Crime as Communication


		Sign up for the AMDCO Newsletter.

Crime as Communication: An Interpretive Theory of the Insanity Defense and the Mental Elements of Crime.

BENJAMIN B. SENDOR *

* Assistant Professor of Public Law and Government, Institute of Government, University of North Carolina at Chapel Hill. A.B. 1973; J.D. 1976, Harvard University. For their helpful comments on an earlier draft of this article, I thank Richard J. Bonnie, Elizabeth Dowling, Herbert Fingarette, Jean Hampton, F. Patrick Hubbard, Robert E. Rosen, and Roderick M. Stewart. I also thank David Bear for his valuable research suggestions.

SUMMARY:
... This article adopts Austin's suggestion and focuses on a particular legal excuse, the insanity defense. ... It considers the insanity defense as an excuse that negates criminal responsibility by rebutting common-sense, everyday inferences about the meaning of conduct, inferences drawn in criminal law through the elaborate interpretive system of the mental elements of crime. ... The commentary to an earlier draft of the test explained that an effort to exclude nondeterrable defendants from punishment must take account of impairment by mental illness of both cognitive and volitional capacities. ... In summary, the insanity defense arises when an offender's mental illness deprives him of the cognitive or volitional capacity required to adhere to relevant legal or moral norms. ... The interpretive analysis of the insanity defense and other excuses raises an exciting question: If the core of criminal responsibility is the disrespect shown by actions that injure the legally protected interests of others, and if excuses represent conduct that does not convey this disrespect, then how does the criminal law account for normal cases, where the conduct in question does convey such disrespect? The answer, I propose, is that criminal law addresses meaning through the intricate system of mental elements known as mens rea and the voluntariness component of actus reus. ...

TEXT:
[*1371] I. INTRODUCTION

In his essay, A Plea for Excuses, philosopher J. L. Austin suggested that a study of excuses can illuminate the nature of human action. He wrote: "[T]o examine excuses is to examine cases where there has been some abnormality or failure: and as so often, the abnormal will throw light on the normal, will help us to penetrate the blinding veil of ease and obviousness that hides the mechanisms of the natural successful act." n1 This article adopts Austin's suggestion and focuses on a particular legal excuse, the insanity defense. It attempts to understand the underlying rationale of this excuse and use it as a compass to point beyond itself and orient us to fundamental issues concerning the nature of activity subject to the criminal sanction.

The traditional view adopted by most contemporary courts and commentators characterizes the insanity defense as an excuse that negates criminal responsibility by making an exception, where a mental disease or defect is present, to the premise that a person is an autonomous agent who freely chooses his own conduct. n2 The capacity of free choice of free will is typically seen as resting on either or both of two other human capacities: cognition and volition. n3 This article presents a different perspective of the insanity defense. It considers the insanity [*1372] defense as an excuse that negates criminal responsibility by rebutting common-sense, everyday inferences about the meaning of conduct, inferences drawn in criminal law through the elaborate interpretive system of the mental elements of crime.

The article presents the thesis that criminal law uses the insanity defense -- as well as other excuses and, indeed, the very framework of the mental elements of crime -- to acknowledge the importance of the meaning a person conveys to other people through conduct. The meaning relevant to criminal law is the respect a person shows for legally protected interests of other people or the community by acting in a way that avoids injuring those interests. Crime is regarded as a kind of conduct that expresses disrespect for such interests. Mens rea elements and the voluntariness component of actus reus, including their intricate classification of states of mind and excuses, function together as the criminal law's systematic guide for interpreting such expression. Excuses, including the insanity defense, arise when a defendant's action does not convey the disrespect normally conveyed by criminal conduct. By proposing this view of the insanity defense, other excuses, and the mental elements of crime -- a view I call the "interpretive theory" -- I do not intend to reject the importance of choice, cognition, or volition in determining criminal responsibility. Rather, I intend to use this focus on the meaning an individual conveys through his actions to help understand why choice, cognition, and volition are relevant in assessing criminal responsibility.

II. OVERVIEW OF VARIOUS APPROACHES TO THE INSANITY DEFENSE

A. HISTORY OF CRIMINAL RESPONSIBILITY AND THE INSANITY DEFENSE PRIOR TO M'NAGHTEN'S CASE

A review of the history of the insanity defense shows that the view of the defense as resting on cognitive and volitional incapacities has an ancient pedigree, a tradition that can enlighten discussion today. This part discusses general mental criteria of criminal responsibility and specific criteria of the insanity defense that courts and commentators used or proposed before the landmark decision of M'Naghten's Case. n4 This discussion identifies the capacities historically considered important in assessing criminal responsibility.

The framework of Anglo-American analysis of excuses, including the insanity defense, can be traced to Aristotle. n5 According to Aristotle, "voluntariness" is a necessary condition of blameworthiness; we do not blame a person for "involuntary" acts. n6 He divided involuntary acts into two categories distinguished with respect to cognitive and volitional problems: acts committed "owing to ignorance" [*1373] and acts committed "under compulsion." n7

From Braction in the thirteenth century through Blackstone in the eighteenth century, English commentators adopted Aristotle's identification of the capacities of cognition and volition as the twin bases of justification for excuses in their explanation of the exculpatory character of insanity. n8 Their analyses of the significance of cognition and volition fall into three patterns: (1) cognitive capacity regarded as a necessary element of criminal responsibility; (2) both cognitive and volitional capacities regarded as necessary elements of criminal responsibility; and (3) volitional capacity regarded as the ultimate criterion of criminal responsibility, with cognitive capacity in turn serving as a condition of volitional capacity.

The first pattern, stressing the primacy of cognitive capacity, is represented by Fitzherbert and Coke. Fitzherbert, writing in the sixteenth century, identified memory and discretion as cognitive capacities essential for criminal liability. n9 Coke, writing in the seventeenth century, focused on the cognitive capacities of memory and under standing as criteria of criminal responsibility. n10

[*1374] Bracton, Lambard, and Dalton fit the second pattern, identifying both cognition and volition as elements of criminal responsibility. Bracton cited the volitional concepts of will, intention, and purpose as necessary conditions of criminal liability generally. n11 With specific reference to cases in which a defendant's sanity is in question, Bracton invoked the cognitive capacities of sense and reason. n12

At the end of the sixteenth century, Lambard also selected cognition and volition as elements of criminal responsibility, mentioning the cognitive capacities of knowledge and understanding (including knowledge and understanding of good and evil) and the volitional capacity of will. n13 Similarly, in his seventeenth century work The Country Justice, Dalton identified the cognitive capacities of [*1375] knowledge -- of good and evil -- and the volitional capacities of intent, will, and purpose as conditions of responsibility. n14

Hale, Hawkins, and Blackstone represent the third pattern among pre-M'Naghten commentators: those who viewed volitional capacity as the basic mental criterion of criminal liability, and who saw cognitive capacity as a secondary criterion, a condition of volitional capacity. n15 In The History of the Pleas of the Crown, Hale mentioned the cognitive capacities of understanding, memory, and reason, and the volitional capacity of will, as crucial factors in criminal liability. Regarding the connection between cognition and volition, he wrote:

The content of the will is that, which renders human actions either commendable or culpable . . . . And because the liberty or choice of the will presupposeth an act of the understanding to know the thing or action chosen by the will, it follows that, where there is a total defect of the understanding, there is no free will in the choice of things or actions. n16

Against this analytic background, stressing volition as the direct criterion of [*1376] responsibility, Hale then focused on cognitive disability as the feature of insanity that precludes volitional capacity and, therefore, excuses conduct: "[T]he best measure I can think of is this; such a person as labouring under melancholy distempers hath yet ordinarily as great understanding, as ordinarily a child of fourteen years hath, is such a person as may be guilty of treason or felony." n17 Similarly, Hawkins stressed volition as the major mental criterion of criminal liability, regarding cognitive capacity as a condition of volition. n18 Again, within that analytic context, he then identified cognitive capacity as the key factor in the insanity defense. n19

Finally, Blackstone, whose Commentaries played a major role in the early development of American law, followed the view of Hale and Hawkins concerning the relationship between cognition and volition in criminal liability. He wrote: "All the several pleas and excuses, which protect the committees of a forbidden act from the punishment which is otherwise annexed thereto, may be reduced to this single consideration, the want or defect of will." n20 Then, with specific reference to insanity, he identified defective understanding as the symptom that robs a person of his will and, therefore, precludes criminal liability. n21

Reported cases in the eighteenth and early nineteenth centuries indicate that judges and attorneys in that period shared the view of pre-M'Naghten commentators that the exculpatory character of the insanity defense rests on cognitive and volitional incapacities. Their discussions of the role of cognition and volition can be divided into the same three patterns discussions by pre-M'Naghten commentators. n22

Two cases provide good examples of the first pattern, citing only cognitive [*1377] incapacity as relevant to the insanity defense. In Bowler's Case, n23 a murder case tried in 1812, Justice LeBlanc's jury instructions set forth a cognitive test for insanity that foreshadowed the M'Naghten test. LeBlanc charged the jury to consider whether, at the time of the killing, the defendant was capable of distinguishing between right and wrong, and whether a delusion prevented him from understanding the nature of his act. n24 In Rex v. Offord, n25 an 1831 murder case, Lord Lyndhurst also used a strictly cognitive test in his jury instructions concerning the insanity defense. n26

Two other cases illustrate the second pattern, citing both cognitive and volitional incapacities as factors to be considered in assessing an insanity defense. The imaginative and persuasive case presented by defense counsel Thomas Erskine highlighted an 1800 trial of a would-be assassin of King George III. n27 James Hadfield was a former British soldier who suffered from the delusion that he must die to save the world. Because he also believed that he should not commit suicide, he concluded that execution was his appropriate form of self-sacrifice. He decided to kill the king to bring about his desired martyrdom through execution.

Hadfield's conduct demonstrated that intellectually he understood the nature of his act and the difference between good and evil. Indeed, in Hadfield's mind it was the special nature of his act -- assassinating the king -- and the enormity of its evil that would lead to the execution he sought and the greater good of world redemption. Accordingly, Erskine argued in favor of an insanity defense test different from the strictly cognitive test then in use. Instead of a cognitive test based on the criterion of understanding the difference between good and evil, Erskine proposed a mixed cognitive and volitional test. Erskine urged that Hadfield be acquitted because he suffered from an insane delusion that gave him an irresistible impulse to assassinate the king. n28

[*1378] The trial of Regina v. Oxford n29 involved another attempted assassination of royalty, this time an assault against Queen Victoria in 1840. The prosecutors cited a strictly cognitive test of insanity involving knowledge of the nature, quality, and consequence of one's act, and knowledge of whether the act was wrong, a position falling within the first pattern discussed above. n30 Lord Denman's jury instructions, however, contained both cognitive and volitional components. The cognitive portion mentioned knowledge of the nature, character, and consequences of one's act and knowledge of whether the act was wrong. In the volitional segment of his instructions, Denman used the concept of irresistible impulse. n31

Three more cases exemplify the third pattern, in which volitional incapacity is seen as the key feature of insanity and cognitive incapacity is regarded as precluding volitional capacity. In Rex v. Arnold, n32 Justice Tracy used a cognitive test for the specific issue of insanity in his jury instructions, mentioning the cognitive capacities of reasoning, using one's senses, knowing what one is doing, and distinguishing between good and evil. He also told the jury, however, that cognitive capacity is relevant because it is a condition of volitional capacity, the essential mental element directly linked to criminal responsibility. n33

[*1379] The procedure in Ferrers' Case, n34 in which the House of Lords convicted a peer of murder, did not involve jury instructions. Lord Ferrers' argument in his own defense concerning insanity falls within the second pattern, mixing volitional and cognitive elements. Ferrers referred to the cognitive capacity "to know the difference between a moral and immoral action," then he introduced the volitional concept of irresistible impulse to English insanity law. n35 The prosecutor's argument fits within the third pattern. He sought to harness the cognitive capacities of reason, thought, design, understanding the nature of actions, and understanding the difference between good and evil to the service of the volitional capacity to restrain one's passions. n36

Finally, a report of Bellingham's Case n37 also exemplifies the third pattern of cases and commentary. Two different accounts of the jury instructions of Lord Chief Justice Mansfield mention the cognitive capacities of reasoning, judgment, and distinguishing between right and wrong and good and evil. According to one report, Mansfield also viewed cognitive capacity as a condition of the volitional capacity of intention. n38

[*1380] This analysis of the early history of the insanity defense shows that for six centuries before M'Naghten's Case, commentators, judges, and attorneys identified a number of specific capacities as relevant to the exculpatory character of the insanity defense. These capacities can be placed into two basic groups, cognition and volition, with each group containing an assortment of more specific capacities.

B. MODERN INSANITY DEFENSE TESTS

An insanity defense test can perform several functions. First and foremost, it guides a jury in determining whether to acquit a defendant by reason of insanity. It also guides reviewing courts, prosecutors, and defense attorneys in their assessments of whether a defendant has a valid insanity defense. By defining the criteria of insanity, a test also places limits on the types and severity of mental illnesses and incapacities covered by the insanity defense, thereby expressing and promoting legislative or judicial policy about the nature and range of people and conduct subject to criminal liability. n39 Furthermore, a test can guide and limit attorneys and witnesses in preparing testimony to support or oppose the insanity defense in a particular case. n40

This part analyzes the major modern insanity defense tests used or proposed in Anglo-American law, beginning with the M'Naghten test, to discuss yet another task of an insanity defense test: to identify incapacities relevant to exculpation and, thereby, to identify capacities relevant to responsibility. By identifying such incapacities and capacities, a test expresses a view of what it is about mental illness in some cases that prompts the law to excuse conduct that otherwise [*1381] would be criminal. n41 The purpose of this analysis is not to endorse or criticize particular tests, but to ascertain which capacities they implicate as relevant to responsibility and exculpation.

1. M'Naghten Test

M'Naghten's Case triggered a sharp, complex debate over the appropriate criteria for the insanity defense, a debate that continues today. Significantly, the framework of the discussion remains largely unchanged from the pre-M'Naghten period, with courts and commentators still focusing on cognitive and volitional capacities as the possible key elements of the insanity defense. Under the M'Naghten test, a defendant is insane if, at the time of his criminal conduct in question, he was "labouring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing; or, if he did know it, that he did not know he was doing what was wrong." n42 As the earlier review of pre-M'Naghten cases and commentary demonstrates, M'Naghten's Case did not introduce the concepts embraced by this test. Instead, it crystallized them into a formula derived from tests already used in previous cases.

Plainly, the M'Naghten test is a cognitive test. What are the cognitive functions encompassed by the M'Naghten test? The report of the House of Lords session in which the test was presented sheds little light on the question. The meanings of the two parts of the test -- knowledge of the nature and quality of an act and knowledge that an act is wrong -- are ambiguous. Fortunately, in the 143 years since M'Naghten the meanings of the two parts have been explored in a rich history of interpretation by courts and commentators.

Courts have interpreted the test in a variety of ways, ranging from restrictive to expansive. For example, one court has defined the first part of the test narrowly to refer only to mere awareness of the physical nature and likely physical consequences of an act. n43 Other courts, however, have interpreted the first part more broadly, to include such cognitive faculties as understanding one's relation to other people; n44 power to comprehend rationally the nature or consequences of an act; n45 possession of mind, capacity, reason, and understanding sufficient to judge the nature, character, and consequences of an act; n46 capacity for rational appraisal of the nature and consequences of an act; n47 and capacity to appreciate rationally the significance of an act. n48 These examples demonstrate that cognition, under the M'Naghten test, may be viewed not only as mere perceptual and [*1382] intellectual awareness of the physical nature of conduct, but also as the capacity to reason actively and understand the impact and significance of one's conduct for others. n49

Some courts have interpreted the second part of the M'Naghten test narrowly to refer only to knowledge of the legal wrongfulness of conduct. n50 Other courts, however, have read the second part of the test to encompass knowledge of both the legal and moral wrongfulness of conduct. Under the latter view, a defendant is acquitted by reason of insanity if he lacks either type of knowledge. n51 Furthermore, some courts have considered the relevant knowledge to include emotional or spiritual understanding of the wrongfulness of conduct, beyond mere intellectual knowledge of wrongfulness. n52

Commentators also have subjected the cognitive capacity test encompassed by M'Naghten to a variety of interpretations. For example, Williams considered the first part of the test to refer merely to knowledge of physical circumstances and consequences. n53 Other commentators have interpreted this part more expansively. Hall and Perkins and Boyce, for instance, regard it as referring to the capacity for rationality. n54 Weihofen suggested that the capacity assessed under the M'Naghten test should include capacity for emotional appreciation of the significance of conduct. He wrote: "[I]f the word 'know' were given this broader interpretation, so as to require knowledge 'fused with affect' and assimilated by the whole personality -- so that, for example, the killer was capable of identifying with his prospective victim -- much of the criticism of the knowledge test would be met." n55 In 1883, Stephen urged adopted of a broad test of cognitive capacity. He proposed that the relevant knowledge should include knowledge of moral as well as legal wrongfulness of conduct. n56 He also suggested that the pertinent capacity should include active moral reasoning -- "the power of passing a rational judgment on the moral character of [an] act" n57 -- as well as mere intellectual awareness of its wrongfulness. Livermore and Meehl have similarly maintained that knowledge under the M'Naghten test should entail the capacity to engage in the active reasoning that enables a person to take account [*1383] of the wrongfulness of conduct when acting. n58

In sum, courts and commentators have identified a broad range of cognitive functions relevant to criminal responsibility under the M'Naghten test. These range from simple perceptual and intellectual awareness of the physical circumstances, nature, consequences, and legal wrongfulness of conduct; to awareness of the moral wrongfulness of conduct; to the capacity to incorporate such awareness into reasoning about conduct; to emotional appreciation of the circumstances, nature, consequences, and wrongfulness of conduct. Although there is no consensus about a single correct interpretation of the pertinent cognitive functions, this discussion has shown the possible range of such functions. n59

2. Irresistible Impulse Test

The classic exposition of the irresistible impulse test was given in 1886 in Parsons v. State. n60 In Parsons, Judge Somerville wrote that criminal responsibility requires both cognitive capacity and "freedom of will." That is, a mentally ill person is not criminally responsible if his illness destroys his power to choose between right and wrong, even if he understands the difference between right and wrong. n61 As Somerville wrote, a defendant should be acquitted by reason of insanity if he satisfies the M'Naghten criteria or "[i]f, by reason of the duress of such mental disease, he had so far lost the power to choose between the right and wrong, and to avoid doing the act in question, as that his free agency was at the time destroyed." n62 The irresistible impulse test has not been used as a substitute [*1384] for the M'Naghten test. Instead, its function has been to supplement the M'Naghten test in cases in which a defendant has the requisite cognitive capacity for criminal responsibility, but lacks the requisite volitional capacity. n63

As with the M'Naghten test, it will be useful to determine which capacities are encompassed by the irresistible impulse test. Some courts and commentators have stated that the test refers only to sudden and unplanned impulses, thereby excluding conduct preceded by planning, reflection, or brooding. n64 Other courts and commentators, however, have adopted a more expansive view of the test. They regard it more generally as a test of a defendant's capacity to control behavior, to refrain from wrongful conduct. This broader interpretation includes both sudden acts and planned acts within the ambit of the test. n65

A significant feature of all interpretations of the volitional capacity relevant to the test is the analysis of what might be called positive and negative elements. n66 The positive element is the will, intent, or purpose to engage in particular conduct. The negative element is refraining from engaging in particular conduct. Every voluntary act involves both the positive capacity to engage in conduct and, simultaneously, the negative capacity to refrain from committing other acts. Any choice requires both choosing to do something and choosing not to do other things. For example, when I drive a car, my conduct includes stepping on the accelerator -- but also not stepping on it too hard or too softly -- and steering to direct the car where I want to go -- while avoiding road hazards, pedestrians, and wrong turns. Although we often describe our conduct in positive terms, a full view of conduct includes negative elements as well.

The positive element of volition rarely is at issue in insanity defense cases. As several commentators have observed, many insane defendants fully intend to commit the criminal acts for which they are charged. Such people do not lack the positive volitional capacity to engage in criminal conduct. n67 Instead, volitional defense tests -- such as the irresistible impulse test and, as we shall see, the volitional prong of the Model Penal Code test -- focus on the second component [*1385] of volition, the negative capacity to control, resist, or avoid wrongful conduct. The psychological perspective underlying these tests is that some insane people lack the degree of self-control necessary to choose not to commit criminal acts. n68

3. Product Test

The product test set forth in Durham v. United States n69 was a modern restatement of a test introduced in 1870 in State v. Pike. n70 As Judge Bazelon stated in Durham, the product test provides "simply that an accused is not criminally responsible if his unlawful act was the product of mental disease or mental defect." n71 The salient characteristic of the product test for this inquiry is that it rejects the traditional legal approach of identifying particular capacities as essential to the exculpatory significance of mental illness. A major goal of the test, as explained in Durham, is to avoid linking nonresponsibility with specific symptoms. Although the Durham court acknowledged that specific cognitive and volitional incapacities would continue to be relevant, it sought to deemphasize the law's traditional focus on these incapacities as crucial elements in assessing responsibility. n72 The court sought to open the door for jury instructions and expert [*1386] testimony to all facts regarding a defendant's mental illness that might bear on whether that illness caused his criminal conduct. n73

[*1387] 4. Model Penal Code Test

The Model Penal Code test for the insanity defense provides: "A person is not responsible for criminal conduct if at the time of such conduct as a result of mental disease or defect he lacks substantial capacity either to appreciate the criminality [wrongfulness] of his conduct or to conform his conduct to the requirements of the law." n74 The express purpose of this test is to identify and excuse defendants who, because of mental illnesses, cannot be deterred by penal sanctions. n75 The test comprises both cognitive and volitional incapacities as criteria of insanity. The commentary to an earlier draft of the test explained that an effort to exclude nondeterrable defendants from punishment must take account of impairment by mental illness of both cognitive and volitional capacities. n76 This position rests on the view that both cognition and volition are vital functions in enabling a person to respond to the deterrent influence of punishment. n77 Thus, the Model Penal Code test follows common law in resting the insanity defense on the twin pillars of cognitive and volitional capacities. Accordingly, the drafters designed the test as a two-pronged rule providing for acquittal of a defendant who lacks either the requisite cognitive capacity or the requisite volitional capacity for criminal responsibility.

