—they must also establish a chain of inferences no link of which is based on a propensity inference.”); People v. Thompson, 611 P.2d 883, 889 (Cal. 1980) (refusing to allow a “criminal disposition” to “establish any link in the chain of logic connecting the uncharged offense with a material fact”).
  51. .Gomez, 763 F.3d at 856.
  52. .For a similar theory, see United States v. Claxton, 276 F.3d 420 (8th Cir. 2002), in which evidence of drugs and drug trafficking found in the defendant’s apartment was admissible “for the purpose of showing [defendant’s] motive for possessing the guns and was relevant to the issue of the ownership of the guns.” Id. at 423. The Claxton case differs from our hypothetical, though, as no character inference was necessary to establish that the Claxton defendant was involved in drug trafficking at the time of his alleged possession of the firearm—police officers found both the drug evidence and the firearms in the same search. Id. at 422.
  53. .This may not always be so. For instance, where addiction evidence is admitted to show a motive to have committed a particularly heinous crime, or one which carries the potential for a particularly severe sentence, see, e.g., State v. Hughes, 191 P.3d 268, 273, 278 (Kan. 2008) (felony murder), the risk that nullification prejudice will result from addiction evidence may be lower.
  54. .See United States v. Rubio-Estrada, 857 F.2d 845, 853 (1st Cir. 1988) (Torruella, J., dissenting) (lamenting that courts may “circumvent[] the ban” whenever enumerated other purposes are at issue “without making explicit the specific logical progression” necessary to prove the relevant fact); see also Morris, supra note 34, at 208 (arguing that many contemporary uses of uncharged-misconduct evidence require a “chain” that “necessarily includes an inference” of character and that “[t]he cases simply cannot be squared with the plain language of Rule 404(b)”).
  55. .Fed. R. Evid. 404(a)(1). As Professor Imwinkelried has argued, so long as American courts remain purportedly committed to the prohibition of character evidence despite calls for its relaxation, if the rule is to be walked back, “it should be done explicitly in a straightforward fashion—not by legerdemain.” Imwinkelried, supra note 34, at 602–03.
  56. .Imwinkelried, supra note 29, § 3.16 (observing “numerous” such cases); see also Landis, supra note 15 (collecting cases).
  57. .The accuracy of this statement is, for the most part, beyond the scope of this Note. But it is worth noting that according to the National Institute on Drug Abuse, as of 2013, 9.4% of the population of the United States was estimated to have used an illicit drug in the past month. Ctr. for Behav. Health Stat. & Qual., U.S. Dep’t Health & Human Servs., Results from the 2013 National Survey on Drug Use and Health: Summary of National Findings (Sept. 2014), [].
  58. .See infra notes 60–61 and accompanying text.
  59. .Leonard, supra note 11, at 524. Note that general-motive addiction evidence is distinct from drug use admitted as intrinsic to, or “inextricably intertwined” with, the charged offense. But the inextricably-intertwined theory is sometimes offered as additional support for the admission of addiction evidence on a general-motive theory. See, e.g., United States v. Cody, 498 F.3d 582, 590–91 (6th Cir. 2007) (affirming the admission of addiction evidence in prosecution for bank robbery because defendant’s drug habits were “inextricably intertwined with the bank robbery, . . . and thus extremely probative of motive”); United States v. Lafferty, 372 F. Supp. 2d 446, 463 (W.D. Pa. 2005) (finding, in prosecution for burglary of firearms, “addiction to/use of heroin” admissible both as “part of a ‘single criminal episode’” and as evidence of motive), rev’d on other grounds, 503 F.3d 293 (3d Cir. 2007). For a comprehensive discussion of the inextricably-intertwined theory, see generally Edward J. Imwinkelried, The Second Coming of Res Gestae: A Procedural Approach to Untangling the “Inextricably Intertwined” Theory for Admitting Evidence of an Accused’s Uncharged Misconduct, 59 Cath. U. L. Rev. 719 (2010).
  60. .United States v. Cunningham, 103 F.3d 553, 557 (7th Cir. 1996).
  61. .See also, e.g., United States v. Petrillo, No. 15-cr-192-01-JL, 2016 WL 4444726, at *2 (D.N.H. Aug. 23, 2016) (holding that, in the context of a trial for making false statements to a federal agency, evidence of addiction may be admissible to show “a motive to lie on an application for a job that would provide access to pharmaceutical drugs,” but withholding the evidence upon finding risk of unfair prejudice); State v. Collins, 528 P.2d 829, 831 (Ariz. 1974) (holding evidence of defendant’s heroin addiction admissible to prove a motive for stealing heroin in a robbery-murder).