Although they retained the basic common law approach, the drafters also sought to cure some of the perceived flaws and ambiguities in the M'Naghten and irresistible impulse tests. The changes create an expansive blend of those two tests. With respect to the cognitive prong, the test uses the word "appreciate" rather than the word "know" used in the M'Naghten test. "Appreciate" was selected to indicate that the cognitive prong should include emotional awareness as well as intellectual awareness. n78 In regard to the relevant object of appreciation, [*1388] the test provides an option for legislatures or appellate courts to select either "criminality" or "wrongfulness"; the drafters thereby avoided taking a position on whether the requisite appreciation for responsibility refers to the immorality of conduct as well as to the illegality of conduct. n79

The volitional prong of the test uses the phrase "conform his conduct to the requirements of the law" rather than "irresistible impulse." The new language was selected to clarify that the test applies to planned acts as well as to sudden acts. n80 Another significant feature of the Model Penal Code test is its provision excusing a defendant who lacks "substantial" cognitive or volitional capacity. The drafters made it clear that by using the term "substantial" they intended to excuse a defendant whose cognitive or volitional capacity is "greatly impaired" as well as a defendant whose capacity is totally impaired. n81

5. New Federal Test

Congress enacted the Insanity Defense Reform Act of 1984 as part of the Comprehensive Crime Control Act of 1984. n82 Under the new federal insanity defense test, a defendant is not responsible for criminal conduct if, "as a result of a severe mental disease or defect, [he] was unable to appreciate the nature and quality or the wrongfulness of his acts." n83 The language of the test shows that it is a cognitive test, containing no volitional prong. n84 It blends elements of the [*1389] M'Naghten test and the cognitive prong of the Model Penal Code test. Congress adopted the Model Penal Code's use of the term "appreciate" n85 to designate the cognitive capacity at issue. The new test incorporates both the M'Naghten test's reference to awareness of the "nature and quality" of an act n86 and the Model Penal Code's reference to awareness of the "wrongfulness" of an act n87 to describe the types of appreciation in question. n88 To emphasize that nonpsychotic behavioral disorders or neuroses do not suffice to establish the defense, the test states that the defendant's mental illness must be "severe" to be exculpatory. n89 The federal test also deletes the Model Penal Code qualification that an incapacity due to mental illness is exculpatory if it is "substantial." n90

[*1390] C. THE IRRATIONALITY OF MENTALLY DISORDERED CONDUCT

This review of the major modern insanity defense tests has shown that most of those tests employ cognitive and/or volitional criteria. n91 The relative virtues and flaws of the tests have been debated vigorously since M'Naghten was announced. Nonetheless, the differences among the tests and the intensity of the debate surrounding them should not blind us to their underlying unity. It is important to note that most of the tests do rest on cognitive and/or volitional incapacities. n92 Instead of regarding cognitive and volitional incapacities merely as competing bases for tests, we should take a step back and consider why courts and commentators over the course of seven centuries have focused on these two types of incapacities as key features of legal insanity. As we have seen, Aristotle grasped this point and Anglo-American courts and legal commentators beginning with Bracton have repeatedly affirmed it. This leads us to the question: In what ways are cognition and volition related such that both types of capacities are relevant to criminal responsibility? The goal of this inquiry is not to develop yet another insanity defense test. Instead, its purpose is the more theoretical one of identifying the exculpatory characteristics of criminal conduct excused by the insanity defense.

Several commentators, beginning with Stephen in the nineteenth century, have identified irrationality as the feature that relates cognitive and volitional incapacities to each other and links them to responsibility. Stephen premised criminal responsibility on a general capacity for self-control, which he defined as the "power to attend to distant motives and general principles of conduct and to connect them rationally with the particular act under consideration." n93

In our time, Fingarette and Moore have written extensive analyses of the irrational quality of mentally disordered conduct. Fingarette defines rationality as the capacity to respond to factors that are morally and legally relevant to one's situation. Fingarette calls these morally and legally relevant factors "essential relevance," explaining: "It is this responsiveness to essential relevance which, in [*1391] the last analysis, constitutes the root notion, though not the entire meaning, in the concept of responsibility." n94 Accordingly, he defines legal insanity as the "failure to respond relevantly to what is essentially relevant by virtue of a grave defect in capacity to do so inherent at least for the time in the person's mental makeup." n95 The capacity for rational conduct -- for "rationally tak[ing] into account the implications of the act relevant to criminality" n96 -- plainly encompasses the full range of cognitive and volitional capacities discussed above, but it is not reducible to any of these capacities. Instead, Fingarette regards the capacity for rational conduct as the underlying principle that unites cognitive and volitional capacities and relates them to criminal responsibility. n97 That is, it is the impact of cognitive and volitional incapacities on the capacity for rational conduct that makes cognitive and volitional impairments exculpatory. n98

Similarly, Moore has proposed that the distinctive feature of mentally disordered behavior is impairment of what he terms -- drawing on Aristotle -- "practical [*1392] reasoning." n99 According to Moore, practical reasoning entails "the formulation of ends and the selection of means to the attainments of those ends" in light of circumstances known to the actor. n100 Moore views conduct involving competent practical reasoning as rational conduct. n101 He identifies several essential attributes of sound practical reasoning with respect to criminal responsibility: intelligibility of ends, consistency of ends, transitivity (ordering by preference) of ends, consistency and ends according to objective standards, intelligibility of means, consistency of means, coherence of means, consistency and coherence of means over time, and causal autonomy to act so as to attain selected ends. n102

In Moore's view, mental illness impairs various cognitive and volitional capacities such as perception, memory, imagination, learning, reasoning, feeling emotion, and the will to fulfill emotions and desires. n103 Because these capacities are necessary for practical reasoning, their impairment can preclude rational conduct. n104 In turn, because rational conduct is a condition of criminal responsibility, impairment of those cognitive and volitional capacities can preclude responsibility. n105 Other commentators such as Morse, n106 Livermore and [*1393] Meehl, n107 Hall, n108 and Feinberg, n109 have identified irrationality as the distinctive and exculpatory feature of mental illness. n110

The thread that runs through these similar views of the irrational and exculpatory quality of mentally disordered conduct can be summarized as follows: the distinctive character of mental illness with respect to criminal responsibility is the incapacity of a mentally ill person to guide his conduct according to the myriad of legally and morally relevant factors that can deter other people -- "normal" people -- from committing crimes. n111 An insanity acquittee lacks the capacity to take such factors into account, so that we are inclined not to blame him. n112

The legally and morally relevant factors that act as a deterrent to criminal conduct are varied and numerous. They function in all spheres of life: private, interpersonal,and social. They include sensory perceptions of and factual knowledge about events, circumstances, objects, and beings; moral values learned from and reinforced by family and community; moral values developed through reflection about one's own experience; religious doctrine; emotional concern about others, manifested by love, friendship, care, sympathy, and empathy; fear of shame; fear of guilt; n113 fear of jeopardizing friendships, social status, employment, or wealth; persuasion by others; fear of retaliation; fear of denunciation; legal prohibitions imposed by statutes, common law, ordinances, and agency regulations; fear of force used by law enforcement officers; and fear of fines, imprisonment, or execution. All these, and more, constitute features of our [*1394] social environment that the vast majority of people take into account in choosing whether to commit criminal acts.

The capacity to take account of such factors, to be guided by such influences in acting, requires many more specific capacities -- capacities so natural and normal that, as Austin suggested in the passage quoted in the first paragraph of this article, we easily can and do ignore their operation until one or more of them fail us. They include the cognitive capacities of sensory perception, factual and conceptual knowledge, memory, imagination, theoretical and practical reasoning, n114 and emotional awareness about circumstances and consequences of conduct. They also include the volitional capacities of voluntary muscle control, the selection and use of particular means to attain desired ends, and the avoidance of other means and ends, such as undesirable, immoral, or prohibited ends. In short, the capacity to bring the many morally and legally relevant factors to bear involves the full range of cognitive and volitional capacities discussed earlier.

Just as the capacity to respond to deterring factors requires a variety of specific cognitive and volitional capacities, so the incapacity to take account of these factors can be manifested in a variety of ways, such as lack of sensory, factual, or conceptual knowledge; lack of memory; lack of imagination; impaired reasoning; lack of emotional appreciation; lack of voluntary muscle control; and lack of control over the selection of means and ends of conduct. These specific impairments are examples of more general incapacity rather than definitive, limiting cases of incapacity. A particular mentally ill person might well exhibit symptoms relating to only one or a few of these faculties. n115 Each specific faculty, however, is relevant to assessing criminal responsibility because each bears on the central issue of a person's capacity to attend to factors that normal people can take into account when choosing to commit or not commit criminal acts.

III. A NEW APPROACH

A. THE INTERPRETIVE THEORY OF THE INSANITY DEFENSE

Why do we excuse people who, by reason of mental illness, lack the capacity to take account of morally and legally relevant factors in their conduct? Some courts and commentators have suggested that we excuse them out of fairness: [*1395] we should not blame such people because we cannot fairly expect them to have acted differently. n116 The distinctive quality of mental illness that prompts us to treat mentally ill offenders fairly is, according to some commentators, the inability to make a choice: our sense of fairness requires that we excuse a person whom we cannot expect to have chosen to comply with the law. n117 While I agree with that position, I maintain that we should explore still deeper to discover what is special about choice that makes it an essential ingredient of responsibility.

Utilitarian theorists have maintained that criminally insane defendants should be excused because punishment would not deter them and because excusal would not weaken the law's deterrent effect on normal people. n118 H.L.A. Hart has effectively criticized this position, however, pointing out that while the threat of punishment of insane defendants might not specifically deter those defendants [*1396] from committing crimes, punishment of such defendants nevertheless might well serve as an effective general deterrent to others. n119

Hart has proposed that excuses such as the insanity defense are appropriate because they promote a vision of society in which people can live with confidence and satisfaction about their capacity to control their lives through choices. Recognition of excuses signals to citizens that they have control over the possibility of facing punishment for transgressions. n120 Through this approach Hart seeks to explain the role of excuses as a matter of political theory, identifying a political value of our culture that he believes should help shape substantive criminal law. Although it is an appealing position as a matter of policial theory, his view does not explain our very real inclination to excuse people -- in daily life as well as in the criminal justice system. n121 To put it bluntly, we should consider why we do excuse people as well as why we should excuse them.

Other commentators have suggested that we excuse mentally ill offenders because we only blame beings who are like us; insane offenders are not sufficiently like us to warrant blame. We do not blame trees for falling on houses, boulders for tumbling down hills, animals for acting out aggressive instincts, or small children for their occasionally undisciplined behavior. Similarly, we do not [*1397] blame mentally ill offenders because they are not sufficiently similar to us to warrant regarding them as agents of moral responsibility. n122 This "jurisdictional" view of the insanity defense has historical roots in the early common law doctrines that likened insane people to children and "brutes." n123 It is an attractive view, identifying a significant element of our inclination to excuse mentally ill offenders. The issue arises, though, as to what distinguishes the vast majority of "normal" people (and some mentally ill people who are not criminally insane) whom we do blame, from those mentally ill people whom we do not blame. n124

I suggest that a key factor in our inclination to excuse mentally ill offenders lies in the special human capacity to convey meaning through conduct and to interpret meaning in the conduct of other human beings. We live in a world that is infused with meaning or significance for us. By "meaning," I refer, quite broadly, to the ways in which the world matters to us and takes form for us. n125 Of course, given such an expansive definition of meaning, perhaps all organisms live in world of meaning to some extent, from insects responding to one another's chemical or visual signals, to a bird responding to a mating song, to a dog tracking a scent, to a person reading a law review article. For human beings, language -- both written and spoken -- is a special and powerful tool for conveying meaning. n126 Conduct, however, also can express meaning. Throughout our [*1398] daily lives, we convey meaning through our actions and we interpret it in the actions of other people. n127

In this article I focus on a particular kind of meaning conveyed by conduct: the attitudes or dispositions an actor expresses toward the interests of other individuals and his community through his behavior. In everyday life, such attitudes can include love, friendship, care, concern, empathy, sympathy, respect, lack of care, indifference, disrespect, dislike, contempt, hostility, and others. I suggest that the specific attitude relevant to criminal responsibility is a particular type of respect: the quality of respect a defendant has shown toward the legally protected interests of other individuals and the community. n128 Criminal law demands [*1399] that a person act with adequate respect for those interests by acting in a way that avoids injuring them. This is the normative requirement imposed by the criminal law. It presents a vision of the ways in which we should, and even more importantly should not, act toward one another and toward our community. n129

[*1400] A person expresses his attitude toward those legally protected interests through his conduct. A wrongdoer's disrespect expressed by committing a wrongful act evokes our anger and prompts us to blame him, both in everyday life and through the criminal justice system. n130 Disrespect for legally protected interests can be expressed through other specific attitudes. These range from hostility, expressed by acting with intent to injure interests or with knowledge that one's actions will injure interests, to indifference (or recklessness), expressed by acting with knowledge that one's conduct might injure interests, to carelessness, expressed by acting without sufficient caution to avoid an unknown risk to interests. To use a simple example of conduct at one end of this range, a robber who commands a victim at gunpoint to hand over all the money in his pockets plainly shows hostility toward the victim's interests in life and property. Despite the obvious differences among these specific attitudes, they all constitute forms of disrespect for the legally protected interests of another person or the community. The differences among the specific attitudes are matters of degree.

Expression of an attitude toward the legally protected interests of other individuals and the community relates directly to two issues discussed above: (a) the legally and morally relevant factors that can deter criminal conduct, and (b) cognitive and volitional capacities. The legal and moral factors together convey the criminal law's normative standards of respect and, if adhered to, serve to shape or influence behavior to conform to those standards. These factors do not [*1401] deter us from committing crimes solely by rote obedience. They deter us also by prompting us to respect legally protected interests.

An actor with sufficient cognitive and volitional capacities to appreciate legally and morally relevant factors expresses his attitude toward legally protected interests when he uses such capacities either to conform to or to ignore those factors in his conduct. "Capacity" in this sense means both underlying cognitive and volitional abilities ingrained in an actor's character and situational cognitive and volitional abilities to conform to legally and morally relevant factors. A corollary of this principle -- crucial for our upcoming analysis of excuses -- is that the capacity to take account of those factors is a necessary condition of the meaning or expressiveness of conduct with regard to the issue of respect. Only a person who appreciates the circumstances, nature, and consequences of his conduct, and who can control his conduct, expresses respect or disrespect through his conduct. In sum, a phenomenological analysis of the components of meaningful criminal conduct indicates that cognitive and volitional capacities are essential conditions of conveying, through conduct, the type of meaning pertinent to criminal responsibility. If an actor lacks these capacities necessary for guiding his behavior according to legally and morally relevant factors, then his conduct does not express either respect or disrespect for legally protected interests.

Before further exploring the meaning relevant to criminal responsibility, we should briefly consider a preliminary matter: How can such meaning be ascertained reliably or authoritatively in the case of a particular defendant's conduct? The key to answering this question is understanding that in criminal law the meaning of action is not determined entirely subjectively or entirely objectively. It is an intersubjective meaning, created and interpreted by both actors and observers. n131

The intersubjective nature of the meaning of conduct has two important aspects. First, although in both everyday life and criminal law an actor and an observer often agree about the meaning of the actor's behavior, it is also possible for the actor's behavior to convey an attitude toward the interests of other people or the community that differs from the attitude he intended or realized. That is, even though it is the meaning of the actor's conduct that is at issue, his own understanding of his conduct does not necessarily exhaust the meaning of that conduct. Carelessness about another person's safety, for instance, conveys a message about the actor's disrespect for the other person's well-being, even though the actor himself is not aware of that message at the time of his conduct. Meaning, then, is not equivalent to a subjective "state of mind." It is not identical to the content of an actor's consciousness at the time of his action. Although an actor's state of mind is relevant to the meaning conveyed by his conduct, it does not conclusively determine that meaning. n132

[*1402] A second aspect of the intersubjective nature of the meaning relevant to criminal law is that Anglo-American law already has a long-standing interpretive procedural framework or mechanism that assigns the task of authoritative interpretation to particular participants in a criminal proceeding. In the context of a criminal case, several interpretive perspectives come actively into play, vying with one another for the claim to authoritativeness: those of the defendant, the victim (except in a victimless crime), witnesses, prosecuting and defense attorneys, the jury, and trial and appellate judges. Each of these participants may have a pertinent interpretation of the defendant's act, an interpretation that can help shape the final, authoritative interpretation. Anglo-American criminal procedure, however, designates the jury (or the trial judge in a bench trial) as the virtually authoritative interpretative voice, subject to judicial review only for serious errors. n133 As I discuss below, n134 the jury's interpretation is informed, [*1403] guided, and limited by the official definition of the alleged crime established by legislation or common law, the indictment or other charging document, evidence, rules of evidence, jury instructions, and attorneys' arguments. The important point here, however, is that the criminal law regards the jury's interpretation as the authoritative interpretation of the relevant meaning of a defendant's conduct. It is for the jury to interpret the defendant's behavior to determine whether he showed disrespect.

To summarize the analysis thus far, criminal law assigns to the jury the role of making a normative interpretation of the defendant's conduct. n135 An actor, such as a defendant, takes a position about the law's normative standards and expresses his attitude toward the legally protected interests they represent when he uses his cognitive and volitional capacities to take account of a large array of morally and legally relevant factors -- either by shaping his conduct to conform with those factors or by ignoring them. These capacities are necessary conditions for conveying the meaning through conduct relevant to this analysis of criminal law. n136

Using this analytic foundation, we may turn to situations that can arise both in everyday life and in criminal law in which an actor lacks the requisite capacities to take legally and morally relevant factors into account in acting. In such cases, his conduct does not convey the meaning ordinarily conveyed by identical physical action. It is in these situations, when the inferences about attitudes toward legally protected interests that we normally draw from a person's physical acts are inaccurate, that excuses -- including the insanity defense -- come into play. n137

A few examples illustrate this point. If a person fires a shotgun at another person, we naturally infer that he intends to hurt or kill the victim. Nevertheless, if the actor mistakenly believes that he is shooting at a deer, then we do not infer a similar intent to harm the victim's legally protected interest in life. Instead, we excuse him if he made a reasonable mistake. n138 If the hunter really [*1404] does aim at a deer, but the victim runs suddenly and unexpectedly into his line of fire as the trigger is pulled, too late for the hunter to avoid firing or to avert his aim, we once again excuse the hunter, this time because an accident occurred. In both of these cases of excuse, the hunter lacks the cognitive awareness of the circumstances and/or consequences of his actions so as to conform his conduct to legally and morally relevant factors. Consequently, his conduct does not convey disrespect for the victim's interest in life. n139

Similarly, if a five-year-old child shoots someone while playing with his family's pistol, we excuse him because his cognitive incapacity to understand the circumstances and consequences of his act prevents him from conforming his conduct to the appropriate legal and moral factors. As in cases of adult mistakes and accidents, his behavior does not express the disrespect for the victim's interest in life normally inferred from such actions. If a woman drives a getaway car for an armed bank robber and later takes a share of the booty, we infer that she shared the gunman's attitude regarding the interests of the robbery victims. If the same woman is forced to drive the robber away because the robber is holding a gun to her head, however, we do not infer from her conduct that she shared the robber's attitude toward the interests of the robbery victims. Instead, we excuse her on the ground of duress. In this case, the driver lacks the volitional capacity to control her conduct so as to conform to relevant moral and legal factors. Her conduct, therefore, does not express disrespect for the victims' interests. n140

[*1405] This analysis of excuses also applies to the insanity defense. n141 We excuse a defendant by reason of insanity, if, at the time he engaged in criminal conduct, [*1406] he lacked the cognitive or volitional capacity to conform his conduct to the appropriate relevant moral and legal factors. We excuse him because his cognitive or volitional incapacity leads us not to interpret his actions as expressing disrespect for legally protected interests. n142 Certainly many mentally ill people have the capacity to express respect and disrespect for the legally protected interests [*1407] of other people and the community. Moreover, many mentally ill people who lack the capacity to express respect nonetheless have sufficient capacity to convey meaning other than respect and disrespect through their conduct. To use the Model Penal Code's extreme example, a man who strangles his wife while believing sincerely that he is squeezing lemons n143 does not convey disrespect for her interest in life, but he might express some meaning through his conduct -- at the very least his desire to obtain lemon juice. Because the meaning of conduct relevant to criminal responsibility is expression of respect or disrespect for legally protected interests, we excuse only those mentally ill offenders who, as a result of cognitive or volitional incapacity, cannot convey disrespect of those protected interests. We do not excuse any other type of mentally ill offenders. n144

The following cases illustrate the interpretive theory of the exculpatory significance of insanity. First, consider the case of Joy Baker, described by Bonnie:

According to [Mrs. Baker's] account -- which no one has ever doubted -- she became increasingly agitated and fearful during the days before the shooting; she was worried that her dogs, her children (ages eight and 11), and her neighbors were becoming possessed by the devil and that she was going to be "annihilated." On the morning of the shooting, after a sleepless night, she ran frantically around the house clutching a gun to her breast. Worried about what the children might do to her if they became demonically "possessed" and about what she might do to them to defend herself, she made them read and reread the 23rd Psalm. Suddenly her aunt arived unexpectedly. Unable to open the locked front door, and ignoring Mrs. Baker's frantic pleas to go [*1408] away, the aunt came to the back door. When she reached through the broken screening to unlock the door, Mrs. Baker shot her.

The aunt then fell backward into the mud behind the porch, bleeding profusely. "Why, Joy?" she asked. "Because you're the devil, and you came to hurt me," Joy answered. Her aunt said, "Honey, no, I came to help you." At this point, Mrs. Baker said, she became very confused and "I took the gun and shot her again just to relieve the pain she was having because she said she was hurt."