  62. .See, e.g., United States v. Washam, 468 F. App’x 568, 572 (6th Cir. 2012) (bank robbery); United States v. Cody, 498 F.3d 582, 591 (6th Cir. 2007) (bank robbery); United States v. Laflam, 369 F.3d 153, 157 (2d Cir. 2004) (bank robbery); United States v. Bitterman, 320 F.3d 723, 727 (7th Cir. 2003) (bank robbery); United States v. Brooks, 125 F.3d 484, 500–01 (7th Cir. 1997) (bank robbery); United States v. Miranda, 986 F.2d 1283, 1285 (9th Cir. 1993) (bank robbery); United States v. Kadouh, 768 F.2d 20, 21–22 (1st Cir. 1985) (heroin trafficking); Donovan v. State, 32 S.W.3d 1, 9 (Ark. Ct. App. 2000) (theft); State v. Hughes, 191 P.3d 268, 278–79 (Kan. 2008) (felony-murder and burglary); People v. Jones, No. 324512, 2016 WL 4129097, at *4 (Mich. Ct. App. Aug. 2, 2016) (felony-murder); People v. King, No. 324500, 2016 WL 555860, at *3 (Mich. Ct. App. Feb. 11, 2016) (home invasion).
  63. .Washam, 468 F. App’x at 572 (quoting Cody, 498 F.3d at 591).
  64. .United States v. Cyphers, 553 F.2d 1064, 1069 (7th Cir. 1977); King, 2016 WL 555860, at *3; Donovan, 32 S.W.3d at 9.
  65. .Bitterman, 320 F.3d at 727.
  66. .Kadouh, 768 F.2d at 21.
  67. .State v. Henness, 679 N.E.2d 686, 694 (Ohio 1997).
  68. .That is, curable by a limiting instruction. See, e.g., LaFlam, 369 F.3d at 157; Bitterman, 320 F.3d at 727.
  69. .38 F.3d 747 (4th Cir. 1994).
  70. .Id. at 751.
  71. .Id. at 752.
  72. .See, e.g., Leger v. Commonwealth, 400 S.W.3d 745, 751 (Ky. 2013) (suggesting that the admissibility of a drug habit, on a motive theory, requires “evidence of insufficient funds to support that habit”); People v. Jones, 326 N.W.2d 411, 413 (Mich. Ct. App. 1982) (“[T]he legal relevance of heroin addiction to motive for a theft offense is dependent on two factors: (1) that defendant was addicted at or near the time of the offense . . . , and (2) that defendant lacks sufficient income from legal sources to sustain his or her continuing need for heroin.”).
  73. .See, e.g., United States v. Mullings, 364 F.2d 173, 175–76 (2d Cir. 1966) (rejecting the argument that defendant’s use of narcotics despite making less than $65 per week was sufficient to admit evidence of narcotics use because “[t]here was no evidence how often [defendant] took narcotics, or what the maintenance of such a habit would cost him”); State v. Powell, 459 S.E.2d 219, 226 (N.C. 1995) (affirming admission, to show motive for felony-murder, of evidence that defendant had a $100-per-day drug habit and no longer received monthly government assistance); Biera v. State, 391 S.W.3d 204, 213 (Tex. App.—Amarillo 2012, pet. ref’d) (allowing the inference to motive for robbery from drug use combined with evidence of unemployment).
  74. .See supra notes 40–42 and accompanying text.
  75. .191 P.3d 268 (Kan. 2008).
  76. .Id. at 278.
  77. .Id.; see also United States v. LaFlam, 369 F.3d 153, 157 (2d Cir. 2004) (“[B]ecause the identity of the bank robber was in dispute, evidence [of drug use] establishing that he had a motive to commit those robberies was material and relevant.”).
  78. .United States v. Cunningham, 103 F.3d 553, 556 (7th Cir. 1996) (emphasis added). Judge Posner, writing in Cunningham, acknowledged that motive evidence “overlap[s]” with character evidence when the evidence of past bad acts proves a “taste” for engaging in crime. Id. But even where motive overlaps with character, his solution “for preventing . . . abuse is Rule 403, not Rule 404(b).” Id. at 556–57. And, Judge Posner specifically posits that evidence of past drug convictions used to show motive in a robbery case raises no character-evidence problems. Id. at 557.
  79. .People v. King, No. 324500, 2016 WL 555860, at *3 (Mich. Ct. App. Feb. 11, 2016).
  80. .United States v. Brooks, 125 F.3d 484, 500 (7th Cir. 1997).
  81. .LaFlam, 369 F.3d at 156.
  82. .The difficulty of fairly estimating probative value and prejudicial effect is notorious. See H. Richard Ulliver, Evidence of Character to Prove Conduct: Illusion, Illogic, and Injustice in the Courtroom, 130 U. Pa. L. Rev. 845, 846 (1982) (referring to the characterization of judging what is “probative” and what is “prejudicial” as “an ornamented baroque partita on a two note theme”). If nothing else is clear from a survey of courts’ treatment of addiction evidence, it serves as further indication of such difficulty.
  83. .People v. Cardenas, 31 Cal. 3d 897, 907 (1982) (quoting People v. Davis, 233 Cal. App. 2d 156, 161 (Cal. Ct. App. 1965)). The Cardenas court went on to observe “that the public generally is influenced with the seriousness of the narcotics problem in this community, and has been taught to loathe those who have anything to do with illegal narcotics in any form or to any extent.” Id.
  84. .United States v. Madden, 38 F.3d 747, 753 (4th Cir. 1994). More recently, the Supreme Court of West Virginia rejected the admissibility of addiction evidence while collecting a litany of judicial observations to the effect that narcotics offenses are by their nature unduly prejudicial. State v. Taylor, 593 S.E.2d 645, 650 (W. Va. 2004).
  85. .United States v. Sutton, 41 F.3d 1257, 1259–60 & n.3 (8th Cir. 1994).
  86. .State v. LeFever, 690 P.2d 574, 578 (Wash. 1984) (en banc), rev’d on other grounds, State v. Brown, 782 P.2d 1013 (Wash. 1989) (en banc).
  87. .That said, this Note argues that courts should not reach the Rule 403 inquiry because addiction evidence should be facially inadmissible as character evidence.
  88. .Imwinkelried, supra note 29, § 3:15 (positing that motive evidence has substantial probative value only where “[m]any other persons presumably had no motive.”).
  89. .Id.
  90. .Cf. 1 Timothy 6:10 (King James) (“[T]he love of money is the root of all evil.”).
  91. .Edwin H. Sutherland et al., Principles of Criminology 90 (11th ed. 1992).
  92. .Id.
  93. .Id.
  94. .A 2009 study analyzing a random, stratified sample of the U.S. population found that people with drug addictions were “seen as more dangerous and fear evoking than those with” physical or other mental disorders. Patrick W. Corrigan et al., The Public Stigma of Mental Illness and Drug Addiction, 9 J. Soc. Work 139, 143, 145 (2009). And despite the steady march of science towards understanding addiction as a disease, the mere state of intoxication is prejudicial: a “person under the influence of alcohol or drugs is seen as unpredictable, and thus anxiety-provoking.” Robin Room, Stigma, Social Inequality, and Alcohol and Drug Use, 24 Drug & Alcohol Rev. 143, 150 (2005). Moreover, the continued moralization of addiction results in the lay-understanding that addiction is a “causal agen[t]” in “violence, calamities, and failure in major social roles.” Id. at 149.
  95. .766 A.2d 1176 (N.J. Super. Ct. App. Div. 2001).
  96. .Id. at 1180.
  97. .Id. at 1181; see also State v. J.M., 102 A.3d 1233, 1237 (N.J. Super. Ct. App. Div. 2014) (“[P]roof of a defendant’s drug addiction to show motive for committing a burglary or theft is inadmissible on the theory that drug addicts are perpetually in need of money.”).
  98. .977 A.2d 454 (N.H. 2009).
  99. .Id. at 460. The court noted the “strong circumstantial evidence identifying the defendant as the perpetrator,” and a witness who could identify the defendant, which indicated that the addiction evidence would be used to support the prosecution’s theory of the case, not to prove identity. Id.
  100. .Id. at 459 (“In the absence of some identification of the defendant as the intruder, his heroin addiction, though introduced to show motive, would necessarily fill in the missing logical gaps that Rule 404(b) requires a prosecutor to fill.”); see also Gould v. State, 579 P.2d 535, 539 (Alaska 1978) (holding that, because the “issue of identification” was “hotly contested” by defendant, evidence that defendant was “unemployed and had a $300 a day heroin habit” was not admissible to prove a motive for burglary).