All the psychiatrists who examined Mrs. Baker concluded that she was acutely psychotic at the time she killed her aunt. The police who arrested her and others in the small rural community agreed that she must have been crazy because there was no rational explanation for her conduct. She was acquitted. n145

This summary of the evidence indicates that Baker's sensory perception probably was intact and that she had considerable understanding of her physical and even social circumstances. Nevertheless, she also suffered from serious cognitive impairment in the forms of delusions and defective practical reasoning. For example, she had paranoid delusions regarding her children and aunt. She reasoned in bizarre fashion that she could achieve the end of protecting herself from her children through the means of forcing them to read the Twenty-third Psalm repeatedly. Similarly, she reasoned that she could achieve the end of relieving her aunt's pain from the first shot by shooting her a second time. Baker's delusions and impaired reasoning prevented her from guiding her conduct according to the many legal and moral factors -- from love to fear of punishment -- that normal people take into account when acting to avoid injuring interests such as another's interest in life. Because Baker's cognitive incapacity prevented her from acting with such concern, her conduct in shooting her aunt does not express to us -- that is, we do not interpret -- disrespect for her aunt's interest in life. A person such as Baker, who cannot act with care regarding legally protected interests because of mental illness, does not express or communicate to us an attitude of disrespect for those interests when she injures them.

As a second example, consider a similar case of mental illness, described by Livermore and Meehl:

A 35-year-old-woman without previous history of mental illness develops a depressive episode of psychotic proportion with the usual symptoms of extreme sadness of mood, weight loss, sleep disturbances, loss of interest in her usual activities, preoccupation with negative thoughts of guilt and hopelessness of the future, and psychomotor retardation. During the deepest phase of the depression, the psychomotor retardation is such that she is massively inhibited from doing almost anything, including anything dangerous to herself or others. As the depression begins to lift, however, the psychomotor retardation is somewhat reduced, permitting her to make a few decisions and take a few actions, [*1409] and since her affective state is still one of very black mood and her thinking is still along lines of hopelessness, she kills both of her children and makes an unsuccessful suicide attempt. On that day, she kept the children home from school, called up the cleaning woman and told her not to come as scheduled, and in various other ways showed clear evidence of systematic planning and premeditation of the killings. Emerging from the depression, she is at the time of trial essentially "normal" in all respects, and has a recollection although a somewhat dim one both of the events that transpired and of her mental state at the time. Insofar as she can recall how she felt and what she thought, her recollection is in accordance with the statements she made to police officers at the time, namely that while God did not command her to do anything and no voices were heard, it seemed perfectly clear to her that the situation was hopeless and that the killing was the only right thing to do because of the terrible state of the world and the terrible kind of mother she had been and would no doubt continue to be. The question as to whether she was doing something forbidden by law did not enter her consciousness in any form, although obviously she recognized the necessity of performing the act in some sense surreptitiously, when her husband and the cleaning lady and so on were not around. The evidence indicated, and the psychiatric testimony was to the effect that the suicide attempt was genuine and miscarried quite inadvertently. n146

As in the case of Joy Baker, this psychotic parent's sensory perception appears to have been intact and she had considerable understanding of her physical and social circumstances. In contrast to Baker, the parent her acted after careful planning, indicating at least sufficient practical reasoning to select effective means to attain the end of killing her children. As Livermore and Meehl point out, her efforts to perform the killings secretly suggest that she understood the illegality of her conduct. n147 Like Baker, however, this parent suffered from substantial cognitive impairment in the form of delusions and faulty practical reasoning. She had delusions about both the world and herself. She also reasoned inappropriately that the means of killing her children and herself would attain the end of saving them all from the horror of human life. The mother's delusions and defective practical reasoning prevented her from guiding her behavior according to the moral and legal factors that normally deter people from injuring the legally protected interests involved. In this case, the effect of the parent's cognitive incapacity was particularly tragic, since it twisted her evident love for her children -- normally a factor that deters people from harming their children -- into a decision to kill them. Because this cognitive incapacity prevented the mother from acting in accord with the moral and legal factors that can shape [*1410] respectful conduct, we do not interpret in her acts a sense of disrespect for her children's interest in life.

As a third example of cognitive incapacity, consider Hadfield's Case, discussed earlier. n148 The report of the case indicates that Hadfield did not suffer from perceptual disability and that he understood the physical and social circumstances of his act. As in the case of the psychotic parent, Hadfield's careful planning demonstrated considerable capacity for practical reasoning: to attain the end of his own execution, he chose the ordinarily effective method of assassinating the king. To attain the end of assassinating the king, he carefully planned the appropriate means of attending, while armed, the Drury Lane theater for a performance her knew the king was scheduled to attend. Furthermore, like the depressive parent, Hadfield understood the illegality of his conduct. Indeed, he chose his course of conduct precisely because of its serious, unquestionable illegality: it would guarantee his desired goal of execution. n149 Like both Baker and the depressive parent, however, Hadfield also suffered from serious cognitive incapacity -- in his case, a delusion that his own martyrdom through execution was the necessary means to save humanity.

Hadfield's cognitive impairment prevented him from conforming his conduct to the moral and legal factors that can deter normal people from committing such desperate acts. In his case, the cognitive incapacity had a pathetic quality because Hadfield committed his blatantly immoral and illegal act for the lofty, moral purpose of saving humanity. Because Hadfield's cognitive disability blocked him from conforming his behavior to the appropriate legal and moral norms, we do not interpret his conduct as expressing disrespect for the king's legally protected interests. A person with intact cognitive and volitional capacities who assessinates or tries to assassinate a government leader for political or financial reasons plainly conveys disrespect for the leader's interest in life and the country's interest in stable leadership. n150 Hadfield's disordered conduct, however, did not convey those attitudes.

With regard to the exculpatory character of volitional incapacity, note first that in a shared sense, the three examples of cognitive incapacity just discussed can also be seen as involving volitional incapacity of the kind posited by the third category of pre-M'Naghten cases and commentary, discussed above. n151 That is, we could say the cognitive disability of the defendants prevented them from controlling conduct by conforming their actions to the relevant moral and legal factors. There are also cases, however, including perhaps the case of Joy Baker, in which volitional incapacity manifests itself more directly. While these cases might well involve cognitive impairment as well, n152 volitional disability stands [*1411] out as a noticeable feature of the defendant's conduct and experience.

As an example of cases involving volitional incapacity, consider People v. Gorshen. n153 Nicholas Gorshen, a Russian immigrant and a religious, family man, killed Red O'Leary, his longshoreman boss, after a quarrel at work. Gorshen had appeared at work after an unusual episode of drinking. O'Leary accused him of being too intoxicated to work and ordered him to leave. The two men argued and then fought, leaving Gorshen unconscious on a ship's deck. During the fight, O'Leary had shouted a Russian sexual obscenity at Gorshen. When Gorshen regained consciousness, he threatened in the presence of other workers that he would go home, get a gun, return to the dock, and shoot O'Leary. Indeed, Gorshen carried out his threat, shooting O'Leary to death despite the presence of police called to the dock in an attempt to thwart his plan.

According to defense psychiatrist Dr. Bernard Diamond, n154 Gorshen had suffered for years from occasional trances, lasting for a few minutes at a time, in which he had visual hallucinations of demons, strange animals, and bizarre and deformed people. He experienced visions of odd surgical experiments and mystical transformations of organs. He also experienced bizarre sexual temptations. Diamond wrote:

For the previous two years, Gorshen had been having progressive difficulty with his sexual potency and he was beginning to doubt his capacity to return from his world of visions and devils. Slowly his confidence in his masculinity, his confidence in his ability to withstand the perverse sexual temptations of his trances was lessening and becoming more dependent on his being able to work hard as a longshoreman. When the gang boss ordered him off the ship, it was the climax of his worst fears. In the dazed emotional confusion of the fight, the Russian obscenity was too much for him. He returned to consciousness with only one thought, one emotion: the overwhelming compulsion to kill O'Leary. Only through killing could he retain his sanity and avoid eternal incarceration in the hell of his visions. n155

In Diamond's view, Gorshen suffered from an uncontrollable compulsion to kill O'Leary in order to retain his slipping hold on his own sanity. n156 Accepting for the sake of argument Diamond's analysis of Gorshen's conduct, it appears that Gorshen usually had normal cognitive capacities with respect to sensory perception, legal and noral norms, and practical reasoning. Over time, however, his cognitive incapacities during the trances undermined his sense of masculinity and confidence in his ability to control his life. The emotional pressure eventually became so intense that O'Leary's conduct on the day of the shooting drove [*1412] Gorshen beyond the point of control and prevented him from guiding his conduct according to the appropriate moral and legal norms.

Although cognitive problems certainly played a major role in Gorshen's conduct, his feeling of an uncontrollable urge to kill O'Leary played a significant part in his ultimate actions. Gorshen's desperation about his situation created extreme duress. The emotional pressure to shoot O'Leary ultimately overwhelmed his ability to resist. To the extent we believe the pressure was inrresistble, we would not interpret Gorshen's conduct as expressing disrespect for O'Leary's interest in life.

A psychiatrist testifying in the case of Joy Baker gave a similar explanation of the relation between Baker's cognitive disabilities and her impulse to shoot her aunt:

When Mrs. Baker ordered the children to read the 23rd Psalm, she was trying to assure that they would not be the ones to attack her; in this sense, putting the Bible in their hands was an act of self-defense. When the victim drove up, Mrs. Baker did everything she could to prevent her aunt from coming in the house. However, once her aunt put her hand through the door, Mrs. Baker had no options left, psychologically speaking. She was in a state of extreme anxiety and was fearful of imminent attack. The impulse for protective action was, if you will, irresistible. n157

This testimony explains that Baker's cognitive impairments led to her fear of attack and exerted overwhelming pressure on her to defend herself against her children and her aunt. The emotional pressure was so intense that it prevented her from guiding her conduct according to usual legal and moral norms, including in this case Baker's apparent love for her aunt and children. If we believe that the pressure was irresistible, we interpret her conduct as a desperate act of self-defense, lacking the volitional character necessary to express disrespect for her aunt's interest in life. n158

The cases of Gorshen and Baker involved cognitive incapacities that undermined volitional capacity to control behavior. In order cases, however, the volitional incapacity appears to operate even more directly, not merely as a derivative of cognitive disorders. Consider, for example, the following account, written by an assailant years after his murder of a colleague:

On the Saturday night (or early morning) I suddenly awoke with what I can only describe as strong homicidal tendencies for which I could in no way account. I lay in bed for some time trying to overcome them, till at last giving way to a powerful natural want, I left may bed to attend to it, and found myself quite unable to return to my couch. After a time (under what seemed a strong compulsion) I lighted a candle and [*1413] put on my dressing-gown, and wandered about my bedroom trying to overcome "myself," going down on my knees praying to be delivered from such an awful tragedy. I managed to lay down on the outside of my bed, but found myself unable to stay there. On getting on to my feet, I found myself opposite my dressing-table, when, much against my will, I possessed myself of a razor, which I could not get rid of, try as I would; my fingers seemed to clutch it as though they were hermetically sealed. Even then I stayed about in my bedroom for some time, taking up my position in front of my chest of drawers. After staying there some time, I tried to get to my bed, in doing which I had to pass my lighted candle, when I felt compelled to take it up (though it was far from my intention), and with it left the room, trying all the time as I walked along the passage to return to my room, but it was not to be. I may add that Mrs. X. [the victim's wife] came out of her room and tried to get the weapon away from me, but I (unaccountably) held it out of her reach. She then retired and locked the door after her. I stood about for some time and tried to get down the stairs, which were just outside her door, but it was of no avail. Then trying the handle, I found it unlocked, with Mrs. X. standing close by. I tried to get her to take the razor away, but she seemed totally unable to do so. I then walked on to the hearthrug and stood there for some time.

All this time Mr. X. lay in bed, and neither spoke nor made any effort to leave his bed to throw me out of the room. I at length turned to leave the room, but instead of going straight out, I felt compelled to turn to the left, which with another turn brought me to the head of Mr. X.'s bed, where I stood for some time trying to get the mastery of myself, . . . but it was of no use, for at length I felt compelled to make use of the weapon on his throat. If felt sure he would continue lying there, but to my surprise he suddenly jumped up and sat edgeways on his pillow with his feet out of bed, and there I left him. n159

In this case, the assailant suffered from an uncontrollable urge to kill the victim, unaccompanied by any apparent cognitive disorder. If we once again accept for the sake of argument the claim that the impulse was irresistible, we would not interpret the assailant's conduct as conveying disrespect for the victim's interest in life. Instead, the assailant's experience that he had no choice, that he was powerless to resist his homicidal urge, precludes us from interpreting his conduct as expressing disrespect. n160

[*1414] In summary, the insanity defense arises when an offender's mental illness deprives him of the cognitive or volitional capacity required to adhere to relevant legal or moral norms. When he lacks such capacity, his criminal conduct does not express the disrespect for legally protected interests of another person or the community normally associated with the activity in question.

The interpretive theory can explain why commentators and courts have regarded the absence of free choice, and irrationality, as key features of the exculpatory character of insanity. Cognitive and volitional capacities to weight legally and morally relevant factors before acting are necessary to make an authentic choice of conforming to those factors. But the question of whether the conduct is chosen, by itself, is not important as an ultimate criterion of responsibility. [*1415] Choice is important because it invests an actor's conduct with meaning that can be interpreted by others. Freely chosen conduct that injures the legally protected interests of another person or the community expresses the actor's disrespect for those interests. Conduct that injures those interests, but is not freely chosen, does not express such disrespect.

Irrationality is a vital aspect of the exculpatory nature of insanity because rationality is an essential attribute of intelligible conduct, of behavior an observer, such as a jury, can interpret. Thus, rationality is not important standing alone, but rather as a condition for engaging in conduct that others can interpret. When a mentally ill person injures legally protected interests through irrational conduct, we excuse the individual because we do not interpret his conduct as expressing disrespect for those interests.

B. THE INTERPRETIVE THEORY AND MENTAL ELEMENTS OF CRIME

The interpretive analysis of the insanity defense and other excuses raises an exciting question: If the core of criminal responsibility is the disrespect shown by actions that injure the legally protected interests of others, and if excuses represent conduct that does not convey this disrespect, then how does the criminal law account for normal cases, where the conduct in question does convey such disrespect? n161 The answer, I propose, is that criminal law addresses meaning through the intricate system of mental elements known as mens rea and the voluntariness component of actus reus. I will discuss the application of the interpretive theory first to mens rea and then to the mental component of actus reus.

In developing this view of the mental elements of crime, it will be helpful at the outset to consider the historical context of the issue. Sayre traced the early development in England of the concept of mens rea as a general element of evil intent or moral blameworthiness required for felony liability. Through several centuries of common law development, the concept of mens rea proliferated into particularized states of mind peculiar to specific felonies. n162

The modern discussion of the definition and role of mens rea began with Oliver Wendell Holmes. Underlying Holmes' view of mens rea was the belief that criminal law has a utilitarian purpose of deterring undesirable conduct rather than the purpose of assessing moral blameworthiness. As part of his utilitarian perspective, Holmes argued in favor of an "external" or "objective" theory of criminal liability. As Hall has noted, Holmes used those concepts ambiguously. Sometimes he used them to signify that the criminal law employs external tests as evidence to infer actual mental states. n163 More often, however, Holmes used these terms fundamentally -- and controversially -- as normative terms to argue that criminal law requires each person to adhere to a specified standard of conduct, regardless of his actual "inner" or "subjective" state of mind. n164 The core of Holmes' theory was that criminal law imposes liability when a person either knows or should know the tendency of his acts to cause proscribed results.

[*1416] Holmes acknowledged that a defendant's actual state of mind could be relevant to criminal liability. For example, a defendant's actual knowledge of circumstances that would make a particular act dangerous, or his actual foresight of harm, would be factors in determining his liability. In Holmes' view, however, actual states of mind were relevant only as they contributed to a tendency of certain conduct to cause a prohibited harm. Holmes wrote: "The mens rea, or actual wickedness of the party, is wholly unnecessary, and all reference to the state of his consciousness is misleading if it means anything more than the circumstances in connection with which the tendency of his act is judged are the circumstances known to him." n165 In sum, Holmes contended that a defendant was liable if he engaged in proscribed conduct either with an actual criminal state of mind about the harmfulness of his behavior or if he should have understood the harmfulness of the conduct.

Holmes' objective theory, then, saw the role of mens rea as serving the criminal law's normative and utilitarian purpose of deterring conduct the community deems harmful. It is "objective" not in the natural scientific sense of positing an empirically verifiable truth, but by its contrast to a "subjective" theory that either focuses descriptively on a defendant's actual state of mind or allows each individual to establish and apply his own normative standards to his conduct. n166

In contrast to Holmes' objective theory, some commentators have advanced a subjective theory, or what Fletcher more helpfully terms a "descriptive" theory, of mens rea. n167 Descriptive theorists, such as Stephen, n168 Sayre, n169 Turner, n170 and Williams, n171 have conceived of mens rea simply as the mental element required by common law or statute as part of the definition of an offense. Under this approach, mens rea is a morally neutral concept that entails an inquiry into the actual state of mind of a particular defendant at the time he allegedly engaged in prohibited conduct. According to descriptive theorists, "there is no general concept of mens rea or culpability, but merely an array of diverse mentes reae or 'kinds of culpability.'" n172 This view rejects both Holmes' objective theory and the early common law view that a criminal proceeding is an inquiry into a defendant's moral guilt. Descriptive theorists instead conceive of a criminal proceeding as a nonmoral inquiry into whether the defendant's conduct and actual accompanying state of mind satisfy the designated criteria of the alleged offense. n173

Yet a third approach, exemplified by Hall and Kadish, regards mens rea as [*1417] both a normative and a subjective concept. As a subjective concept, it differs from Holmes' view by giving more weight to a defendant's actual state of mind and by focusing on the defendant's moral blameworthiness. As do pure descriptive theorists, these modern subjective/normative theorists acknowledged the diversity of mental states designated as elements of crimes. They part company with pure descriptive theorists, however, by maintaining that such states are parts of a common normative inquiry into a defendant's blameworthiness. For example, Hall has written: "[T]he fact that the mens rea of murder differs from that of robbery, e.g., is not inconsistent with the presence of a common, 'essential' characteristic, expressed in the voluntary doing of a morally wrong act . . . ." n174

This brief historical review establishes a context for understanding the interpretive theory of mens rea proposed here. This theory rests on the normative principle advanced earlier in the article: that the criminal law requires that each person respect legally protected interests by acting in a way that avoids injuring those interests. A person who injures such interests, but who lacks the cognitive or volitional capacity to show such respect, is not held criminally liable. A person who injures those same interests, however, while possessing sufficient cognitive and volitional capacities expresses his disrespect for those interests and is therefore criminally liable. n175

Under the interpretive theory, the various specific states of mind encompassed by the term mens rea, such as intent, knowledge, malice, recklessness, and negligence, represent the criminal law's rules for interpreting the defendant's attitude toward various protected interests. Through these rules, as incorporated into statutes, common law definitions of crimes, and jury instructions, the legislature and the courts guide the jury's deliberation about the meaning or attitude conveyed by the defendant's alleged conduct. The jury's inquiry is both descriptive and normative. It engages in a descriptive inquiry to determine whether the defendant's actual state of mind matches the state of mind required as an element of the offense. The jury, however, makes this descriptive inquiry for the normative purpose of interpreting the attitude conveyed by the defendant's conduct. When a jury finds that a designated mental element accompanied a defendant's wrongful conduct, its duty is to interpret that conduct as conveying the type of attitude condemned by the criminal law. Mens rea, then, is the mechanism through which the moral quality of a defendant's behavior becomes an integral concern of a criminal proceeding, by interpreting the defendant's attitude toward legally protected interests as expressed through his conduct. Thus, [*1418] the interpretive theory of mens rea represents a synthesis of the descriptive and normative theories.

While there are important differences among the diverse mental states recognized in common law and statutes, they represent only differences in degree of disrespect for legally protected interests. For example, negligence regarding circumstances or consequences of conduct involves a minimal, though sometimes quite harmful, disrespect, while specific intent to cause a prohibited consequence represents an extreme form of disrespect. These differences are reflected within a particular category of offense by linking the levels of grading and punishment to the degree of departure from the normative standard. For example, first- and second-degree murder and voluntary and involuntary manslaughter are the common law grades of homicide, distinguished by virtue of different states of mind, and hence involving differing degrees of disrespect for the victim's protected interest in life. n176

In addition, various categories of criminal offenses involve different forms of mens rea. For example, the mental elements of the several common law grades of homicide -- specific intent to kill, general intent, malice, and criminal negligence -- plainly differ from the mental element of common law larceny -- specific intent to deprive the victim of rightful possession of property. Why do different criminal offenses involve different forms or combinations of mens rea? While I cannot give a complete answer at this time, my preliminary response is that the constellation of mental elements peculiar to a given offense represents the legislatures' or courts' descriptive determination of the ways in which the legally protected interests involved are commonly violated. That is, they identify the conduct, circumstances, and consequences that tend to characterize violations of such interests. n177 Furthermore, the elements represent the legislatures' or courts' normative determination of the quality of respect appropriate to prevent injury to specific protected interests, and the states of mind necessary to express disrespect for those interests. Again, this is only a preliminary answer and requires further development -- beyond the scope of this article -- exploring the nature of the interests protected by the criminal law.