  101. .Costello, 977 A.2d at 459.
  102. .As discussed above, the sole purpose served by addiction evidence is proof of identity. See supra notes 40–42, 60–61, 74–77 and accompanying text.
  103. .For instance, in United States v. Cepeda Penes, 577 F.2d 754 (1st Cir. 1978), baseball star Orlando Cepeda appealed his conviction of possession of marijuana with intent to distribute. Id. at 755. Cepeda claimed that “he was the innocent recipient of contraband” and unaware that the packages he received contained marijuana. Id. at 760. As such, evidence that Cepeda had not filed tax returns and owed back taxes was admissible to prove his motive, which in turn proved intent. Id.
  104. .Cf. Leonard, supra note 11, at 527 (“‘Once a murderer, always a murderer’ is precisely the type of reasoning forbidden by the character rule. But is drug addiction the same? Arguably not.”).
  105. .370 U.S. 660 (1962).
  106. .Id. at 660, 662, 667.
  107. .Id. at 667.
  108. .Id. at 666.
  109. .Id. at 667.
  110. .Id. at 666.
  111. .Id. at 667–76 (Douglas, J., concurring).
  112. .Id. at 669–70.
  113. .Id. at 671. Even Justice Douglas, though, conceded that addicts may be punished for “acts of transgression.” Id. at 674. But it was unclear whether he would consider mere use of narcotics, by a narcotic addict, to constitute such a transgression—in dissent, Justice White expressed concern that the majority’s holding would “place the use of narcotics beyond the reach of the States’ criminal laws.” Id. at 686 (White, J., dissenting).
  114. .392 U.S. 514 (1968).
  115. .Id. at 533.
  116. .Id. at 535.
  117. .Id. at 568 (Fortas, J., dissenting).
  118. .See William J. Stuntz, The Uneasy Relationship Between Criminal Procedure and Criminal Justice, 107 Yale L.J. 1, 69 (1997) (“Powell v. Texas abandoned the broader reading of Robinson, and today the criminalization of low-level addictive behavior is routine.”).
  119. .Stephen J. Morse, Addiction, Choice and Criminal Law, in Addiction and Choice 426, 435 (Nick Heather & Gabriel Segal eds., 2017).
  120. .Traynor v. Turnage, 485 U.S. 535, 536 (1988).
  121. .Id. at 550 (stating that the D.C. Circuit “accurately characterized” the medical community’s position on the disease model as “a substantial body of medical literature that even contests the proposition that alcoholism is a disease, much less that it is a disease for which the victim bears no responsibility” (quoting McKelvey v. Turnage, 792 F.2d 194, 200–01 (D.C. Cir. 1986))).
  122. .Id. (citing Herbert Fingarette, The Perils of Powell: In Search of a Factual Foundation for the “Disease Concept of Alcoholism”, 83 Harv. L. Rev. 793, 802–08 (1970)).
  123. .Montana v. Egelhoff, 518 U.S. 37 (1996).
  124. .Id. at 49.
  125. .Model Penal Code § 2.01(2)(d) (Am. Law Inst., 1962); see also Deborah W. Denno, Crime and Consciousness: Science and Involuntary Acts, 87 Minn. L. Rev. 269, 277 (2002) (collecting statutes and observing that “[m]ost states have an explicit requirement or a provision that approximates” the MPC’s).
  126. .486 F.2d 1139 (D.C. Cir. 1973).
  127. .Id. at 1144 (Wilkey, J., concurring in the judgment).
  128. .See id. at 1151 (“Moore could never put the needle in his arm the first and many succeeding times without an exercise of will. His illegal acquisition and possession are thus the direct product of a freely willed illegal act.”). But see id. at 1243 (Wright, J., dissenting) (“[N]o matter how the addict came to be addicted, once he has reached that stage he clearly is sick, and a bare desire for vengeance cannot justify his treatment as a criminal.”).
  129. .See id. at 1150 (Wilkey, J., concurring in the judgment) (“[W]hile addiction may be a ‘compelling propensity to use narcotics,’ it is not necessarily an irresistible urge to have them.”).
  130. .See id. at 1208 (Robb, J., concurring in the judgment) (“The doctrine espoused by the minority would license an addict to commit any criminal act—including the sale of drugs—that he considers necessary to support and maintain his habit.”).
  131. .Id. at 1242 (Wright, J., dissenting).
  132. .See, e.g., Hernandez v. Johnson, 213 F.3d 243, 250 (5th Cir. 2000) (“Texas courts have consistently ruled that alcoholism may not be the basis for an involuntary intoxication defense . . . .”); See v. State, 757 S.W.2d 947, 950 (Ark. 1988) (collecting cases and observing that “most jurisdictions have held that an irresistible compulsion to consume intoxicants caused by a physiological or psychological disability does not render the ensuing intoxication involuntary”); see also Morse, supra note 119, at 436 (summarizing contemporary law as “avoid[ing] expanding a defense based on addiction raised by the Moore dissenters”).
  133. .See Wayne R. LaFave, Criminal Law 481 (4th ed. 2003) (“The mere fact that the defendant is an alcoholic or addict is not sufficient to put his intoxicated or drugged condition into the involuntary category.” (footnotes omitted)); see also Sanford H. Kadish, Excusing Crime, 75 Calif. L. Rev. 257, 287 (1987) (“There are enough conscious, purposive actions in the characteristic behavior of addicts (including abstinence when the motivation is great enough) that it cannot possibly be considered involuntary.”).
  134. .Daniel M’Naghten’s Case [1843] 8 Eng. Rep. 718 (PC), 722 (appeal taken from HL); see also Paul H. Robinson & Tyler Scot Williams, Insanity Defense, in Mapping American Criminal Law: Variations Across the 50 States (forthcoming 2018) (manuscript at 2–3) (surveying jurisdictions), [].
  135. .Model Penal Code § 4.01(1) (Am. Law Inst., 1962); see also Robinson & Williams, supra note 134 (manuscript at 3–5) (surveying jurisdictions). The MPC rule also allows an insanity defense where the defendant shows he lacked substantial capacity to “appreciate” either the “criminality” or “wrongfulness” of his conduct. Model Penal Code § 4.01(1).
  136. .731 F.2d 243 (5th Cir. 1984).
  137. .The case was decided before the Insanity Defense Reform Act of 1984, Pub. L. No. 98-473, 98 Stat. 2057 (codified as amended at 18 U.S.C. § 17 (2012)) went into effect.
  138. .Lyons, 731 F.2d at 245 (quoting Bailey v. United States, 386 F.2d 1, 4 (5th Cir. 1967)).
  139. .Id.; see also Kadish, supra note 133, at 286 (rejecting the theory that addiction could serve as grounds for an insanity defense because “[i]t is hard to see how addiction could qualify as a disease of the mind in the sense of a condition negating moral agency”).
  140. .See, e.g., Commonwealth v. Tate, 893 S.W.2d 368, 371 (Ky. 1995) (relying on “dissension in the medical community as to whether addiction is a mental disease” to reject an addiction-as-insanity defense); Commonwealth v. Herd, 604 N.E.2d 1294, 1298 (Mass. 1992); State v. Ingram, 607 S.W.2d 438, 441 (Mo. 1980); State v. Herrera, 594 P.2d 823, 830 (Or. 1979).
  141. .See, e.g., Cal. Penal Code § 29.8 (2017) (“In any criminal proceeding in which a plea of not guilty by reason of insanity is entered, this defense shall not be found by the trier of fact solely on the basis of . . . an addiction to, or abuse of, intoxicating substances.”); Morse, supra note 119, at 437 (noting that multiple U.S. jurisdictions “explicitly exclude addiction (or related terms) as the basis for an insanity defense despite the inclusion of this class of disorder in the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders”). The federal Insanity Defense Reform Act, too, has been interpreted to preclude consideration of voluntary substance abuse. United States v. Garcia, 94 F.3d 57, 61–62 (2d Cir. 1996); United States v. Knott, 894 F.2d 1119, 1121 (9th Cir. 1990).
  142. .Lyons, 171 F.2d at 247.
  143. .Jeff Feix & Greg Wolber, Intoxication and Settled Insanity: A Finding of Not Guilty by Reason of Insanity, 35 J. Am. Acad. Psychiatry & L. 172, 173 (2007) (emphases added).
  144. .People v. Kelly, 516 P.2d 875, 881, 883 (Cal. 1973) (en banc).