Within this elaborate interpretive framework the mens rea elements of criminal offenses might be termed first-order rules of interpretation. They are the jury's hermeneutic tools for interpreting the relevant meaning or attitude of the [*1419] raw "text of the defendant's conduct. n178 The law also equips the jury with [*1420] what can be called second-order rules of interpretation. These are the familiar inferences regarding mental elements that the jury learns through jury instructions. For example, given the frequent difficulty of gauging a defendant's intent, courts often instruct juries that they may infer that a person intends the ordinary consequences of his voluntary conduct. The inference serves as a second-order interpretive rule to guide the jury's assessment of the defendant's intent, which, in turn, serves as a first-order interpretive rule to guide the jury's assessment of the defendant's attitude toward the protected interests involved. n179 As another illustration, consider the standard inference in jury instructions about malice in homicide cases: the jury may infer malice from the intentional infliction of a wound with a deadly weapon when there is no evidence of mitigation, justification, or excuse. This inference can help the jury resolve the potentially difficult issue that forms the core of malice -- whether the defendant acted with a conscious or deliberate disrespect for the victim's interest in avoiding great bodily injury or death. n180

[*1421] We can apply the same analysis used to examine means rea to the voluntariness component of actus reus. The full dimension of the mental elements of crime extends beyond the scope of traditional mens res elements to include the issue of voluntariness commonly regarded as falling within the concept of actus reus. According to traditional criminal theory, human activity is subject to criminal sanction only if it consists of a voluntary act or omission. In other words, voluntariness is a necessary condition for conduct to be considered a genuine human act. n181 An act is generally considered voluntary if it is "an exertion of the will [*1422] manifested in the external world," n182 or if it is "a [willed] muscular contraction." n183 It involves both the will to engage in a particular physical movement -- regardless of the purpose or motive of that movement -- and consciousness that one is so acting -- regardless of awareness of the circumstances of such action. n184 As H.L.A. Hart has written, it means conscious control over muscular movements. n185 Involuntary acts are those committed without such will or consciousness, such as reflexes, convulsions, bodily movement during unconsciousness or sleep, and conduct during hypnosis or resulting from hypnotic suggestion. n186

Voluntariness is one criterion for the meaning or expressive significance of criminal conduct. It is a facet of the quality of respect expressed by an actor through his conduct. For example, if a person voluntarily pulls the trigger of a gun pointed at another person, the voluntary nature of his conduct contributes to the actor's expression of an attitude toward the victim's interest in life. When coupled with mens rea factors such as the actor's knowledge that the gun is loaded and either his knowledge that the gun is pointed at the victim or his intent to point it at the victim, his voluntary pulling of the trigger expresses the actor's disrespect for the victim's interest in life. Nevertheless, if the actor acts without either the requisite physical control over his movement or consciousness of his movement, then his conduct does not convey such meaning. For example, if one person physically forces another person's finger to squeeze a trigger, or if an actor shoots another person during an episode of sleepwalking, we would not interpret the triggerman's act as expressing disrespect for the victim's interest in life. In such a case we would acquit the actor either by failing to find the essential element of actus reus or by applying an "involuntary act" defense. n187

To summarize, the various mental states categorized under the umbrella of mens rea and the voluntariness component of actus reus function as guides to the jury in its role of interpreting the meaning of a defendant's conduct. When those states of mind that constitute essential mental elements are missing from a defendant's conduct, as in cases of involuntary action, mistake, and accident, then [*1423] his conduct does not convey disrespect toward protected interests and the jury should acquit him. When all essential mental elements are present, but they still do not accurately guide the jury's interpretation of the defendant's conduct, as in cases of duress, infancy, and insanity, then once again the defendant's conduct does not convey disrespect toward the protected interests and the jury should acquit him. n188

[*1424] C. IMPLICATIONS OF THE INTERPRETIVE THEORY

As suggested at the beginning of this article, our inquiry into the rationale of the insanity defense and other excuses has revealed the interpretive function of excuses in criminal law and has pointed beyond that function to identify the interpretive role of the mental elements of crime. The interpretive theory also can provide insight into other important issues in substantive criminal theory. This concluding part addresses implications of the theory for five issues: the irrelevance of a defendant's good motive to his criminal liability; whether negligence should be a ground for criminal liability; whether strict liability should be a ground for criminal liability; the purposes of punishment and treatment; and alternative approaches to mentally disordered criminal conduct.

1. Irrelevance of Good Motive

Turning first to the issue of motive, note the general rule that a defendant's allegedly good motive for committing a criminal act does not constitute a defense to criminal liability. n189 To state the point more fully, a defendant's motive can be relevant as circumstantial evidence on the question of whether he committed the crime of which he stands accused. His motive is not, however, an additional element of an offense. Once the prosecution establishes that the defendant voluntarily committed a prohibited act with a culpable mental state, the prosecution has established a prima facie case of guilt, even without showing the defendant's motive. Finally, after the prosecution establishes the essential elements of actus reus and mens rea, proof of the defendant's good motive will not function as a defense. n190 Why is it that a defendant's good motive does not serve as a defense?

Consider the example of a man who commits an armed robbery of a bank to obtain money to finance a charitable program. Without question, his motive for the robbery is laudable. Nevertheless, to obtain the money to achieve his worthy goal, out latter-day Robin Hood injures property interests and threatens the interest in life of the bank staff and customers in the bank at the time of the robbery. Thus, even though his ultimate purpose is good, he expresses the same disrespect for legally protected interests as a robber who acts out of gardenvariety greed. Since the law condemns such expression, regardless of motive, our philanthropic robber is as guilty as a greedy robber. n191 A judge might take the [*1425] differences in motive into account in sentencing, but they are equally guilty.

2. Negligence as a Basis of Criminal Liability

Turning to the issue of negligence, we see that commentators have debated at length whether negligence is an appropriate basis of criminal liability. n192 The interpretive theory of mens rea shows that negligence and the conscious forms of mens rea, purpose and knowledge, rest on a common foundation. Consider, for example, the case of a man who commits a homicide through negligent conduct, such as pulling the trigger of a loaded gun pointed at another person while believing it is unloaded and without first checking its condition. At the time of his act, the defendant is unaware of the risk created by his conduct. If he could have and should have been aware of the risk, however, he is liable for involuntary manslaughter.

As H.L.A. Hart has argued, it is fair to hold this defendant criminally liable because he failed to exercise his capacity and opportunity to avoid creating the risk. n193 The interpretive theory of mens rea supports Hart's position. It is appropriate to hold the negligent defendant criminally liable because by creating or taking a risk he could have avoided, or by making an unreasonable mistake, he has expressed disrespect for the victim's interest in life. n194 Although this defendant's negligence represents a less serious degree of disrespect than an intentional killing, both negligent and intentional homicides express disrespect. The difference is a matter of the strength of the meaning expressed by such conduct. n195

[*1426] 3. Strict Criminal Liability

The question of criminal liability based on strict liability, with no proof of mens rea required, has also generated great controversy. Favoring strict liability, Lady Wootton has advocated limiting the function of a criminal trial to a determination of whether a defendant committed the prohibited physical conduct, shifting all consideration of culpability to a posttrial disposition proceeding. n196 In opposition, other commentators have argued that a strict liabiity offense unfairly imposes criminal liability on a defendant for conduct beyond his control. n197 Courts and legislatures have imposed strict liability in three types of cases: (1) public regulatory offenses; (2) a small number of basic offenses (usually moral or sexual offenses such as bigamy and statutory rape, although felony murder is also in this category); and (3) cases in which ignorance of the law does not function as a defense. n198

Under the interpretive theory, criminal liability without fault is not appropriate. As shown above, when a person acts without the capacity to conform his conduct to legal and moral factors, his conduct does not express disrespect for legally protected interests. Indeed, a court might find a defendant guilty of a strict liability offense despite making heroic efforts to avoid injuring the interests protected by the offense. n199 By eliminating consideration of a defendant's mental state from the determination of guilt, strict liability offenses eliminate the relevance of a feature of criminal conduct that contributes to its criminal character -- the meaning this conduct conveys. n200

[*1427] 4. Purposes of Punishment and Treatment

We have seen that the inquiry into the meaning of human conduct can explain the role of excuses and mental elements of crime in characterizing acts as criminal. It can also provide insight into reasons for punishing people who commit criminal acts. Major goals traditionally attributed to punishment include retribution, general deterrence, specific deterrence, incapacitation, and rehabilitation. n201 The role of expression, as interpreted by the jury, in characterizing conduct as criminal points to yet another value of punishment: as a meaningful response to the expressive content of a criminal's behavior. Just as an offender conveys meaning to his victim or to the community through his conduct, that is, through his disrespect for protected interests, so the community -- through its agents the sentencing judge and corrections system -- responds by conveying meaning through the vehicle of punishment. Punishment can convey either a limited message or a morally rich message. Functioning as a limited signal, it conveys the message that people should avoid engaging in certain conduct if they want to avoid painful consequences to themselves. Philosopher Jean Hampton has likened this limited deterrent function to the purpose served by an electrified fence in a cow pasture. A cow that runs up against the fence "experiences pain and is conditioned, after a series of encounters with the fence, to stay away from it and thus remain in the pasture. A human being in the same pasture will get the same message and learn the same lesson -- 'if you want to avoid pain, don't try to transgress the boundary marked by this fence.'" n202

Functioning as a richer signal, punishment can also tap our human capacity for communicating to an offender not only that conduct is wrong and will result in painful consequences for the offender, but also why it is wrong. As Hampton explains:

[U]nlike the animal in the pasture, a human being will also be able to reflect on the reasons for that fence's being there, to theorize about why there is this barrier to his freedom.

Punishments are like electrified fences. At the very least they teach a person, via pain, that there is a "barrier" to the action she wants to do, and so, at the very least, they aim to deter. But because punishment "fences" are marking moral boundaries, the pain which these "fences" administer (or threaten to administer) conveys a larger message to beings who are able to reflect on the reasons for these barriers' existence: they convey that there is a barrier to these actions because they are morally wrong. n203

The moral message conveyed by punishment responds to the message conveyed by the offender. The offender, by committing a crime, has indicated his disrespect for the protected interests of the victim or the community. n204 Society [*1428] responds, through punishment, with a message that the offender has not acted with the respect necessary to avoid injuring protected interests. As Hampton writes:

Consider the fact that someone who (for no moral reason) violates a positive duty to others is not acting out of any interest in the others' well-being. A teenager who steals from a passer-by because she needs the money, a man who rapes a woman so he can experience a sense of power and mastery -- such people are performing immoral acts in order to satisfy their own needs and interests, insensitive to the needs and interests of the people they hurt. The way to communicate to such people that there is a barrier of a very special sort against these kinds of actions would seem to be to link performance of the actions with what such people care about most -- the pursuit of their own pleasure. Only when disruption of that pursuit takes place will a wrongdoer appreciate the special force of the "mustn't" in the punisher's communication. n205

Regardless of one's position on whether punishment is a necessary way to communicate this message, Hampton at least highlights the communicative role of punishment. In this role, then, punishment is as an expressive response by the community to an offender's expressive conduct as interpreted by the community. The jury serves as the community's interpretive agent, and the sentencing judge and corrections system serve as the community's expressive agents. n206 In short, crime and punishment are both forms of communication. n207

It is important to note that the expressive function of punishment promotes the goals of other functions of punishment. It expresses the collective anger at issue in retribution. Consistent with both general and specific deterrence, it seeks to prevent future criminal acts by the criminal and others -- not only by the simple deterrent of threatening people with the "electrified fence" of time in prison, but also through normative communication. n208 Finally, it seeks to rehabilitate [*1429] offenders by helping them appreciate the reasons their conduct was wrongful.

By viewing punishment as a form of normative communication, we can understand why we do not punish mentally ill offenders or those to whom other excuses apply. We punish offenders who express their disrespect by ignoring legal and moral factors that should guide their conduct. We do not, however, punish people who lack the capacity to guide their behavior in that way. As a result of their incapacity to so guide their conduct, mentally ill defendants do not express the disrespect that prompts us to punish. Accordingly, we do not punish them. We excuse them instead. In the case of people we excuse by reason of insanity, we respond in an additional way: we generally commit them to mental institutions to treat them with the goal of enhancing their capacity to guide their conduct according to relevant legal and moral factors. n209

[*1430] 5. Alternative Approaches to Mentally Disordered Criminal Conduct

In addressing the impact of mental disorder on criminal conduct, courts and commentators have recognized a variety of alternatives to the insanity defense. These approaches include those designed to expand the relevance of mental disorder to criminal responsibility, and others fashioned to eliminate or narrow it. An expansive strategy preserves the insanity defense as an exculpatory defense and, in addition, recognizes mental disorder as a partial or mitigating defense that reduces the degree of a defendant's culpability in some cases. The two major types of mitigating defenses for mentally ill offenders are the diminished capacity and diminished responsibility defenses.

The diminished capacity defense operates in a fashion similar to the partial defense of intoxication. Using this defense, a defendant can introduce evidence to show that he lacked the capacity to form a mental state essential to the definition of the alleged offense. Diminished capacity, then, is a negativing defense designed to rebut prosecution evidence on an essential mental element of a crime. n210 Put more simply, diminished capacity, like the defenses of mistake and accident, is more a rule of logical relevance than a defense: if psychiatric evidence can show that the defendant lacked an essential state of mind, it is as relevant as nonpsychiatric evidence on that issue. n211 Theoretically, a defendant may use the diminished capacity defense to negate the mental state required for any degree of crime, thereby resulting in outright acquittal in some cases. In practice, however, courts generally have limited use of the defense to reduce the defendant's culpability to a lower grade of offense. n212 Under the diminished responsibility defense, evidence of mental disorder is admissible to show that while a particular defendant's mental illness might not entitle him to an acquittal by reason of insanity and might not negate an essential mental state, it might nevertheless reduce his degree of responsibility, resulting in conviction of a lower grade of offense. n213

[*1431] The major argument in favor of recognizing the defense of diminished capacity is that fundamental fairness should allow a defendant to introduce relevant evidence concerning an essential element of his alleged crime. n214 The chief argument in support of the diminished responsibility defense is that it permits juries to make more finely tuned judgments of culpability. n215 Opponents of the defenses have argued, however, that psychiatric evidence about a defendant's capacity to form specific mental states or about his degree of culpability is unreliable n216 and that these defenses jeopardize the criminal law's maintenance of social control. n217

Under the interpretive theory, both types of mitigating defenses are appropriate as a matter of theory. As indicated above in the discussion of grading of offenses according to different degrees of mens rea, n218 the mental state with which an act is committed affects the meaning conveyed by that act. Careful examination of conduct may reveal that it does not convey the degree of meaning normally conveyed by such conduct. This may arise because the defendant could not form a designated mental state -- a case of diminished capacity. Alternatively, it may arise because mental illness reduced the defendant's cognitive or volitional capacity to guide his conduct according to relevant moral and legal factors -- a case of diminished responsibility. In either situation, it is theoretically appropriate to find that the defendant was less culpable than other people committing similar acts. In other words, the meaning conveyed by human conduct is rarely an all-or-nothing matter; gradations in conveying such meaning should theoretically result in gradations in culpability. n219

At the other end of the spectrum of alternative approaches to the relationship between mental illness and criminal responsibility is the proposal to abolish the insanity defense. Abolition could come in two forms. The more extreme form, advocated by Lady Wootton and discussed above, n220 entails removing the issue of a defendant's mental state -- mental elements as well as insanity -- entirely [*1432] from the inquiry into guilt. As explained earlier, n221 this strict liability version of abolition would not be appropriate under the interpretive theory because it ignores a pivotal component of guilt: the meaning conveyed by conduct.

A more limited abolitionist strategy would involve elimination of the insanity defense while permitting a defendant to present psychiatric evidence to negate an essential mental element of his alleged offense. n222 Stated differently, this proposal calls for replacing the insanity defense with a diminished capacity defense. Opponents of this approach have pointed out that it would arbitrarily exclude many mentally ill defendants from a mental disorder defense. This system would convict defendants, such as Hadfield, n223 who entertain the requisite "descriptive" state of mind for guilt, but who nevertheless should be excused. n224 As the American Bar Association stated in a recent report, assessing criminal blameworthiness entails not only an inquiry into mental elements alone, but also an inquiry into the "quality of knowledge and intent, going beyond a minimal awareness and purposefulness." n225

The interpretive theory supports those opposing the limited abolitionist strategy. Under the interpretive theory, the limited, "elements" approach to abolition would be inappropriate because it fails to recognize the interpretive role served by mens rea. As explained above, n226 required mental elements function as interpretive rules for the jury, guiding its assessment of the meaning of a defendant's conduct. As shown by cases such as Hadfield's Case, though, these interpretive rules are not always reliable. The insanity defense is necessary to address cases in which a defendant's conduct lacks, as a result of mental illness, the requisite meaning for guilt despite the presence of the requisite mental elements. The limited abolitionist approach ignores the value of the insanity defense as an occasionally necessary exception to the criminal law's general interpretive rules. n227

[*1433] Still another approach to the relationship between mental disorder and criminal responsibility is to permit a jury to render a verdict of "guilty but mentally ill." In some jurisdictions, this verdict is available as a fourth alternative to the verdicts of guilty, not guilty, and not guilty by reason of insanity. n228 In a few jurisdictions, it replaces the verdict of not guilty by reason of insanity. n229 As a fourth alternative, it is appropriate under the interpretive theory as essentially a form of diminished responsibility verdict. n230 As a substitute for the insanity defense, however, it is inappropriate as a variant of the limited abolitionist position criticized above. n231

IV. CONCLUSION

Over the course of centuries, the insanity defense has remained one of the most controversial issues in Anglo-American substantive criminal law. This article has explored the theoretical foundation of the insanity defense in terms of the meaning conveyed by conduct. The insanity defense arises when a defendant, due to mental illness, lacks the cognitive or volitional capacity to guide his conduct according to moral and legal factors so as to avoid injuring legally protected interests of another person or the community. It is this lack of capacity that prompts us, through the jury as our representative, to not interpret in the defendant's conduct a sense of disrespect for those interests. Pursuing this line of inquiry further, the mental elements of crime are the law's mechanisms for interpreting the significance of disrespect expressed by the conduct of normal, or nondisordered, offenders. Taken together, mental elements and excuses, including the insanity defense, operate as the law's intricate system for interpreting the quality of respect or disrespect expressed by the conduct of defendants. n232 Perhaps [*1434] the deep connection between the insanity defense and the mental elements of offenses helps explain why the defense has endured despite the controversy surrounding it: the quality of respect or disrespect for legally protected interests expressed by a defendant's conduct lies at the heart of criminal responsibility, and the insanity defense serves as an important tool for the jury's task of normative, interpretive inquiry into such respect or disrespect. Crime is a form of communication, and the system of mental elements and excuse defenses, including the insanity defense, serves as the law's body of rules for interpreting the message conveyed by criminal conduct.

FOOTNOTES:
n1 Austin, A Plea for Excuses, 57 ARISTOTELIAN SOC'Y PROC. 1, 5-6 (1956-1957), reprinted in FREEDOM AND RESPONSIBILITY 6, 8 (H. Morris ed. 1961). For analyses of how examination of the abnormal can contribute to our understanding of the normal, see M. HEIDEGGER, BEING AND TIME 102-07 (J. Macquarrie & E. Robinson trans. 1962); R. PALMER, HERMENEUTICS 132-37 (1969).

n2 E.g., United States v. Brawner, 471 F.2d 969, 985-86, 1002 (D.C. Cir. 1972) (en banc); Durham v. United States, 214 F.2d 862, 876 (D.C. Cir. 1954), overruled, United States v. Brawner, 471 F.2d 969, 991 (D.C. Cir. 1972); People v. Wolff, 61 Cal. 2d 795, 814, 394 P.2d 959, 971, 40 Cal. Rptr. 271, 283 (1964) (en banc) (superseded by statute as stated in People v. Sanders, 154 Cal. App. 3d 487, 201 Cal. Rptr. 411 (1984)) (opinion withdrawn by order of court); State v. Sikora, 44 N.J. 453, 470, 210 A.2d 193, 202 (1965); S. GLUECK, LAW AND PSYCHIATRY 5-19 (1962); H. GROSS, A THEORY OF CRIMINAL JUSTICE 306 (1949); H.L.A. HART, PUNISHMENT AND RESPONSIBILITY 182-83 (1968); D. HERMANN, THE INSANITY DEFENSE 74-75 (1983); P. LOW, J. JEFFRIES & R. BONNIE, CRIMINAL LAW 692-93 (1982); H. PACKER, THE LIMITS OF THE CRIMINAL SANCTION 132-34 (1968); Bonnie, The Moral Basis of the Insanity Defense, 69 A.B.A. J. 194, 195 (1983); Bonnie & Slobogin, The Role of Mental Health Professionals in the Criminal Process: The Case for Informed Speculation, 66 VA. L. REV. 427, 448-49 (1980); Hall, Psychiatry and Criminal Responsibility, 65 YALE L.J. 761, 765 (1956); Kadish, The Decline of Innocence, 26 CAMBRIDGE L.J. 273, 273-75, 287 (1968).

n3 Brawner, 471 F.2d at 979-81; MODEL PENAL CODE § 4.01 commentary at 156-57 (1955); S. GLUECK, supra note 2, at 70-71; S. GLUECK, MENTAL DISORDER AND THE CRIMINAL LAW 97-98 (1925); A. GOLDSTEIN, THE INSANITY DEFENSE 45-46, 75-76 (1967); D. HERMANN, supra note 2, at 74-75; 2 J. STEPHEN, A HISTORY OF THE CRIMINAL LAW OF ENGLAND 183 (1883); Bonnie, supra note 2, at 195-96.

n4 10 Cl. & Fin. 200, 8 Eng. Rep. 718 (1843).

n5 For further discussion of excuses in the ancient world, see D. HERMANN, supra note 2, at 18-20.

n6 ARISTOTLE, ETHICA NICOMACHEA 1109b (W. Ross trans. 1925). In words that span the centuries to speak clearly to observers of criminal law today, Aristotle wrote:

Since virtue is concerned with passions and actions, and on voluntary passions and actions praise and blame are bestowed, on those that are involuntary pardon, and sometimes also pity, to distinguish the voluntary and the involuntary is presumably necessary for those who are studying the nature of virtue, and useful also for legislators with a view to the assigning both of honours and of punishments.