  145. .People v. Skinner, 228 Cal. Rptr. 652, 660–61 (1986); see also Cal. Penal Code § 29.8 (2017) (prohibiting insanity defenses “solely on the basis of . . . an addiction to, or abuse of, intoxicating substances”).
  146. .Powell v. Texas, 392 U.S. 514, 535 (1968).
  147. .See supra subpart III(B).
  148. .See supra subpart III(C).
  149. .See supra subpart III(A).
  150. .Stuntz, supra note 118, at 69.
  151. .See supra subpart I(A).
  152. .Leonard, supra note 11, at 452; see also supra notes 43–47 and accompanying text.
  153. .U.S. Const. amends. V, VI.
  154. .In re Winship, 397 U.S. 358, 364 (1970).
  155. .Fed. R. Evid. 609(a)(1) (governing the impeachment of witnesses with evidence of criminal convictions “punishable by death or by imprisonment for more than one year”). The Rule tracks criminal law’s definition of crimes that are “generally regarded as felony grade.” Fed. R. Evid. 609 advisory committee’s note to subdiv. (a). For instance, the Armed Career Criminal Act, which prohibits felons from purchasing firearms, defines a “violent felony” as a violent crime “punishable by imprisonment for a term exceeding one year.” 18 U.S.C. § 924(e)(2)(B) (2012).
  156. .See Fed. R. Evid. 609(a)(2) (governing the impeachment of witnesses with convictions for which “the elements of the crime require[] proving . . . a dishonest act or false statement”); see also Fed. R. Evid. 609 advisory committee’s note to 2006 amendments (“Ordinarily, the statutory elements of the crime will indicate whether it is one of dishonesty or false statement.”); United States v. Lewis, 626 F.2d 940, 946 (D.C. Cir. 1980) (excluding evidence of conviction for narcotics distribution offered to impeach witness because, “[w]hile narcotics may be sold in a manner that is ‘deceitful,’ . . . the statutory elements of offenses under the Controlled Substances Act do not require that the drug be sold or possessed in a manner that involves deceit, fraud or breach of trust” (footnotes omitted)); Logan v. Drew, 790 F. Supp. 181, 183 (N.D. Ill. 1992) (holding evidence of misdemeanor conviction for unlawful use of a credit card admissible to impeach witness because the crime “explicitly includes as an element the intent to defraud”).
  157. .See United States v. Jefferson, 623 F.3d 227, 234 (5th Cir. 2010) (reciting the rule that prior convictions are “automatically admissible” if, per the elements of a criminal offense, the convictions were for crimes involving dishonesty or false statements); Walker v. Horn, 385 F.3d 321, 333 (3d Cir. 2004) (same).
  158. .Restatement (Third) of Torts § 14 (Am. Law Inst. 2010).
  159. .Young v. Red Clay Consol. Sch. Dist., 122 A.3d 784, 843 (Del. Ch. 2015). The Young court relied on the fact that tort law looks to the criminal law in the realm of negligence per se for its conclusion that “criminal statutes . . . remain relevant as evidence of the floor for permissible electoral conduct.” Id. at 843.
  160. .Restatement (Second) of Contracts § 176 (Am. Law Inst. 1981).
  161. .See Unif. Trade Secrets Act, § 1(1), 14 U.L.A. 537 (1985) (defining “improper means” as including theft and bribery); Robert G. Bone, A New Look at Trade Secret Law: Doctrine in Search of Justification, 86 Calif. L. Rev. 241, 258 (1998) (explaining that trade secret law justifies treating misappropriation of trade secrets “through criminal wrongdoing, such as theft or fraud” as giving rise to trade secret liability because “[i]ndependently wrongful conduct of this sort improperly invade[s] the owner’s zone of secrecy”).
  162. .See H.L.A. Hart, Legal Responsibility and Excuses, in Punishment and Responsibility: Notes in the Philosophy of Law 28, 35–40 (1968); H.L.A. Hart, Negligence, Mens Rea and Criminal Responsibility, in Punishment and Responsibility: Notes in the Philosophy of Law 136, 149–57. The positivist position is not without its detractors. E.g., Stephen J. Morse, Culpability and Control, 142 U. Pa. L. Rev. 1587, 1587–88 (1994) (“If it is true that an agent really could not help or control herself and was not responsible for the loss of control, blame and punishment are not justified on any theory of morality and criminal punishment.”).