Id.

n7 Id. at 1109b-10a.

n8 For further discussions of the early English historical development of the insanity defense, see J. BIGGS, THE GUILTY MIND 81-88 (1955); J. HALL, GENERAL PRINCIPLES OF CRIMINAL LAW 475-77 (2d ed. 1960); D. HERMANN, supra note 2, at 21-31; R. PERKINS & R. BOYCE, CRIMINAL LAW 950-51 (3d ed. 1982); 1 N. WALKER, CRIME AND INSANITY IN ENGLAND 15-73 (1968); H. WEIHOFEN, INSANITY AS A DEFENSE IN CRIMINAL LAW 17-21 (1933); Crotty, The History of Insanity as a Defence to Crime in English Criminal Law, 12 CALIF. L. REV. 105 (1924); Sayre, Mens Rea, 45 HARV. L. REV. 974, 1004-07 (1932).

n9 Fitzherbert wrote:

He who is of unsound Memory, hath not any manner of Discretion; for if he kill a Man, it shall not be a Felony, not Murder, nor he shall not forfeit [sic] his Lands or Goods for the same, because it appeareth that he hath not Discretion; for if he had Discretion, he should be hanged for the same, as an Infant who is of the Age of Discretion, who committeth Murder or Felony, shall be hanged for the same.

A. FITZHERBERT, THE NEW NATURA BREVIUM 466 (8th ed. 1755), quoted in H. WEIHOFEN, supra note 8, at 18 n.15.

n10 In his First Institute, Coke described four types of people as non compos mentis:

Non compos mentis is of four sorts; 1. Ideota, which from his nativitie, by a perpertuall infirmatie, is non compos mentis. 2. Hee that by sickness, griefe, or other accident, wholly loseth his memorie and understanding. 3. A lunatic that hath sometime his understanding and sometime not, aliquando gaudel lucidis intervallis, and therefore he is called non compos mentis, so long as he hath not understanding. Lastly, hee that by his owne vitious act for a time depriveth himselfe of his memorie and understanding, as he that is drucken.

1 E. COKE, INSTITUTES OF THE LAWS OF ENGLAND *247a (R. Small ed. 1853), quoted in J. BIGGS, supra note 8, at 84-85. Although the quoted passage refers only to civil matters, Coke subsequently applied these principles to criminal cases:

[I]n criminal causes, as felony, the act and wrong of a madman shall not be imputed to him, for that in those causes, actus non facit reum, nisi mens sit rea, and he is amens (id est) sine mente, without his mind or discretion; and furioso solo furore punitur, a madman is only punished by his madnesse. And so it is of an infant, untill he be of the age of fourteene, which in law is accounted the age of discretion.

1 E. COKE, supra, at *247a-47b, quoted in J. BIGGS, supra note 8, at 85.

In his Third Institute, Coke wrote that a person who is "non compos mentis and totally deprived of all compassings and imaginations, cannot commit high treason by compassing or imagining the death of the king . . . but it must be an absolute madnesse, and a totall deprivation of memory." 3 E. COKE, supra, at *6. Regarding suicide, Coke wrote, "If a man lose his memory by the rage of sicknesse or infirmity, or otherwise, and kill himself while he is non compos mentis, he is not Felo de se: for as he cannot commit murder upon another, so in that case he cannot commit murder upon himself." Id. at *54.

n11 For example, Bracton wrote, "Remove will and every act will be indifferent. It is your intent that differentiates your acts, nor is a crime committed unless an intention to injure exists; it is will and purpose which distinguish maleficia." 2 H. BRACTON, ON THE LAWS AND CUSTOMS OF ENGLAND 23 (S. Thorne trans. 1968). Bracton further wrote: "[Thus] we must consider intention and purpose, as well as what is done or said, in order to ascertain what action follows and what punishment. For remove intention and every act will be indifferent; it is your intent which distinguishes your acts . . . ." Id. at 289-90. With respect to the interpretive theory presented infra, it is interesting to note that the clause, "it is your intent which distinguishes your acts," was translated by Sayre as "your state of mind gives meaning to your act." Sayre, supra note 8, at 985 (emphasis added). Bracton also wrote, "will and purpose are the distinctive marks of crimes." 2 H. BRACTON, supra, at 375. Finally, Bracton wrote with reference to theft: "It is will and purpose which mark maleficia, nor is a theft committed unless there is an attempt to steal." Id. at 384.

n12 Bracton's full statement with respect to the exculpatory significance of insanity is

But what shall we say of a madman bereft of reason? And of the deranged, the delirious, and the mentally retarded? or if one labouring under a high fever drowns himself or kills himself? Quaere whether such a one commits felony de se. It is submitted that he does not, nor do such persons forfeit their inheritance or their chattels, since they are without sense and reason and can no more commit an injuria or a felony than a brute animal, since they are not far removed from brutes, as is evident in the case of a minor, for if he should kill another while under age he would not suffer judgment. That a madman is not liable is true, unless he acts under pretense of madness while enjoying lucid intervals.

2 H. BRACTON, supra note 11, at 424 (footnotes omitted). It is interesting to note that Bracton recognized the possibility of temporary insanity. See 1 N. WALKER, supra note 8, at 29.

Walker has argued that Bracton, by referring to brute animals, probably intended to draw an analogy between the cognitive capacities of insane people and the cognitive capacities of animals, rather than to suggest that insane people act in a frenzied manner. Id. at 28-29. Bracton's comparison of an insane person with a "brute animal," however, cited by both Coke and Hale, was later lifted from its context as a mere illustration and converted into the "wild beast" test of insanity. Platt & Diamond, The Origins and Development of the "Wild Beast" Concept of Mental Illness and its Relation to Theories of Criminal Responsibility, 1 J. HIST. BEHAVORIAL SCI. 355, 359 (1965). Platt and Diamond have explained that it was Judge Tracy, in his jury instructions in Rex v. Arnold, 16 How. State Tr. 695 (Eng. 1724), who played the pivotal, albeit perhaps unintended, role in transforming Bracton's descriptive remark into legal dogma by using the words "wild beast." Tracy told the jury that to be acquitted by reason of insanity: "[I]t must be a man that is totally deprived of his understanding and memory, and doth not know what he is doing, no more than an infant, than a brute, or a wild beast." Id. at 765. Several modern commentators have pointed out that even Tracy did not seek to create a "wild beast" test of insanity. His full charge to the jury in Arnold stressed the cognitive capacities of reason, sense, memory, and understanding (including understanding of the distinction between good and evil) and the volitional capacities of will and intention. As Platt and Diamond have observed, Tracy's jury instructions "represented a medley of legal theories of responsibility mixed with popular superstitions about mental illness." Platt & Diamond, supra, at 360; see also D. HERMANN, supra note 2, at 28-29 (analysis of Judge Tracy's instructions); H. WEIHOFEN, supra note 8, at 20 n.24 (same).

n13 Lambard wrote several editions of his treatise, entitled Eirenarcha. In one edition he stated,

If a mad man or a natural foole, or a lunatike in the time of his lunacie, or a childe y apparently hath no knowledge of good nor evil do kil a ma, this is no felonious acte, nor any thing forfeited by it . . . for they canot be said to haue any understanding wil. But if upo examinatio it fal out, y they knew what they did, & y it was ill, the seemeth it to be otherwise.

W. LAMBARD, EIRENARCHA Cap. 21.218 (1581), quoted in J. BIGGS, supra note 8, at 84 and in United States v. Currens, 290 F.2d 751, 764 (3d Cir. 1961). In another edition of his treatise, Lambard wrote:

In all these sorts of voluntarie Manslaughter, being exempted from the fault of Felonie (as that is, which we say to be done se defendo) there is no person to be punished, to whome the law hath denied a will, or minde to doe the harme: as a mad man: hee that is borne both deafe, and dumbe: nor an infant under the age of 12 years, unlesse it may by some evident token appeare, that hee had understanding of good and evill: for then, in him Malicia supplebit atafem: and to these (by the opinion of H. Bracton) you may adde the Lunatike during that disease for a fourth . . . .

W. LAMBARD, EIRENARCHA 228 (London 1599); see also United States v. Freeman, 357 F.2d 606, 616 (2d Cir. 1966) (describing shift in 18th century from knowledge of "good or evil" to emphasis on "knowledge"). Biggs has noted that Lambard's treatise marks the first appearance in English law of an insanity test based on cognition of good and evil. J. BIGGS, supra note 8, at 84.

n14 Dalton wrote,

If one that is non compos mentis, or an ideot, kill a man, this is no felony; for they have not knowledg [sic] of good and evil, nor can have a felonious intent, nor a will or mind to do harm: and no felony or murder can be committed without a felonious intent and purpose . . . .

And again, actus non facit reum, nisi mens sit rea., and a mad-man is Amens, id est sine mente, without his mind or discretion, and is only, and enough punished by his madness. . . .

So it is, if a Lunatick person killeth another during his lunacy, it is no felony. . . for all acts done by him in his lunacy, are as the acts of an ideot . . . .

A. DALTON, THE COUNTRY JUSTICE 283 (1666).

n15 This view of cognition as playing a supporting role to volition has been voiced by modern commentators as well. E.g., MODEL PENAL CODE § 4.01 commentary at 156 (1955); S. GLUECK, supra note 2, at 70-71; H. PACKER, supra note 2, at 134; 2 J. STEPHEN, supra note 3, at 170.

n16 1 M. HALE, THE HISTORY OF THE PLEAS OF THE CROWN 14-15 (1778). The echoes of Aristotle sound clearly in the full passage in which the quotation in the text appears:

Man is naturally endowed with these two great faculties, understanding and liberty of will, and therefore is a subject properly capable of a law so called, and consequently obnoxious to guilt and punishment for the violation of that law, which in respect of these two great faculties he hath a capacity to obey: the content of the will is that, which renders human actions either commendable or culpable; as where there is no law, there is no transgression, so regularly, where there is no will to commit an offense, there can be no transgression, or just reason to incur the penalty or sanction of that law instituted for the punishment of crimes or offenses. And because the liberty or choice of the will presupposeth an act of the understanding to know the thing or action chosen by the will, it follows that, where there is a total defect of the understanding, there is no free act of the will in the choice of things or actions.

Id.

n17 Id. at 30. Hale also discussed the distinction between "total" and "partial" insanity, a distinction that has been traced in English law as far back as 1592 to a pamphlet written by attorney-priest Richard Cosin. 1 N. WALKER, supra note 8, at 38-39. While total insanity would excuse criminal conduct, Hale was not certain about the legal effect of partial insanity. 1 M. HALE, supra note 16, at 30-32. This distinction foreshadowed the modern development of the diminished capacity and diminished responsibility defenses. See infra notes 210-17 and accompanying text. In addition, Hale concluded that temporary as well as permanent insanity would excuse a person from criminal liability. 1 M. HALE, supra note 16, at 32.

n18 Hawkins wrote: "The guilt of offending against the law whatsoever, necessarily supposing a wilful disobedience, can never justly be imputed to those who are either incapable of understanding it, or of conforming themselves to it." 1 W. HAWKINS, PLEAS OF THE CROWN 1 (1716).

n19 Hawkins cited the cognition of good and evil insanity test introduced by Lambard: "[I]t is to be observed, that those who are under a natural disability of distinguishing between good and evil, as infants under the age of discretion, ideots, and lunaticks, are not punishable by any criminal prosecution whatsoever." Id. at 1-2. Hawkins concluded, as Hale had, that both temporary and permanent insanity are exculpatory. Id. at 2 n.2.

n20 2 W. BLACKSTONE, COMMENTARIES *20 (emphasis in original).

n21 With respect to the insanity defense, Blackstone wrote:

The second case of a deficiency in will, which excuses from the guilt of crimes, arises, also, from a defective or vitiated understanding, viz., in an idiot or a lunatic. For the rule of law, as to the latter, which may easily be adapted also to the former, is, that "furiosus furore solum punitur" (a madman is punished by his madness alone). In criminal cases, therefore, idiots and lunatics are not chargeable for their own acts, if committed when under these incapacities: no, not even for treason itself.

Id. at *24 (emphasis in original).

n22 For further discussions of these cases, see D. HERMANN, supra note 2, at 28-34; 1 N. WALKER, supra note 8, at 52-83; H. WEIHOFEN, supra note 8, at 20-24; Crotty, supra note 8, at 114-17; Sayre, supra note 8, at 1006.

n23 For a discussion of Bowler's Case, see G. COLLINSON, A TREATISE ON THE LAW CONCERNING IDIOTS, LUNATICS, AND OTHER PERSONS NON COMPOSMENTIS 673 (London 1812), cited in J. BIGGS, supra note 8, at 92-93; Crotty, supra note 8, at 115-16.

n24 LeBlanc instructed the jury to determine whether, at the time of the act, the defendant

was or was not incapable of distinguishing right from wrong, or under the influence of any illusion in respect of the prosecutor which rendered his mind at the moment insensible of the nature of the act he was about to commit, since in that case he would not be legally responsible for his conduct.

J. BIGGS, supra note 8, at 92.

n25 5 Car. & P. 168, 172 Eng. Rep. 924 (1831).

n26 Lord Lyndhurst's vague but plainly cognitive test was whether Offord "did not know, when he committed the act, what the effect of it, if fatal, would be, with reference to the crime of murder. The question was, did he know that he was committing an offence against the laws of God and nature?" Id. at 168, 172 Eng. Rep. at 925. The report then states that Lyndhurst referred with approval to the jury instructions given in Bellingham's Case, (discussed infra notes 37-38 and accompanying text), but without specifying the portion of those instructions to which Lyndhurst referred.

n27 Hadfield's Case, 27 How. State Tr. 1281 (Eng. 1800).

n28 Erskine stated in his opening statement to the jury:

Gentlemen, it has pleased God so to visit the unhappy man before you; to shake his reason in its citadel; to cause him to build up as realities, the most impossible phantoms of the mind, and to be impelled by them as motives irresistible; the whole fabric being nothing but the unhappy vision of his disease -- existing no where else -- having no foundation whatsoever in the very nature of things.

Id. at 1315 (emphasis in original).

Elsewhere in his opening address, Erskine also sowed the seeds of the modern insanity defense formulation known as the product test, discussed infra notes 69-73 and accompanying text: "I must convince you, not only that the unhappy prisoner was a lunatic, within my own definition of lunacy, but that the act in question was the immediate, unqualified offspring of the disease . . . ." Id. at 1314 (emphasis in original). Oddly, the court, with the prosecutor's consent, halted the trial in the middle of Erskine's case. One of the judges indicated that the testimony thus far had shown that Hadfield was quite insane and he suggested to the jury that it acquit Hadfield. The jury obliged by returning a verdict of not guilty by reason of insanity. Id. at 1355.

n29 9 Car. & P. 525, 173 Eng. Rep. 941 (1840).

n30 One prosecutor defined the issue of insanity as whether the defendant

was insane at the time he did the act -- whether he was unconscious of what he was doing, so that he did not know right from wrong. If he was labouring under some delusion, and did not know the consequences of what he was doing -- if he was insane at the time, then he is not an accountable agent.

Id. at 530, 173 Eng. Rep. at 943. Another prosecutor argued to the jury that Oxford should be acquitted by reason of insanity only if he showed that he was "in a state in which he did not know right from wrong -- in which he did not know the nature and quality of the act he was committing." Id. at 545, 173 Eng. Rep. at 949.

n31 Denman instructed the jury:

If some controlling disease was, in truth, the acting power within him which he could not resist, then he will not be responsible . . . . On the part of the defence it is contended, that the prisoner at the bar was non compos mentis, that is (as it has been said) unable to distinguish right from wrong, or, in other words, that from the effect of a diseased mind he did not know at the time that the act he did was wrong . . . . The question is, whether the prisoner was labouring under that species of insanity which satisfies you that he was quite unaware of the nature, character, and consequences of the act he was committing, or, in other words, whether he was under the influence of a diseased mind, and was really unconscious at the time he was committing the act, that it was a crime.

Id. at 546-47, 173 Eng. Rep. at 950.

n32 16 How. State Tr. 695 (Eng. 1724).

n33 Tracy instructed the jury that the question for them to consider was

whether this man hath the use of his reason and sense? If he was under the visitation of God, and could not distinguish between good and evil, and did not know what he did, though he committed the greatest offence, yet he could not be guilty of any offence against any law whatsoever; for guilt arises from the mind, and the wicked will and intention of man. If a man be deprived of his reason, and consequently of his intention, he cannot be guilty . . . . On the other side, we must be very cautious; it is not every frantic and idle humour of a man, that will exempt him from justice, and the punishment of the law. When a man is guilty of a great offence, it must be very plain and clear before a man is allowed such an exemption; therefore it is not every kind of frantic humour or something unaccountable in a man's actions, that points him out to be such a madman as is to be exempted from punishment: it must be a man that is totally deprived of his understanding and memory, and doth not know what he is doing, no more than an infant, than a brute, or a wild beast, such a one is never the object of punishment; therefore I must leave it to your consideration, whether the condition this man was in, as it represented to you on one side, or the other, doth shew a man, who knew what he was doing, and was able to distinguish whether he was doing good or evil, and understood what he did . . . . You are to consider what he was at this day, when he committed this fact. There you have a great many circumstances about the buying the powder and the shot; his going backward and forward: and if you believe he was sensible, and had the use of his reason, and understood what he did, then he is not within the exemptions of the law, but is as subject to punishment as any other person.

Id. at 764-65.

n34 19 How. State Tr. 886 (Eng. 1760). This trial marks the first instance in England or America of recorded testimony given by an expert psychiatric witness. The expert was Dr. John Monro, the Physician Superintendent of Bethlem Asylum. 1 N. WALKER, supra note 8, at 58-62.

n35 In a written summary, Ferrers stated:

If [the witnesses] have not directly proved me so insane as not to know the difference between a moral and an immoral action, they have at least proved that I was liable to be driven and hurried into that unhappy condition upon very slight occasions . . . . If I could have controuled my rage, I am answerable for the consequences of it. But if I could not, and if it was the mere effect of a distempered brain, I am not answerable for the consequences . . . .

19 How. State Tr. at 945 (Eng. 1760).

n36 The prosecutor stated:

If there be a total permanent want of reason, it will acquit the prisoner. If there be a total temporary want of it, when the offence was committed, it will acquit the prisoner: but if there be only a partial degree of insanity, mixed with a partial degree of reason; not a full and complete use of reason, but (as Lord Hale carefully and emphatically expresses himself) a competent use of it, sufficient to have restrained those passions, which produced the crime; if there be thought and design; a faculty to distinguish the nature of actions; to discern the differences between moral good and evil; then, upon the fact of the offence proved, the judgment of the law must take place.

Id. at 947-48.

n37 A Full Report of the Trial of John Bellingham for the Murder of Hon. Spencer Perceval (1812) [hereinafter Report of Bellingham's Case], cited in J. BIGGS, supra note 8, at 90-91. G. COLLINSON, supra note 23, at 636.

n38 According to Report of Bellingham's Case, supra note 37, Mansfield instructed the jury: "The single question was whether, at the time this fact [sic] was committed, he [Bellingham] possessed a sufficient degree of understanding to distinguish good from evil, right from wrong, and whether murder was a crime not only against the law of God, but against the law of his country." J. BIGGS, supra note 8, at 91; 1 N. WALKER, supra note 8, at 271. Collinson reported Mansfield's jury charge as follows:

If a man were deprived of all power of reasoning, so as not to be able to distinguish whether it was right or wrong to commit the most wicked transaction, he could not certainly do an act against the law. Such a man, so destitute of all power of judgment, could have no intention at all. In order to support this defense, however, it ought to be proved by the most distinct and unquestionable evidence, that the criminal was incapable of judging between right and wrong. It must, in fact, be proved beyond all doubt, that at the time he committed the atrocious act with which he stood charged, he did not consider that murder was a crime against the laws of God and nature. There was no other proof of insanity which would excuse murder or any other crime.

G. COLLINSON, supra note 23, quoted in J. BIGGS, supra note 8, at 90-91.

n39 As noted recently by an American Bar Association committee, an insanity test can be designed to avoid "moral mistakes" by setting limits on the scope of the insanity defense. A.B.A. STANDING COMMITTEE ON ASSOCIATION STANDARDS FOR CRIMINAL JUSTICE, PROPOSED CRIMINAL JUSTICE MENTAL HEALTH STANDARDS 329 (1984) [hereinafter A.B.A. REPORT]. For example, one of the reasons cited by the A.B.A. committee for omitting a volitional factor from its proposed test is the difficulty of determining whether an impulse was irresistible or simply not resisted. Id. at 330; cf. Bonnie, supra note 2, at 196 ("there is, in short, no objective basis for distinguishing between offenders who were undeterrable and those who were merely undeterred, between the impulse that was irresistible and the impulse not resisted"); accord, AMERICAN PSYCHIATRIC ASSOCIATION, STATEMENT ON THE INSANITY DEFENSE 11 (1982) [hereinafter A.P.A. STATEMENT]. This purpose also can be served by adjusting the burden of persuasion. A.B.A. REPORT, supra, at 378-83; Bonnie, supra note 2, at 197; Morse, Excusing the Crazy: The Insanity Defense Reconsidered, 58 S. CAL. L. REV. 777, 824-25 (1985).

n40 See Durham v. United States, 214 F.2d 862, 874-75 (D.C. Cir. 1954), overruled, United States v. Brawner, 471 F.2d 969, 991 (D.C. Cir. 1972); see also S. GLUECK, supra note 2, at 92-93; Bonnie, supra note 2, at 196; cf. A. GOLDSTEIN, supra note 3, at 53-58.

n41 See generally H. FINGARETTE, THE MEANING OF CRIMINAL INSANITY 173-215 (1972); S. GLUECK, supra note 2, at 103-04; M. MOORE, LAW AND PSYCHIATRY 244-45 (1984).

n42 10 Cl. & Fin. 200, 210, 8 Eng. Rep. 718, 722 (1843).

n43 State v. Brosie, 113 Ariz. 329, 331, 553 P.2d 1203, 1205 (1976). Subsequently, in State v. Chavez, 143 Ariz. 238, 693 P.2d 893 (1985), the court overruled Brosie, substituting a test of knowledge of "the probable results of his actions" for Brosie's test of knowledge of "the nature and quality of the act." 143 Ariz. at 239, 693 P.2d at 894.

n44 People v. Jenko, 410 Ill. 478, 483, 102 N.E. 783, 785 (1951).

n45 State v. Gramenz, 256 Iowa 134, 138, 126 N.W.2d 285, 288 (1964).

n46 State v. Conte, 157 Conn. 209, 210-11, 251 A.2d 81, 82 (1968) (common law test applied), cert. denied, 396 U.S. 964 (1969). Note that in State v. Toste, 178 Conn. 626, 424 A.2d 293 (1979), the common law test on insanity was held to have been superseded by a 1967 statute adopting the model penal code definition of insanity in Connecticut.

n47 Cole v. State, 212 Md. 55, 58, 128 A.2d 437, 439 (1957).

n48 Bryant v. State, 207 Md. 565, 588, 115 A.2d 502, 512 (1955).

n49 For a similar discussion of the potential breadth of interpretation given by courts to the first part of the M'Naghten test, see A. GOLDSTEIN, supra note 3, at 49-51.

n50 See, e.g., State v. Haman, 285 N.W.2d 180 (Iowa 1979); Regina v. Windle, 2 All E.R. 1, 2 (Crim. App. 1952). See generally cases cited in A. GOLDSTEIN, supra note 3, at 52 n.23.

n51 See, e.g., People v. Schmidt, 216 N.Y. 324, 338-40, 110 N.E. 945, 949-50 (1915); State v. Law, 270 S.C. 664, 667, 244 S.E.2d 302, 304 (1978); State v. Cameron, 100 Wash. 2d 520, 526, 674 P.2d 650, 654 (1983) (en banc). See generally cases cited in A. GOLDSTEIN, supra note 3, at 52 n.24 and H. FINGARETTE, supra note 41, at 153 n.34. Biggs has noted that the pre-M'Naghten test based on knowledge of good and evil -- the predecessor of the second part of the M'Naghten test -- plainly refers to knowledge of moral standards. J. BIGGS, supra note 8, at 91.

n52 E.g., People v. Wolff, 61 Cal. 2d 795, 800-01, 394 P.2d 959, 962, 40 Cal. Rptr. 271, 274 (1964) (enbanc) (knowledge under M'Naghten includes "realization or appreciation of the wrongness of [seriously] harming a human being") (quoting J. HALL, supra note 8, at 520.

n53 G. WILLIAMS, CRIMINAL LAW 477-82 (2d ed. 1961).

n54 Hall, supra note 2, at 773-74; R. PERKINS & R. BOYCE, supra note 8, at 963.

n55 H. WEIHOFEN, MENTAL DISORDER AS A CRIMINAL DEFENSE 77 (1954); accord, Hall, supra note 2, at 780-81. For criticism of this position, see Livermore & Meehl, The Virtues of M'Naghten, 51 MINN. L. REV. 789 (1967).

n56 2 J. STEPHEN, supra note 3, at 167. Stephen used Hadfield as an example of a plainly insane defendant who knew that his conduct was legally wrong, but who believed it was morally proper. Id.

n57 Id. at 163.

n58 Livermore & Meehl, supra note 55, at 808.

n59 The response of the judges in M'Naghten to a question posed by the House of Lords concerning the legal significance of partial delusions seems at first glance to indicate that the judges did not intend to include capacity for active reasoning in either part of the test, but only a defendant's capacity for static, intellectual awareness or knowledge. The judges stated that a partial delusion should exculpate the defendant only if his conduct would have been justified if the facts conformed to his delusion:

[W]e think he must be considered in the same situation as to responsibility as if the facts with respect to which the delusion exists were real. For example, if under the influence of his delusion he supposes another man to be in the act of attempting to take away his life, and he kills that man, as he supposes, in self-defence, he would be exempt from punishment. If his delusion was that the deceased had inflicted a serious injury to his character and fortune, and he killed him in revenge for such supposed injury, he would be liable to punishment.

10 Cl. & Fin. at 211, 8 Eng. Rep. at 723. Some commentators have criticized this position as ignoring the possibility that mental illness could impair not only a person's knowledge of the nature, quality, and wrongfulness of his conduct, but also his reasoning or judgment based on such deluded knowledge. See J. BIGGS, supra note 8, at 152; J. STEPHEN, supra note 3, at 156-57; 1 N. WALKER, supra note 8, at 99-100. As Perkins and Boyce have pointed out, however, the judges took pains to stress that they were tailoring this response precisely to the Lords' question about partial delusions. That is, their answer to the Lords' question about delusions concerned only a defendant who "labours under such partial delusion only, and is not in other respects insane." 10 Cl. & Fin. at 211, 8 Eng. Rep. at 723. By limiting their recommendation concerning delusions to this narrowly defined situation, the judges left the door open to interpreting their recommended insanity defense test as referring to active reasoning as well as to knowledge. R. PERKINS & R. BOYCE, supra note 8, at 966-68; see also Ryan v. People, 60 Colo. 425, 433, 153 P. 756, 759 (1915) ("It is, in effect saying to the jury, the prisoner was mad when he committed the act, but he did not use sufficient reason in his madness.") (quoting State v. Jones, 50 N.H. 369, 387-88 (1854)).

n60 81 Ala. 577, 2 So. 854 (1886).

n61 Id. at 585, 2 So. at 859.

n62 Id. at 597, 2 So. at 866 (emphasis in original). The first recorded attempt to use a test of this nature was in 1760 in Ferrers' Case, 19 How. State Tr. 886 (Eng. 1760), as discussed supra text accompanying note 35. Erskine also included the concept in his argument in 1800 in Hadfield's Case, 27 How. State Tr. 1281 (Eng. 1800), as discussed supra text accompanying note 28. Lord Denman incorporated a species of the test in 1840 in his jury instructions in Regina v. Oxford, 9 Car. & P. 525, 173 Eng. Rep. 941 (1840), discussed supra text accompanying note 31. The doctrine made its American debut in 1834 in State v. Thompson, Wright's Ohio Rep. [1884] 617-22 (1834), quoted in H. WEIHOFEN, supra note 8, at 135 (to be responsible defendant must have had "power to forbear or to do the act"). Other early American cases in which it appeared include Commonwealth v. Rogers, 48 Mass. (7 Met.) 500, 501-04 (1844); Commonwealth v. Mosler, 4 Pa. 264 (1846); and Dejarnette v. Commonwealth, 75 Va. 867, 878-79 (1881). American physician Isaac Ray was an early, influential advocate of the irresistible impulse test. See I. RAY, A TREATISE ON THE MEDICAL JURISPRUDENCE OF INSANITY § 187 (B. Bailyn ed. 1962) (1st ed. Boston 1838). For further discussions of the early history of this test, see J. HALL, supra note 8, at 486-91; H. WEIHOFEN, supra note 8, at 46-52, 64-65.

n63 R. PERKINS & R. BOYCE, supra note 8, at 972.

n64 See, e.g., United States v. Freeman, 357 F.2d 606, 620 (2d Cir. 1966); Durham v. United States, 214 F.2d 862, 873 (D.C. Cir. 1954), overruled, United States v. Brawner, 471 F.2d 969, 991 (D.C. Cir. 1972); Snider v. Smyth, 187 F. Supp. 299, 302 (E.D. Va. 1960), aff'd sub nom. Snider v. Cunningham, 292 F.2d 683 (4th Cir. 1961); ROYAL COMMISSION ON CAPITAL PUNISHMENT, REPORT P314, at 110 (1953) [hereinafter ROYAL COMMISSION REPORT].

N65 See, e.g., Wade v. United States, 426 F.2d 64, 67 (9th Cir. 1970); People v. Duckett, 162 Cal. App. 3d 1115, 1125, 209 Cal. Rptr. 96, 103 (1984); State v. Reidell, 14 Del. 470, 472-74, 14 A. 550, 551-62 (1888); State v. White, 58 N.M. 324, 330, 270 P.2d 727, 731 (1954); A. GOLDSTEIN, supra note 3, at 70-75; R. PERKINS & R. BOYCE, supra note 8, at 974; Hall, supra note 2, at 777-78.

n66 J. HALL, supra note 8, at 113.

n67 United States v. Currens, 290 F.2d 751, 774 (3d Cir. 1961); State v. McCullough, 114 Iowa 532, 535, 87 N.W. 503, 504 (1901); H. FINGARETTE, supra note 41, at 158-72; S. GLUECK, supra note 2, at 55-57; 2 J. STEPHEN, supra note 3, at 167; Davidson, Irresistible Impulse and Criminal Responsibility, 1956 J. FORENSIC SCI. 1, 8-9.

n68 United States v. Currens, 290 F.2d at 774; State v. McCullough, 114 Iowa at 535, 87 N.W. at 506; S. GLUECK, supra note 2, at 55-57; A. GOLDSTEIN, supra note 3, at 69-75; R. PERKINS & R. BOYCE, supra note 8, at 968-69; 2 J. STEPHEN, supra note 3, at 167-68; Grano, Voluntariness, Free Will, and the Law of Confessions, 65 VA. L. REV. 859, 883 (1979). The Model Penal Code test is discussed infra notes 74-81 and accompanying text.

n69 214 F.2d 862 (D.C. Cir. 1954), overruled, United States v. Brawner, 471 F.2d 969, 991 (D.C. Cir. 1972).

n70 49 N.H. 399 (1869), overruled on other grounds, Hardy v. Merrill, 54 N.H. 227, 235 (1875). Justice Doe's explanation of the theory underlying the product test mixes elements of the M'Naghten, irresistible impulse, and product tests:

[I]f the alleged act of a defendant was the act of his mental disease, it was not, in law, his act, and he is no more responsible for it than he would be if it had been the act of his involuntary intoxication, or of another person using the defendant's hand against his utmost resistance; if the defendant's knowledge is the test of responsibility in one of these cases, it is the test in all of them. If he does know the act to be wrong, he is equally irresponsible whether his will is overcome, and his hand used, by the irresistible power of his own mental disease, or by the irresistible power of another person. When disease is the propelling, uncontrollable power, the man is as innocent as the weapon, -- the mental and moral elements are as guiltless as the material. If his mental, moral, and bodily strength is subjugated and pressed into involuntary service, it is immaterial whether it is done by his disease, or by another man, or a brute or any physical force of art or nature set in operation without any fault on his part. If a man knowing the difference between right and wrong, but deprived, by either of those agencies, of the power to choose between them, is punished, he is punished for his inability to make the choice -- he is punished for his incapacity; and that is the very thing for which the law says he shall not be punished. He might as well as be punished for an incapacity to distinguish right from wrong, as for an incapacity to resist a mental disease which forces upon him its choice of the wrong.

. . . [I]f the homicide was the offspring or product of mental disease in the defendant, he was not guilty by reason of insanity.

Id. at 441-42. Biggs reports that the trial court instructed the jury that the verdict in this homicide case "should be not guilty by reason of insanity if the killing was the offspring or product of mental disease in the defendant." J. BIGGS, supra note 8, at 114. The Supreme Court of New Hampshire reaffirmed its adoption of the product test in State v. Jones, 50 N.H. 369, 398 (1871).

n71 214 F.2d at 874-75 (Bazelon, J.) (citation ommitted).

n72 The court stated:

In attempting to define insanity in terms of a symptom, the courts have assumed an impossible role, not merely one for which they have no special competence. As the Royal Commission emphasizes, it is dangerous "to abstract particular mental faculties, and to lay it down that unless these particular faculties are destroyed or gravely impaired, an accused person, whatever the nature of his mental disease, must be held to be criminally responsible * * *." In this field of law as in others, the fact finder should be free to consider all information advanced by relevant scientific disciplines.

Id. at 872 (footnotes omitted) (quoting ROYAL COMMISSION REPORT, supra note 64, at 116); see also S. GLUECK, supra note 2, at 74 (Durham recognizes "the great variety of possible linkages between pathological mental states and behavioral manifestations").

n73 The court gave this directive concerning the nature of admissible evidence and the scope of the jury's deliberations in insanity defense cases:

[U]pon a claim of criminal irresponsibility, the jury will not be required to rely on such symptoms as criteria for determining the ultimate question of fact upon which such claim depends. Testimony as to such "symptoms, phases, or manifestations," along with other relevant evidence, will go to the jury upon the ultimate questions of fact which it alone can finally determine. . . . The jury's range of inquiry will not be limited to, but may include, for example, whether the accused, who suffered from a mental disease or defect did not know the difference between right and wrong, acted under the compulsion of an irresistible impulse, or had "been deprived of or lost the power of his will * * *."

214 F.2d at 875-76 (quoting State v. White, 58 N.M. 324, 329, 270 P.2d 727, 730 (1954)).

In its rejection of specific types of incapacities, the Durham test fails to give jurors, judges, attorneys, witnesses, and the public adequate guidance about the excuplatory character of mental illness. That is, Durham fails to identify what it is about mental illness and the conduct of mentally ill people that can excuse criminal acts in some cases. A test recommended by the Royal Commission on Capital Punishment -- "whether at the time of the act the accused was suffering from disease of the mind (or mental deficiency) to such a degree that he ought not to be held responsible," ROYAL COMMISSION REPORT, supra note 64, at 116 -- is subject to the same criticism. As Glueck wrote:

Unlike the M'Naghten and irresistible impulse tests, Durham fails to provide necessary intervening links between mental aberration and irresponsibility. The traditional tests require the jury to find nor merely the presence of mental illness but that the disorder had an effect in destroying or at least greatly limiting the processes of mentation, comprehension, and self-control basic to guided behavior. Durham, on the other hand, jumps directly from the finding of mental disease to the finding of lack of responsibility without specifying that the jury should go through the intermediate stage of assessing the effect of the mental illness on the processes crucial to rational and controlled conduct. This deficiency in Durham is important, because the concepts embodied in the traditional tests -- lack of knowledge of the nature, quality and wrongfulness of a contemplated act, loss of power of controlling antisocial impulses -- are not merely symptomatic of mental disorder at certain stages but are also indications of the probable destruction of the popular basis of moral accountability in the daily traffic of life.

S. GLUECK, supra note 2, at 103-04 (emphasis in original). Wechsler made the same point in an article about Durham:

If the legal criteria are viewed as an attempt to define mental disease by the enumeration of its symptoms, they are, of course, both unintelligible and presumptuous. Why should they be so regarded? Is it the law that mental disorder ipso facto establishes irresponsibility? Could anyone embrace it as the law without assuring himself that the term disorder would imply the relevant impairment of capacity? As courts and commentators have repeated to the point of tedium, the criteria are addressed to the question of when disorder or defect should be accorded the specific legal consequence of a defense to criminal conviction, with the specific differences in dealing with the individual that this specific legal consequence entails. Thus the criteria are not concerned with the indicia of diagnosis of disease; they are concerned with the effects that a disease must have on the defendant if it is to work the exculpation claimed. The effects they make material are those that bear, as I have shown, on the capacity of the actor for self-control. M'Naghten may be challenged, to be sure, upon the ground that it confines inquiry to the actor's cognitive capacity; more is involved in the capacity for self-control. It may be urged that irresistible impulse, in its ordinary formulation, calls for findings too extreme. These are both relevant and arguable matters. But I submit that it cannot be urged with any show of reason that these criteria should be rejected on the ground that they purport to catalogue the symptoms of defect or of disease.

Wechsler, The Criteria of Criminal Responsibility, 22 U. CHI. L. REV. 367, 373 (1955); see also A. GOLDSTEIN, supra note 3, at 84 ("The principal criticism of Durham was that it . . . was really a 'non-rule.' It was said to provide the jury with no standard by which to judge the evidence, directing it to no pathological factors which were relevant to the law's concerns -- impairment of reason and control.").

The Durham court's vague and expansive "but-for" concept of causation also prompted considerable criticism. E.g., Carter v. United States, 252 F.2d 608, 616-17 (D.C. Cir. 1967); Flowers v. State, 236 Ind. 151, 152-66, 139 N.E.2d 185, 192-94 (1956); State v. Lucas, 30 N.J. 37, 67-72, 152 A.2d 50, 66-69 (1959); A. GOLDSTEIN, supra note 3, at 85-86; D. HERMANN, supra note 2, at 47-49; MODEL PENAL CODE § 4.01 commentary at 159 (Tent. Draft No. 3. 1955). On causation generally, see H.L.A. HART & A.M. HONORE, CAUSATION IN THE LAW 59-78 (1959).

n74 MODEL PENAL CODE § 4.01 (Proposed Official Draft 1962).

n75 MODEL PENAL CODE § 4.01 commentary at 156-57 (Tent. Draft No. 3, 1955).

n76 "The draft accepts the view that any effort to exclude the non-deterrables from strictly penal sanctions must take account of the impairment of volitional capacity no less than of impairment of cognition." Id. comment 3 at 157.

n77 The commentary explains,

The rationale of the [M'Naghten rule] is that there are cases in which reason can not operate and in which it is totally impossible for individuals to be deterred. . . .

Jurisdictions in which the M'Naghten test has been expanded to include the case where mental disease produces an "irresistible impulse" proceed on the same rationale. They recognize, however, that cognitive factors are not the only ones that preclude inhibition; that even though cognition still obtains, mental disorder may produce a total incapacity for self-control.

Id. comment 2 at 156-57.

n78 The commentary states,

In modern psychiatric terms, the "fundamental difference between verbal or purely intellectual knowledge and the mysterious other kind of knowledge is familiar to every clinical psychiatrist; it is the difference between knowledge divorced from affect and knowledge so fused with affect that it becomes a human reality." Zilboorg, Misconceptions of Legal Insanity, 9 AM. J. ORTHOPSYCHIATRY 540, 552 [1938].

Id. at 157; see also A. GOLDSTEIN, supra note 3, at 87 ("[The Model Penal Code test] substitutes 'appreciate' for 'know,' thereby indicating a preference for the view that a sane offender must be emotionally as well as intellectually aware of the significance of his conduct.").

n79 MODEL PENAL CODE § 4.01 note on status of section (Proposed Official Draft 1962) ("change was designed to indicate that the Institute does not disapprove the modification of the formulation by a number of groups that have considered it").

n80 As explained in the commentary:

[The draft] also accepts the criticism of the "irresistible impulse" formulation as inept in so far as it may be impliedly restricted to sudden, spontaneous acts as distinguished from insane propulsions that are accompanied by brooding or reflection. See e.g., ROYAL COMMISSION ON CAPITAL PUNISHMENT, REPORT (1953) para. 314, p. 110.

MODEL PENAL CODE § 4.01 comment 3 at 157 (Tent. Draft No. 3, 1955).

n81 MODEL PENAL CODE § 4.01 comment 4 at 159 (Tent. Draft No. 3, 1955). As explained in the commentary:

The law must recognize that when there is no black and white it must content itself with different shades of gray. The draft, accordingly, does not demand complete impairment of capacity. It asks instead for substantial impairment. This is all, we think, that candid witnesses, called on to infer the nature of the situation at a time that they did not observe, can ever confidently say, even when they know that a disorder was extreme.

Id. at 158; see also A. GOLDSTEIN, supra note 3, at 87 ("it is entirely sensible to leave 'mental disease' undefined, at least so long as it is modified by a statement of minimal conditions for being held to account under a system of criminal law").

n82 Pub. L. No. 98-473, 98 Stat. 2057 (1984) (18 U.S.C.A. § 20(a) (West Supp. 1985)).

n83 18 U.S.C.A. § 20(a) (West Supp. 1985). The act provides for a special verdict of "not guilty only by reason of insanity" in such cases. 18 U.S.C.A. § 4242(b)(3) (West Supp. 1985). Prior to the enactment of the new test, federal courts had used the Model Penal Code test as a matter of common law with some variations among the circuits. See S. 1762, 98th Cong., 1st Sess. (1983); H.R. 2151, 98th Cong., 2d Sess. (1984); 2 P. ROBINSON, CRIMINAL LAW DEFENSES § 173(a), at 282 n.4 (1984); Note, The Proposed Federal Insanity Defense: Should the Quality of Mercy Suffer for the Sake of Safety?, 22 AM. CRIM. L. REV. 49, 55 n.42 (1984); A.B.A. REPORT, supra note 39, at 325. Shortly before passage of the new test, the Fifth Circuit had adopted a similar test in United States v. Lyons, 731 F.2d 243, 248 (5th Cir. 1984), cert. denied, 105 S. Ct. 323 (1984).

n84 The legislative history indicates that although Congress acknowledged the moral basis of a volitional test, it decided not to include a volitional component in the new federal test because of the difficulty of proving reliably whether a particular defendant was unable rather than unwilling to exercise self-control. S. REP. NO. 225, 98th Cong., 2d Sess. 225-29, reprinted in 1984 U.S. CODE CONG. & ADMIN. NEWS 3182, 3407-11.

n85 H.R. REP. NO. 577, 98th Cong., 1st Sess. 12 (1983). See supra note 78 and accompanying text for a discussion of the meaning of the term "appreciate" under the Model Penal Code test.

n86 See supra notes 42-49 and accompanying text for a discussion of the meaning of the terms "nature and quality" under the M'Naghten test.

n87 H.R. REP. NO. 577, supra note 85, at 12 n.9. For discussion of the meaning of the term "wrongfulness" under the Model Penal Code test, see supra note 79 and accompanying text.

n88 Note that the cognitive prong of the Model Penal Code test refers expressly only to appreciation of the wrongfulness or criminality of conduct, omitting the M'Naghten test's explicit reference to appreciation of the nature and quality of conduct. Since the Model Penal Code drafters declared their intent to use a broad cognitive prong, free of the perceived limits of the M'Naghten test, see supra note 78 and accompanying text, and since appreciation of wrongfulness or criminality of conduct generally requires appreciation of the nature and quality of conduct, the cognitive prong of the Model Penal Code test should be interpreted to encompass the M'Naghten test. By including the Model Penal Code and M'Naghten formulations explicitly, the new federal test has the virtue of greater clarity on this issue.

n89 See S. REP. NO. 225, supra note 84, at 229, reprinted in 1984 U.S. CODE. CONG. & AD. NEWS 3182, 3411; H.R. REP. NO. 577, supra note 85, at 13-14.

n90 H.R. REP. NO. 577, supra note 85, at 12-13. The new federal test is very similar to the test developed by Richard Bonnie and recommended recently by the American Psychiatric Association. A.P.A. STATEMENT, supra note 39, at 16-17. Bonnie's test provides that a defendant is not responsible for criminal conduct if "as a result of mental disease or mental retardation, he was unable to appreciate the wrongfulness of his conduct at the time of the offense." Bonnie, supra note 2, at 197. The A.B.A. recently recommended a virtually identical test, providing that "[a] person is not responsible for criminal conduct if, at the time of such conduct, and as a result of mental disease or defect, that person was unable to appreciate the wrongfulness of such conduct." A.B.A. REPORT, supra note 39, at 323. These tests do not include volitional components. They are expansive cognitive tests that use the broad terms "appreciate" and "wrongfulness" introduced by the Model Penal Code. The tests use the term "appreciate" to "encompass 'affective' dimensions of major mental illness," Bonnie, supra note 2, at 197; they thereby take "into account all aspects of the defendant's mental and emotional functioning which relate to the ability to recognize and understand the significance of one's actions." A.B.A. REPORT, supra note 39, at 333. They use the term "wrongfulness" to indicate that they apply to incapacity to appreciate the immoral as well as unlawful character of particular criminal conduct. Id. at 333-34. Along with the new federal test, these tests omit the Model Penal Code's qualification of the relevant incapacity as "substantial" (but without adding the federal test's qualification that the mental illness must be severe). As the A.B.A. Report explains:

This approach has been taken both to simplify the formulation and to reduce the risk that juries will interpret the test too loosely. By using the "substantial capacity," language, the drafters of the ALI standard were trying to avoid the rigidity implicit in the M'Naghten formulation. They correctly recognized that it is rarely possible to say that a mentally disordered person was totally unable to "know" what he was doing or to "know" that it was wrong; even a psychotic person typically retains some grasp of reality. However, it is not necessary to retain the phrase "substantial capacity" in order to take into account these clinical realities. Sufficient flexibility is provided by the term "appreciate," as defined earlier.