  163. .See R. A. Duff, In Defence of One Type of Retributivism: A Reply to Bagaric and Amarasekara, 24 Melb. U. L. Rev. 411, 412 (2000) (arguing that criminal punishment is justified “as an exercise in moral communication”); Dan Kahan, What Do Alternative Sanctions Mean?, 63 U. Chi. L. Rev. 591, 602 (1996) (“The proper retributive punishment is the one that appropriately expresses condemnation and reaffirms the values that the wrongdoer denies.”); see also Joel Feinberg, The Expressive Function of Punishment, 49 Monist 397, 401 (1965) (“That the expression of the community’s condemnation is an essential ingredient in legal punishment is widely acknowledged by legal writers.”).
  164. .See, e.g., Nathan Hanna, Say What? A Critique of Expressive Retributivism, 27 L. & Phil. 123 (2008) (arguing that because “the use of punishment to express criticism is conventional,” the value that such criticism “can play in a justification of punishment” is minimal).
  165. .Harold G. Grasmick & Donald E. Green, Legal Punishment, Social Disapproval and Internalization as Inhibitors of Illegal Behavior, 71 J. Crim. L. & Criminology 325, 334 (1980) (summarizing conclusions of research into the deterrent effects of the criminal law).
  166. .Boldt, supra note 11, at 2322 (“Longstanding notions of blameworthiness rely on the characterization of individual defendants as moral actors.”); John O. Cole, Thoughts from the Land of And, 39 Mercer L. Rev. 907, 911 (1988) (“In the framework of our law, the defendant is seen through a prism of individual responsibility and free choice.”).
  167. .Rebecca Hollander-Blumoff, Crime, Punishment, and the Psychology of Self-Control, 61 Emory L.J. 501, 552 (2012) (“[T]he law draws many fine lines that do not, in fact, delineate control or lack thereof as they purport to do but instead reflect a normative judgment about the type of behavior involved.”).
  168. .Boldt, supra note 11, at 2323–24.
  169. .Powell v. Texas, 392 U.S. 514, 532 (1968).
  170. .Id. at 561 (Fortas, J., dissenting).
  171. .Montana v. Egelhoff, 518 U.S. 37, 50 (1996).
  172. .United States v. Moore, 486 F.2d 1139, 1145 (D.C. Cir. 1973) (Wilkey, J., concurring).
  173. .Id. Indeed, Judge Wilkey put a mathematical point on his assertion that an addict’s strength of character determines whether he will use drugs: “[I]f the addict’s craving is 4 on a scale of 10, and his strength of character is only 3, he will have a resulting loss of self-control . . . .” Id.
  174. .Id. at 1234 (Wright, J., dissenting).
  175. .United States v. Lyons, 731 F.2d 243, 245 (5th Cir. 1984) (quoting Bailey v. United States, 386 F.2d 1, 4 (5th Cir. 1967)).
  176. .Professor Leonard, in the most thoughtful and extensive treatment of addiction evidence to date, analyzed the Seventh Circuit’s decision in United States v. Cunningham, 103 F.3d 553 (7th Cir. 1996). See supra notes 61, 79 and accompanying text (discussing Cunningham). He argued that the decision was proper in light of scientific consensus that “drug addiction can be explained at least in significant part as a brain disorder . . . .” Leonard, supra note 11, at 533.
  177. .Barrett Anderson observed the tension between “the older conception of temperance as a trait of character” and the “newer scientific findings indicating that alcoholism is genetic.” Anderson, supra note 20, at 1956–57. But, although he acknowledged that local criminal laws may inform the question, he concluded that the issue should be resolved according to whether “local moral overtones” against intoxication would prejudice jurors from a given locale. Id.
  178. .Understood this way, the chain of inferences includes an improper inference of the defendant’s “decision.” For a response to the argument that the character rules only prohibit inferences of “action,” not mental processes, see infra section IV(B)(2).
  179. .103 F.3d 553 (7th Cir. 1996).
  180. .Id. at 556–57.
  181. .See infra notes 69–73 and accompanying text (discussing courts that apply such additional requirements).
  182. .This example is drawn from United States v. Miranda, 986 F.2d 1283, 1285 (9th Cir. 1993).
  183. .See id. at 1284.