Id. at 334.

Bonnie's test eliminates the Model Penal Code's volitional prong because of the unreliability of psychiatric assessments of volitional capacity. As Bonnie wrote:

Few people would dispute the moral predicate for the control test -- that a person who "cannot" help doing what he did is not blameworthy. Unfortunately, however, there is no scientific basis for measuring a person's capacity for self-control or for calibrating the impairment of that capacity. There is, in short, no objective basis for distinguishing between offenders who were undeterrable and those who were merely undeterred, between the impulse that was irresistible and the impulse not resisted, or between substantial impairment of capacity and some lesser impairment. Whatever the precise terms of the volitional test, the question is unanswerable, or it can be answered only by "moral guesses."

Bonnie, supra note 2, at 196; see also A.B.A. REPORT, supra note 39, at 329-32; A.P.A. STATEMENT, supra note 39, at 16.

n91 The principal exceptions to this rule are the test proposed by the Royal Commission on Capital Punishment and the Durham test, discussed supra text accompanying notes 69-73. The court in Durham rejected the traditional approach of other tests of referring expressly to cognitive or volitional incapacities -- or any other "symptoms" of mental illness -- as links to culpability. The opinion in Durham, however, still acknowledged the relevance of cognitive and volitional incapacities to responsibility. See Durham, 214 F.2d at 876. For the balance of this article, the term "cognition" will be used to mean the full range of faculties encompassed by the cognitive prong of the Model Penal Code test, including emotional awareness and active reasoning as well as intellectual knowledge and perception.

n92 For a discussion of cognitive and volitional incapacities as the bases of excuse defenses generally and the insanity defense specifically, see 2 P. ROBINSON, supra note 83, at 222-45, 280-303; see also M. MOORE, supra note 41, at 221.

n93 2 J. STEPHEN, supra note 3, at 170.

n94 H. FINGARETTE, supra note 41, at 186-87.

n95 Id. at 203; see also H. FINGARETTE & A. HASSE, MENTAL DISABILITIES AND CRIMINAL RESPONSIBILITY 206-17 (1979) (disability of mind doctrine based on irrationality). Fingarette recognizes that a criminally insane person might well act rationally in many spheres of his life and at most times. H. FINGARETTE, supra note 41, at 205-09. In addition, not all irrational conduct is material to criminal responsibility. What is crucial is rationality of the conduct with which he is criminally charged. As Fingarette has explained: "A criminally insane person must be not merely incapable of rational conduct, but incapable of conduct that is rational with respect to the criminality of that conduct. That is, he must substantially lack capacity to rationally take into account the criminal implications of his act." Id. at 210-11.

n96 H. FINGARETTE, supra note 41, at 211.

n97 Id. at 239-41.

n98 More recently, Fingarette and Hasse have explained the relationship between rationality and criminal responsibility as follows:

The concept of lack of capacity for rational conduct in regard to the criminal significance of the conduct plays a central role in the understanding of Disability of Mind. It is, we maintain, the concept that underlies the long history of practical intuitive judgments that some people are so mentally disabled as to be nonresponsible.

H. FINGARETTE & A. HASSE, supra note 95, at 218. They further explain:

In broadest terms, then, there is a deep and tacit presumption that an individual shares, at least to some minimal practical extent, in a background of basic concepts, perceptions, values, skills, and attitudes common to the members of the community. This is not all the same as presuming that an individual agrees with, or responds in the same way as, everyone else or even anyone else, or that all individuals always act in the same way or support the same specific policies or institutions. When these presumptions are valid, then we can simply assess conduct by asking: Did the person do that voluntarily? Was it done intentionally? Did the person know the particular facts material to that particular act as relevant to standards of law? A few basic questions tell all -- or at least enough for the purpose of the serious matters contemplated by law. For the intention is seen against a (presumed) background of shared understanding and skills and attitudes.

But when there is real doubt as to the existence or integrity of this background of skills, attitudes, and perceptions, then we cannot take for granted the usual crucial significance that these particular features of conduct have for assessing responsibility. It works the other way around as well: when, after reviewing the circumstances, we come to realize that we have serious doubt whether we can take a person's intent or knowledge or deliberation to be criminally significant in the usual way, that is the sign that we have come to see irrationality, come to see that somehow the background skills and attitudes are gravely awry. The discovery is typically not one of temporal sequence but of an emerging awareness simultaneously that the "mental elements" don't have their usual significance, and that the background mental processes are grossly flawed or deranged.

Id. at 231-32 (emphasis in original).

n99 M. MOORE, supra note 41, at 13-14.

n100 Id. at 13.

n101 Id. at 101.

n102 Id. at 101-10. For a similar, means-ends analysis of rationality, see Bice, Rationality Analysis in Constitutional Law, 65 MINN. L. REV. 1, 6-17 (1980). Moore suggests that emotions might be judged as rational or irrational, M. MOORE, supra note 41, at 107-08, and that emotional irrationality can be a feature of mental illness, Id. at 195-96. He states, however, that criminal law rejects the relevance of emotional rationality to culpability. Id. at 107-08.

n103 Moore writes:

Our mentalistic vocabulary is rich in words labeling, by some functional division, various mental powers. We have capacities of perception, of memory, of imagination, of learning, the basic capacities of reasoning and thinking, the capacities to feel emotion, and the capacities of will to have one's emotions and desires issue in one's actions. It is the impairment of these mental functions that we first and foremost have in mind when we speak of someone as being mentally ill.

M. MOORE, supra note 41, at 195-96.

n104 Maintaining that irrationality is the key feature of mental illness, Moore writes:

This kind of pattern of irrational action is the primary symptom of the mental incapacities we label mental illness. For the mental abilities of perception, memory, imagination, and particularly reasoning are necessary in the acquisition of rational beliefs and in maintaining consistency between belief sets and desire sets. Rationality is one of the fundamental properties by which we understand ourselves as persons, that is, as creatures capable of adjusting our actions as reasonably efficient means to rational ends. Being mentally ill means being incapacitated from acting rationally in this fundamental sense.

Id. at 197 (footnote omitted).

n105 Moore explains the connection between mental illness and moral nonresponsibility as follows:

To be mentally ill is to be very seriously irrational . . . . Yet why does severely diminished rationality preclude responsibility? It is because our notions of who is eligible to be held morally responsible depend on our ability to make out rather regularly practical syllogisms for actions. One is a moral agent only if one is a rational agent. Only if we can see another being as one who acts to achieve some rational end in light of some rational beliefs will we understand him in the same fundamental way that we understand ourselves and our fellow persons in everyday life. We regard as moral agents only those beings we can understand in this way.

Id. at 244-45. Applying the same analysis to the relationship between mental illness and legal nonresponsibility, he writes:

The proper definition of legal insanity is one that utilizes this moral criterion. If criminal law is to reflect our shared notions of culpability, an exemption from punishment based on those moral notions ought to utilize those same moral criteria. The only question appropriate to juries is thus one appealing to their moral paradigm of mental illness: Is the accused so irrational as to be nonresponsible?

Id. at 245; see also Moore, Causation and the Excuses, 73 CALIF. L. REV. 1091, 1137-39, 1148-49 (1985) (insane defendants excused because they lack status of moral agents -- they lack moral capacities).

n106 Morse, supra note 39, at 782-88.

n107 Livermore & Meehl, supra note 55, at 805-10. They express reservations, however, about the use in criminal law of a full sense of rationality that would include the capacity for emotional appreciation of the significance of one's act. Id. at 810.

n108 Hall, supra note 2, at 773-75.

n109 J. FEINBERG, DOING AND DESERVING 285-89 (1970).

n110 See also J. GLOVER, RESPONSIBILITY 97-101 (1970) (identifying openness to persuasion as the distinctive feature of mental illnes with respect to responsibility); Grano, supra note 68, at 872 (regarding criminal responsibility as requiring "the opportunity and capacity to be influenced by rational persuasion").

n111 Although these factors can deter "normal" people from committing criminal acts, they do not always do so. As will be discussed more fully below, see infra notes 130-31 and accompanying text, it is these "normal" people whom the law holds criminally responsible for their conduct. See H. FINGARETTE, supra note 41, at 211 (law exists because rational people sometimes elect to flout it); H. FINGARETTE & A. HASSE, supra note 95, at 226-27 (distinguishing between those who make rational decision to flout law and those who are unable to perceive they are doing anything wrong); M. MOORE, supra note 41, at 207 (distinguishing between unintelligible desires (madness) and intelligible but incorrect desires (badness)).

n112 As Fingarette, Moore, and Morse have maintained, the requisite degree of incapacity for excusing a mentally ill defendant is not necessarily total, comprehensive, or permanent. See H. FINGARETTE, supra note 41, at 203-09; M. MOORE, supra note 41, at 197-98; Morse, supra note 39, at 787. See infra note 144 for a discussion of the requisite extent of incapacity that warrants excusing a defendant.

n113 For thought-provoking discussions about the nature of and distinctions between shame and guilt, see E. ERIKSON, CHILDHOOD AND SOCIETY 251-54 (2d ed. 1963); H. LYND, ON SHAME AND THE SEARCH FOR IDENTITY (1958); GUILT AND SHAME (H. Morris ed. 1971); G. PIERS & M. SINGER, SHAME AND GUILT, A PSYCHOANALYTIC AND CULTURAL STUDY (1971).

n114 For a discussion of the meanings of, and differences between, theoretical and practical reasoning, see M. MOORE, supra note 41, at 13-14. Contemporary hermeneutic philosophers also have illuminated distinctions among types of reasoning, based on Aristotle's analyses. See H. GADAMER, TRUTH AND METHOD 278-89 (1975); Bernstein, From Hermeneutics to Praxis, 35 REV. OF METAPHYSICS 823 (1982); Gadamer, The Problem of Historical Consciousness, in INTERPRETIVE SOCIAL SCIENCE 103, 135-45 (P. Rabinow & W. Sullivan ed. 1979).

n115 Goldstein has written that: "However integrated the personality may be, no one has ever suggested that illness must affect every part of it, or must affect every type of functioning in ways which mark the defendant as incapable of complying with the criminal law." A. GOLDSTEIN, supra note 3, at 75; accord H. FINGARETTE, supra note 41, at 216-27 (unity of mind thesis, if valid, does not mean all dimensions of mind deteriorate uniformly); S. GLUECK supra note 2, at 59-60 (exhibiting one sympton of illness, such as absence of knowledge or irresistible impulse, suggests total personality breakdown in impairing person's ability to exist without endangering himself or others); Guttmacher, Principal Difficulties with the Present Criteria of Responsibility and Possible Alternatives, in MODEL PENAL CODE § 4.01 commentary at 174-75 (Tent. Draft 1955) (psychological theory of integration impractical because defects in one area, such as will, may be prominent while defects in another, such as intellect, may be minimal). Disagreeing with this position, Hall has argued that the human personality is an integrated whole in which an incapacity of one kind is likely to affect other capacities. Hall, supra note 2, at 775-81.

n116 Fletcher has expressed this view clearly:

The assessment of attribution and accountability obviously requires the application of standards to the particular situation of the actor . . . . [T]he standard has a variety of forms, but it always recurs to the same normative question: could the actor have been fairly expected to avoid the act of wrongdoing? Did he or she have fair opportunity to perceive the risk, to avoid the mistake, to resist the external pressure, or to counteract the effects of mental illness?

G. FLETCHER, RETHINKING CRIMINAL LAW 510 (1978); see also id. at 492 ("The attribution of wrongdoing to a particular actor turns always on whether it is fair to hold that individual accountable for the wrongful act. The grounds for denying attribution are called excuses.").

Hart has made a similar point:

[M]ost lawyers, laymen and moralists, considering the legal doctrine of mens rea and the excuses that the law admits, would conclude that what the law has done here is to reflect, albeit imperfectly, a fundamental principle of morality that a person is not to be blamed for what he has done if he could not help doing it . . . .

. . . For in most western morality 'ought' implies 'can' and a person who could not help doing what he did is not morally guilty.

H.L.A. HART, supra note 2, at 174-77; see also Holloway v. United States, 148 F.2d 665, 666-67 (D.C. Cir. 1945) ("Our collective conscience does not allow punishment where it cannot impose blame."), cert. denied, 334 U.S. 852 (1948)); J. RAWLS, A THEORY OF JUSTICE 241 (1971) ("Unless citizens are able to know what the law is and are given a fair opportunity to take its directions into account, penal sanctions should not apply to them."); Hart, The Aims of the Criminal Law, 23 LAW & CONTEMP. PROBS. 401, 414-15 (1958) (traditional law doctrines available to solve problems of defendant's mental incapacity negating blameworthiness); Kadish, supra note 2, at 275 (unjust to punish those incapable of choice); Morse, supra note 39, at 788 (law mitigates punishment of extremely crazy people); A.B.A. REPORT, supra note 39, at 338 ("the defense of mental nonresponsibility is necessary to the fair administration of justice").

n117 See, e.g., H.L.A. HART, supra note 2, at 174-77 (fundamental principle of morality that one is not to be blamed for what he could not help doing); Kadish, supra note 2, at 275 (mentally ill persons cannot be deterred by prospect of punishment because they are incapable of choice).

n118 As explained in the Model Penal Code commentary:

[T]he rationale of the position is that these are cases in which reason can not operate and in which it is totally impossible for individuals to be deterred. Moreover, the category defined by the rule is extreme that to the ordinary man the exculpation of the persons it encompasses bespeaks no weaknesses in the law.

MODEL PENAL CODE § 4.01 commentary at 156-57 (Tent. Draft 1955); see also J. BENTHAM, PRINCIPLES OF MORALS AND LEGISLATION 161 (J. Burns & H.L.A. Hart et. 1970) (1st ed. London 1789) (punishment inefficacious where it could produce no effect); G. WILLIAMS, supra note 53, at 495 ("a psychotic who does not know that his act is wrong is not likely to be deterred by the legal prohibition"); A. GOLDSTEIN, supra note 3, at 15 (insane defendant so unlike us that no deterrent purpose is served by punishment); 2 P. ROBINSON, supra note 83, at 225 (condemnatory and deterrent functions of criminal law not undercut by sparing clearly abnormal actor); Wechsler, supra note 73, at 374 (if incapacities extreme, exculpation will not weaken law).

n119 Hart has argued:

Bentham's argument is in fact a spectacular non sequitur. He sets out to prove that to punish the mad, the infant child or those who break the law unintentionally or under duress or even under 'necessity' must be inefficacious; but all that he proves (at the most) is the quite different proposition that the threat of punishment will be ineffective so far as the class of persons who suffer from these conditions is concerned. Plainly it is possible that though (as Betham says) the threat of punishment could not have operated on them, the actual infliction of punishment on those persons, may secure a higher measure of conformity to law on the part of normal persons than is secured by the admission of excusing conditions.

H.L.A. HART, supra note 2, at 19 (emphasis in original); see also id. at 43.

n120 Hart has explained his position as follows:

[B]y attaching excusing conditions to criminal responsibility, we provide each individual with benefits he would not have if we made the system of criminal law operate on a basis of total 'strict liability'. First, we maximize the individual's power at any time to predict the likelihood that the sanctions of the criminal law shall be applied to him. Secondly, we introduce the individual's choice as one of the operative factors determining whether or not these sanctions shall be applied to him. He can weigh the cost to him of obeying the law -- and of sacrificing some satisfaction in order to obey -- against obtaining that satisfaction at the cost of paying 'the penalty'. Thirdly, by adopting this system of attaching excusing conditions we provide that, if the sanctions of the criminal law are applied, the pains of punishment will for each individual represent the price of some satisfaction obtained from breach of law.

Id. at 47; accord, H. PACKER, supra note 2, at 112 (punishment helps create patterns of conforming behavior, thereby reducing number of occasions when choice of criminal to flout law presents itself). Fletcher has made a similar point about the conceptually symbiotic relationship between liability and excuses, though without Hart's stress on a cost-benefit rationale for excuses:

If the criminal law is to be an institution expressing respect as well as compassion, its institutions must be able both to punish the guilty and excuse the weak. These two sentiments depend on each other. Compassion is possible only so far as punishment is the norm. Punishing wrongdoing is possible only so far as we have a concept of accountability for wrongdoing.

G. FLETCHER, supra note 116, at 846. For an insightful observation about punishment for voluntary wrongdoing as a form of respect, see Morris, Persons and Punishment, 52 MONIST 475 (1968), reprinted in H. MORRIS, ON GUILT AND INNOCENCE 36-43 (1976).

n121 As Fletcher has maintained: "The point is simply that the criminal laws should express the way we live. Our culture is built on the assumption that, absent valid claims of excuse, we are accountable for what we do." G. FLETCHER, supra note 116, at 801-02. The question we will pursue shortly is what accounts for that assumption in both criminal law and everyday life.

n122 See, e.g., G. DE TARDE, PENAL PHILOSOPHY 83-89 (1912) (one indispensable condition of responsibility is social similarity; not fulfilled when act emanates from someone who is insane), reprinted in FREEDOM AND RESPONSIBILITY, supra note 1, at 46-49; M. MOORE, supra note 41, at 196, 223, 244-45 (mentally ill are fundamentally different, not sufficiently rational to be fairly blamed). For a fascinating and disturbing example of such a jurisdictional problem arising from cultural differences, see Regina v. Machekequonabe, 28 O.R. 309 (Can. 1898) (manslaughter conviction of Indian who killed another Indian believing him to be an evil spirit clothed in human flesh).

n123 See supra notes 12-13; see also G. FLETCHER, supra note 116, at 497, 836-39 (pre-19th century common law viewed insane more like brutes and children); M. MOORE, supra note 41, at 223 (insane, like children, not sufficiently rational to be fairly blamed or punished.)

n124 Moore contends that the distinction lies in the practical rationality of the behavior of normal, adult human beings. M. MOORE, supra note 41, at 62-66, 223, 244-45. He writes:

Very crazy human beings are not enough like us in one of our essential attributes, rationality, to be considered persons to whom moral and legal norms are addressed. Crazy people are "excused" from responsibility for the harm they cause for the same reason that young children, animals, stones, and perhaps corporations are excused: None of them has the status of being a person, the only kind of entity obligated by moral and legal norms.