  184. .For another instance of contemporaneous addiction evidence, see United States v. Bitterman, 320 F.3d 723 (7th Cir. 2003), in which, “apparently as a result of heron withdrawal,” Bitterman vomited outside the bank he was accused of robbing. Id. at 725. Though his withdrawal symptoms show that Bitterman was addicted while robbing the bank, they were only relevant to the bank robbery if the jury could conclude that he acted to alleviate those symptoms (i.e. to acquire and use heron) by robbing the bank, a conclusion that depends on Bitterman’s propensity to act on his addiction.
  185. .See, e.g., People v. Johnson, 547 N.Y.S.2d 747, 749 (N.Y. App. Div. 1989) (affirming admission of the defendant’s statements to police that “he was drug dependent and needed money to supply his drug habit”).
  186. .For instance, the Sixth Circuit in United States v. Washam, 468 F. App’x 568 (6th Cir. 2012), should not have allowed the admission, in a bank robbery trial, of a defendant’s statement that he had previously robbed a different bank to support his cocaine addiction. See id. at 572.
  187. .See Fed. R. Evid. 404(a)(1), (b)(1).
  188. .See Model Penal Code § 1.13(2) (Am. Law Inst., 1962) (defining an “act” as “a bodily movement”). This Note argues that criminal law should have force in the law of evidence with respect to defining whether conduct is immoral. Hence the need to address criminal law’s definition of “act.”
  189. .Bright-line distinctions between physical acts and mental processes would swallow the character rule. This becomes most apparent in “intent” cases. Intent is a mental state, not an act, and the prosecution must prove intent for all true crimes. So allowing the accused’s bad acts to prove intent (on the theory that character evidence is admissible to prove mental states) would allow the admission of any and all of the accused’s prior convictions. Imwinkelried, supra note 34, at 579–80; see also Thompson v. United States, 546 A.2d 414, 421 (D.C. 1988) (“Intent is an element of virtually every crime. If the ‘intent exception’ warranted admission of evidence of a similar crime simply to prove the intent element of the offense on trial, the exception would swallow the rule.”); Lee E. Teitelbaum & Nancy Augustus Hertz, Evidence II: Evidence of Other Crimes as Proof of Intent, 13 N.M. L. Rev. 423, 431–32 (1983) (arguing that, because intent to keep taken property “is an element of all theft offenses,” allowing character evidence to prove intent regardless of whether it is in dispute “would routinely allow evidence of other thefts by the accused”).
  190. .See supra notes 60–61, 74–77 and accompanying text.
  191. .See supra notes 27–32 and accompanying text.
  192. .Imwinkelried, supra note 34, at 584 (“The application of the laws of the physical sciences can help predict the accused’s physical reaction. It is the mental component of the accused’s conduct which introduces the element of unpredictability.”). Professor Imwinkelried argues that allowing character evidence to prove a mental state, and the mental state to then prove conduct, would allow an end-run around Rule 404(b). See generally id.
  193. .Id. at 583 (observing that a jury’s conclusion that the defendant has a “warped mind inclined to criminal intent” is the “very type of revulsion which the character evidence prohibition is designed to guard against”).
  194. .See supra notes 27–32 and accompanying text.
  195. .See supra notes 165–168 and accompanying text.
  196. .See Teitelbaum & Hertz, supra note 189, at 431 (arguing that distinguishing mental states from acts requires a “strained and improbable” reading of the character rule). Teitelbaum and Hertz rely on the common sense proposition that “most character traits about which we are concerned include some mental element that is essential to their definition.”
  197. .See Michael Louis Corrado, Addiction and the Theory of Action, 25 Quinnipiac L. Rev. 117, 146 (2007) (arguing that if punishment is justified on the bases of “responsibility and desert,” addiction defenses should be recognized); Emily Grant, Note, While You Were Sleeping or Addicted: A Suggested Expansion of the Automatism Doctrine to Include an Addiction Defense, 2000 U. Ill. L. Rev. 997 (2000) (proposing an addiction defense akin to the involuntary-act defense afforded to sleepwalkers); Patrick Eoghan Murray, Comment, In Need of a Fix: Reforming Criminal Law in Light of a Contemporary Understanding of Drug Addiction, 60 UCLA L. Rev. 1006 (2013) (proposing a partial defense of addiction on a “semi-voluntary act” theory).
  198. .Boldt, supra note 11, at 2306; Morse, supra note 119, at 442–43.
  199. .Morse, supra note 119, at 443.
  200. .I am grateful in particular to Professor Steven Goode, TLR’s Notes office, and to KC, who had no idea she was signing up to be a sounding board.