Id. at 65 (emphasis in original). For a discussion of Moore's conception of rationality, see supra notes 99-105 and accompanying text. While I agree with Moore, I also believe that it is important to probe further to discover the link between rationality and responsibility. Why is rationality a condition of responsibility?

n125 Philosopher Martin Heidegger's concept of readiness-to-hand is rooted in this view of meaning. M. HEIDEGGER, supra note 1, at 49-55, 95-122, 172-95. See generally H. GADAMER, supra note 114, at 413-14 ("[I]t is not what is present-at-hand that is calculated or measured in the linguistic composition of the human experience of the world, but it is what exists, what man recognizes as existent as significant, that is expressed in it."); R. PALMER, supra note 1, at 133-34 (1969) (understanding of an object very different from mere intellectual comprehension); P. RICOEUR, FREUD AND PHILOSOPHY 7 (1970) (meaning and the phenomenology of religion). It is interesting to observe that the philosopher Ludwig Wittgenstein, working from a different philosophical tradition, wrote about meaning in a way quite similar to that used by Heidegger. See L. WITTENGENSTEIN, PHILOSOPHICAL INVESTIGATIONS §§ 11, 12, 31 (G. Anscombe trans. 3d ed. 1968); Note, Direct Evidence of State of Mind: A Philosophical Analysis of How Facts In Evidence Support Conclusions Regarding Mental State, 1985 WIS. L. REV. 435, 448-60.

n126 Interpretation of "texts" -- of the already written or already spoken word -- has been a subject of intense intellectual inquiry in the 19th and 20th centuries. The modern field of interpretive scholarship, known as hermeneutics, includes biblical and literary interpretation. Legal analysis by scholars, judges, and lawyers is also an interpretive process. It includes interpretation of legal texts such as constitutions, statutes, and cases and private law documents such as wills and contracts. See generally on the field of hermeneutics, H. GADAMER, supra note 114; D. HOY, THE CRITICAL CIRCLE (1978); R. PALMER, supra note 1. For examples of recent discussions of legal interpretation, see generally Brest, Interpretation and Interest, 34 STAN. L. REV. 765 (1982); Brest, The Misconceived Quest for the Original Understanding, 60 B.U.L. REV. 204 (1980); Fiss, Objectivity and Interpretation, 34 STAN. L. REV. 739 (1982); Stone, Introduction: Interpreting the Symposium, 58 S. CAL. L. REV. 1 (1985); Levinson, Law as Literature, 60 TEX. L. REV. 373 (1982). A major source of the power and versatility of language to convey an enormous range of meaning is its use of symbols that are distinct from the thoughts, feelings, and entities to which they refer. This feature of language is at the root of two aspects of communication that are especially important for criminal law: self-awareness of a person as an individual living in a community and the shared use of language by all members of the community. For a further discussion, see generally H. GADAMER, supra note 114, at 402-14; Taylor, Interpretation and the Sciences of Man, 25 REV. OF METAPHYSICS 3 (1971), reprinted in UNDERSTANDING AND SOCIAL INQUIRY 101 (F. Dallmayr & T. McCarthy eds. 1977).

n127 Modern hermeneutic inquiry has explored the meaning of conduct as well as the meaning of verbal and written texts. E. g., Ricoeur, The Model of the Text: Meaningful Action Considered as a Text, 38 SOC. RES. 529 (1971), reprinted in UNDERSTANDING AND SOCIAL INQUIRY, supra note 126, at 316. Freudian psychology is a prime example of a discipline devoted -- at least in part -- to interpreting the meaning of conduct. See S. FREUD, INTERPRETATION OF DREAMS 277-78 (J. Strachey trans. 1954); M. MOORE, supra note 41, at 301-03; P. RICOEUR, supra note 125; Morse, Failed Explanations and Criminal Responsibility: Experts and the Unconscious, 68 VA. L. REV. 971, 1017 n.135 (1982); Steele, Psychoanalysis and Hermeneutics, 6 INT'L REV. PSYCHO-ANALYSIS 389 (1979). The law also recognizes the expressive, meaningful nature of conduct. See Stone, supra note 126, at 737 ("When we think of interpretation in law, our minds tend to settle, too quickly, on the image of a judge mulling over a statute or other writing. But interpretations of various sorts are ubiquitious throughout our profession, starting with the interpretation of a client's wishes or interests. The law is an interpretation of human action, falling under categories of offer and acceptance, mistake, and criminal intent.") For example, both verbal and nonverbal expressive conduct can qualify as expression for the purpose of first amendment protection. See, e.g., Shuttlesworth v. Birmingham, 394 U.S. 147 (1969) (picketing and parading may constitute methods of expression entitled to first amendment protection). As another example, consider the hearsay rule, which includes assertive conduct as a form of statement covered by the rule. FED. R. EVID. 801(a). Consider also that in recent times, acts of terrorism throughout the world have served as grim examples of the communicative nature of crime. See A. SCHMID & J. DE GRAAF, VIOLENCE AS COMMUNICATION: INSURGENT TERRORISM AND THE WESTERN NEWS MEDIA (1982). These examples involve only intentionally assertive or expressive conduct. I discuss below, infra notes 131-32 and accompanying text, that the meaningful conduct relevant to criminal responsibility includes both intentionally expressive conduct and conduct that is interpreted by others as expressive regardless of the actor's conscious intention or awareness about its meaning.

n128 Philosopher Jean Hampton also has noted the importance of an actor's attitude toward the interests of others as a key feature in the nature of criminal conduct. See, e.g., Hampton, The Moral Education Theory of Punishment, 13 PHIL. & PUB. AFF. 208, 226-27 (1984) (actor's attitude toward interests of others is key feature in nature of criminal conduct). Criminal law acknowledges and protects a variety of interests, including the interests of individual victims in life, safety, and property ownership and possession, and the interests of a community (especially in the case of victimless crimes such as tax evasion and prostitution) in peace, order, safety, moral standards, provision of public services, and even aesthetic and environmental standards. This article does not explore the nature of either the interests protected by criminal law or the ways of violating those interests that criminal law may prohibit. Rather, my purpose here is to examine the ways in which criminal law takes account of offenders' attitudes toward those interests. For discussions of the nature of legally protected interests and of the nature of criminal injury to such interests, see generally J. FEINBERG, OFFENSE TO OTHERS (1985); J. FEINBERG, HARM TO OTHERS (1984); G. FLETCHER, supra note 116; H. PACKER, supra note 2.

n129 Fletcher has aptly distinguished between "descriptive" and "normative" approaches to criminal responsibility. Under the descriptive approach, culpability rests on an "objective" assessment of whether a defendant committed the prohibited act with the particular state of mind that constitutes an element of the alleged crime. In contrast, under the normative approach, culpability rests on a normative assessment of whether a defendant should have avoided the alleged conduct. G. FLETCHER, supra note 116, at 396-401, 491-514. Kadish has maintained that culpability has both descriptive and normative elements. He has written,

The first category we can call mens rea in its special sense. In this special sense mens rea refers only to the mental state which is required by the definition of the offense to accompany the act which produces or threatens the harm . . . . That the absence of mens rea, in this special sense of the required mental state, precludes liability in all of these cases is of course the merest tautology. This is the way these crimes are defined. But it is important to see that they are so defined because the special mens rea element is crucial to the description of the conduct we want to make criminal. And description is crucial in so far as it is regarded as important to exclude from the definition of criminality what we do not want to punish as criminal . . . . The second category of mens rea qualifications to liability is that of legal responsibility, which includes the familiar defences of legal insanity and infancy. These qualifications differ in several particulars from the mens rea qualifications of the first category. In requiring mens rea in the first, special, sense the law is saying that it does not hold a person where he has shown himself by his conduct, judged in terms of its totality, including his mental state, to be no different than the rest of us, or not different enough to justify the criminal sanction. In requiring mens rea in the sense of legal responsibility, the law absolves a person precisely because his deficiencies of temperament, personality or maturity distinguish him so utterly from the test of us to whom the law's threats are addressed that we do not expect him to comply.

Kadish, supra note 2, at 274-75 (emphasis in original); see also H. FINGARETTE, supra note 41, at 131-34 (absence of mens rea in insanity context is effective claim that actor is not responsible agent). See generally H. PACKER, supra note 2, at 108-31.

I agree with the distinction drawn by Fletcher, Kadish, and Packer between the descriptive and normative dimensions of culpability, and with their view that culpability involves a combination of the two dimensions. Indeed, as will be discussed more fully, infra notes 175-88 and accompanying text, I believe that these two dimensions are even more closely related than these commentators maintain. Their pairing as components of culpability is not merely coincidental; rather, they function together as guides for the jury's interpretation of a defendant's conduct.

It should be noted that the normative standards of respect at issue here are peculiar to criminal law. Those standards generally are not exacting. For example, requiring people to refrain from injuring legally protected interests intentionally or knowingly imposes only a minimal normative burden on people. Even at its most demanding, criminal law generally requires only that people steer clear of culpable negligence, that is, of gross deviations from the care that a reasonable person would exercise to avoid injuring legally protected interests (e.g., in cases of criminal negligence or liability due to unreasonable mistake of fact). See generally R. PERKINS & R. BOYCE, supra note 8, at 840-49. This standard represents a different attitude than might be required by moral or religious doctrines. Such doctrines might impose different (and sometimes more exacting) standards of attitude or might seek to protect a greater variety of interests. For example, a moral or religious doctrine might require its adherents to act lovingly toward others, or at least to take greater affirmative responsibility to protect the interests of others.

It is important to emphasize here that by identifying the normative quality of criminal law in this article, I am neither endorsing nor criticizing the normative standards of criminal law. My purpose is not to make a "critical" evaluation of those standards, but rather to note their existence and role in current criminal law. The inquiry in this article rests on an interpretation of criminal law's normative nature, but it is not itself a project of normative hermeneutics, which seeks to interpret texts to obtain guidance for appropriate action. Those who seek to understand criminal law in order to determine a course of action (e.g., a juror deliberating in a criminal trial, an appellate judge reviewing the propriety of a criminal verdict, or a person deciding whether to engage in conduct that might be criminally prohibited) do engage in normative hermeneutics. See Garet, Comparative Normative Hermeneutics: Scripture, Literature, Constitution, 58 S. CAL. L. REV. 35, 39 (1985) (judges seeking guidance in constitutional text and case law engage in normative hermeneutics).

n130 Hart has vividly described the importance for everyday moral judgments of the meaning conveyed by conduct:

Human society is a society of persons; and persons do not view themselves or each other merely as so many bodies moving in ways which are sometimes harmful and have to be prevented or altered. Instead persons interpret each other's movements as manifestations of intention and choices, and these subjective factors are often more important to their social relations than the movements by which they are manifested or their effects. If one person hits another, the person struck does not think of the other as just a cause of pain to him; for it is of crucial importance to him whether the blow was deliberate or involuntary. If the blow was light but deliberate, it has a significance for the person struck quite different from an accidental much heavier blow. No doubt the moral judgments to passed are among the things affected by this crucial distinction; but this is perhaps the least important thing so affected. If you strike me, the judgment that the blow was deliberate will elicit fear, indignation, anger, resentment: these are not voluntary responses; but the same judgment will enter into deliberations about my future voluntary conduct towards you and will colour all my social relations with you. Shall I be your friend or enemy? Offer soothing words? Or return the blow? All this will be different if the blow is not voluntary. This is how human nature in human society actually is and as yet we have no power to alter it. The bearing of this fundamental fact on the law is this. If as our legal moralists maintain it is important for the law to reflect common judgments of morality, it is surely even more important that it should in general reflect in its judgments on human conduct distinctions which not only underly morality, but pervade the whole of our social life.

H.L.A. HART, supra note 2, at 182-83. Perkins and Boyce have made a similar point:

Deeply ingrained in human nature is a tendency to distinguish intended results from accidental happenings. This is the everyday experience of one on the street. One who has been greatly benefited by the act of another may be very much pleased in any event, but his feeling toward the other will not be the same if it was quite an accidental result as if it was the very purpose intended. And one who has been painfully injured by another's act will not have the same personal resentment if it was obviously an accidental injury as he will if the harm was inflicted upon him intentionally. "I didn't mean to" is an explanation accepted so frequently that it is often one of the early acquisitions of small children.

R. PERKINS & R. BOYCE, supra note 8, at 826.

n131 Gadamer has illuminated the shared nature of the project of actor and observer in determining the meaning of texts and conduct. See H. GADAMER, supra note 114, at 91-119, 325-41; see also Taylor, supra note 126. For a helpful guide to Gadamer's discussion of this and related issues, see generally Bernstein, supra note 114. For a similar analysis of the intersubjective nature of meaning, based on the philosophical perspective of Wittgenstein, see generally Note, supra note 125, at 448-60.

n132 Hermeneutic philosophers have discussed this issue at length, particularly with respect to the interpretation of verbal and written texts. See, e.g., H. GADAMER, supra note 114, at 335, 356; Ricoeur, The Hermeneutical Function of Distanciation, in HERMENEUTICS AND THE HUMAN SCIENCES 139 (J. Thompson ed. 1981); Gadamer, The Philosophical Foundations of the Twentieth Century, in H. GADAMER, PHILOSPHICAL HERMENEUTICS 122 (1976) (arguing that author's conscious intent and understanding regarding meaning of his own work does not necessarily determine that meaning conclusively). See D. HOY, supra note 126, at 11-40 (discussing and criticizing emphasis of 19th century hermeneutic scholars Schleiermacher and Dilthey on original meaning of text as intended and understood by its author; analyzing similar views of modern literary critic Hirsch). Modern hermeneutic psychology also stresses the importance of the intersubjective meaning of conduct, thereby rejecting both a subjectivist focus on the contents of an actor's consciousness and a behaviorist failure to regard conduct as meaningful. See Steele, supra note 127, at 389-94 (interpretation of conduct entails a shared project or dialogue between actor and interpreter).

The weight to which an author's own understanding of the meaning of his text is entitled also has generated considerable controversy among legal scholars with respect to legal texts such as constitutions, statutes, and judicial opinions. In such cases, commentators vigorously have debated the importance to be given to the drafters' intent and to the drafters' understanding of their own intent. For a sampling of arguments in favor of the conclusiveness of drafters' intent and understanding of that intent, see generally R. BERGER, DEATH PENALTIES (1982); R. BERGER, GOVERNMENT BY JUDICIARY (1977); Bork, The Impossibility of Finding Welfare Rights in the Constitution, 1979 WASH. U.L.Q. 695; Monaghan, Our Perfect Constitution, 56 N.Y.U. L. REV. 353 (1981); Rehnquist, The Notion of a Living Constitution, 54 TEX. L. REV. 693 (1976). For examples of opposing arguments, see Brest, The Misconceived Quest for the Original Understanding, 60 B.U.L. REV. 204 (1980); Dworkin, Law as Interpretation, 60 TEX. L. REV. 527, 536-40 (1982); Levinson, Law as Literature, 60 TEX. L. REV. 373 (1982); Perry, The Authority of Text, Tradition, and Reason: A Theory of Constitutional "Interpretation," 58 S. CAL. L. REV. 551 (1985). For a historical argument that the framers of the Constitution did not expect future generations to look to the framers' intent in order to interpret the meaning of the Constitution, see Powell, The Original Understanding of Original Intent, 98 HARV. L. REV. 885 (1985).

Wills and contracts are common examples of private law texts for which the drafters' intent is often in question. The relevant meaning of a will is the testator's intent concerning the disposition of his estate, while the relevant meaning of a contract is the intent of the parties concerning an agreement for specified action or forbearance. Note, however, that even though it is the intent of the testator or contracting parties that is at issue in construing a will or contract, neither the testator nor a contracting party is deemed by the law always to be the authoritative voice on the interpretation of that intent. This understanding of his own intent can lose its authoritative power when, for example, a document is ambiguous or when the written content of the document unambiguously conflicts with his orally expressed understanding of the document. Regarding the construction of contracts, see generally RESTATEMENT (SECOND) OF CONTRACTS §§ 200-201 (1981); 3 A. CORBIN, CONTRACTS §§ 536-543, 572B (1960 & Supp. 1971); 4 S. WILLISTON, A TREATISE ON THE LAW OF CONTRACTS §§ 600-603 (3d ed. 1961). On the construction of wills, see generally RESTATEMENT OF PROPERTY § 241 (1940); 4 W. PAGE, THE LAW OF WILLS § 30.7 (rev. ed. 1961). In the realm of interpreting conduct, Freud stands out as a modern pioneer in exploring meaning that is neither intended nor known by the actor. See generally Steele, supra note 127.

n133 As Greenawalt has observed, "The law must concern itself with judgments that strangers (members of a jury) are capable of making." Greenawalt, The Perplexing Borders of Justification and Excuse, 84 COLUM. L. REV. 1897, 1918 (1984); see also United States v. Brawner, 471 F.2d 969, 982 (D.C. Cir. 1972) (en banc) ("The 'moral' elements of the decision [about a defendant's responsibility] are not defined exclusively by religious considerations but by the totality of underlying conceptions of ethics and justice shared by the community, as expressed by its jury surrogate."); accord, State v. Johnson, 121 R.I. 254, 266-68, 399 A.2d 469, 476-77 (1979) (establishing standard for criminal responsibility that preserves jury's role as trier of fact and ultimate decision maker).

n134 See infra notes 161-88 and accompanying text.

n135 "Interpretation" here is used to mean both a descriptive and a normative (evaluative) analysis of conduct. Cf. Bice, supra note 102, at 6 (defining "rationality analysis" in a similar way).

n136 Note that the structure of this analysis is similar to the structure of analysis involved in the third of three categories of pre-M'Naghten cases and commentary discussed supra notes 9-38 and accompanying text. In that analysis, cognitive capacity was deemed to be a necessary condition of volitional capacity which, in turn, was regarded as a necessary condition of culpability. Under the interpretive theory presented here, cognitive and volitional capacities to conform to legally and morally relevant factors are seen as necessary conditions of conveying through conduct the meaning of respect or disrespect for legally protected interests. Such expressive significance or meaning, in turn, is a necessary condition of culpability.

n137 Fletcher has made a similar point: "The distinguishing feature of excusing conditions is that they preclude an inference from the act to the actor's character." G. FLETCHER, supra note 116, at 799. The view I propose here differs from Fletcher's view by focusing on a discrete episode of activity undertaken by the actor rather than the actor's character as developed and manifested through many episodes over time. Although an actor's character may be crucial to moral evaluation of a person, criminal law generally pays attention -- for the purpose of assessing responsibility -- only to distinct instances of an actor's behavior, not to his character over time. His character may become an important issue for sentencing purposes. It may also be used for limited evidentiary purposes in guilt determination. See, e.g., FED. R. EVID. 404. In such cases, however, the defendant's character does not function as an element of an offense, but rather as a type of evidence the jury may use to determine whether the defendant committed the particular crime for which he stands charged. In criminal law as in the Bible, both a saint and an inveterate sinner can be guilty of a single instance of wrongful conduct. See Numbers 20:12; 2 Samuel 11-12.

n138 If his mistake was unreasonable, under a common law grading system he would be convicted for involuntary manslaughter for a negligent homicide rather than the more serious offense of first- or second-degree murder. Note that the law does not always require that mistakes be reasonable to result in exculpation. See generally G. FLETCHER, supra note 116, at 683-91 (discussion of cases and rationales for varying the effects of mistake in criminal law); R. PERKINS & R. BOYCE, supra note 8, at 1044-54 (expansive analysis of effect of both reasonable and unreasonable mistake in case law); 1 P. ROBINSON, supra note 83, at 245-53 (analysis of reckless, negligent, and faultless mistakes and culpability requirements in modern penal codes). As will be discussed infra note 176 and accompanying text, grading of offenses by different mental states reflects the varying degrees of disrespect about the legally protected interests involved.

n139 Fletcher has drawn an interesting distinction between accidents and mistakes: "Accidents occur in the realm of causation. When expected forces go awry, the result is an accident. Mistakes occur in the realm of perception." G. FLETCHER, supra note 116, at 487. For other statements of this distinction, see 2 W. BLACKSTONE, supra note 20, at *26-27 (accident follows in the performance of a lawful act; mistake is a defect of ignorance); 1 P. ROBINSON, supra note 83, at 270 (one mistakes a particular quality or property of an element, but accidentally causes injury); Austin, supra note 1, at 10 n.4, 17 n.14 (noting common misuse of words, defining mistake as error in perception). I suggest one modification in this valuable distinction: accidents do not occur in the realm of causation but in the realm of perception, or cognition, of causation. An event is labeled an accident when expected forces go awry, but in what sense do such forces go awry? They do so when other relevant causal factors, unknown to the actor, operate to alter the expected outcome. Consider Fletcher's example of an accident: I shoot at a tree stump and the bullet ricochets, hitting a man standing nearby. G. FLETCHER, supra note 116, at 487. This unfortunate incident occurs because I did not perceive that the bullet might ricochet off the stump in a direction and with sufficient force to hit and injure the man. An accident has occurred in this example only in the sense that I did not understand the other relevant forces at work. Perhaps the distinction could be phrased this way: mistakes occur in the realm of cognition of stationary circumstances, while accidents occur in the realm of cognition of active causes. For a helpful discussion of circumstances and causes, see Hart & Honore, Causation in the Law, 72 L.Q. REV. 58 (1956), reprinted in FREEDOM AND RESPONSIBILITY, supra note 1, at 325 (clarifying common sense principles of causation, law distinguishes conditions from causes in circumstances of given event).

n140 For discussion of the volitional nature of duress, see G. FLETCHER, supra note 116, at 802-06 (assessment of voluntariness depends in part on nature of competing interests); 2 P. ROBINSON, supra note 83, at 239-41, 351-55 (critical question is whether the threat of harm causes impairment of ability to control behavior). Both Fletcher and Robinson rightly have distinguished between duress or "psychological" compulsion and physical compulsion. G. FLETCHER, supra note 116, at 802-03; 2 P. ROBINSON, supra note 83, at 351; see also H.L.A. HART, supra note 2, at 144 (acts under duress are not involuntary movements, but the action may not be the outcome of the actor's free choice); P. LOW, J. JEFFRIES & R. BONNIE, supra note 2, at 612-14 (duress analysis considers external contraints on actor's capacity to comply with the law and instances where the actor has no choice at all); 2 J. STEPHEN, supra note 3, at 102 (acts without intention, hence involuntary, distinguished from disliked actions voluntarily selected under compulsion); Grano, supra note 68, at 880-84 (all acts under duress are nonetheless voluntary; question for legal responsibility must be degree of impairment of mental freedom that will excuse defendant). As an example of physical compulsion, consider a person who knocks another person to the ground, but only because he himself was pushed by a third person into the victim. As will be discussed infra notes 181-88 and accompanying text, criminal law takes account of such physical compulsion through the concept of actus reus. See P. LOW, J. JEFFRIES & R. BONNIE, supra note 2, at 152-57 (illustrations of acts termed involuntary by the law); 2 P. ROBINSON, supra note 83, at 259-74 (analysis of current state of involuntary action defense). In a case of duress, the actor has the physical capacity to conform his conduct to relevant moral and legal factors. We excuse him because the psychological pressure he faces is so severe that we do not expect him to resist it in order to adhere to these usually relevant factors. For example, the innocent woman who drives the getaway car for the armed bank robber physically could refuse to do so. We regard the threat of death if he refuses to drive the car, however, as exerting such powerful psychological pressure that we excuse her on the grounds of duress.

The issue in duress is a normative decision: given the psychological pressure involved, do we, speaking through the jury, still expect a person to conform to relevant legal and moral factors, or do we consider the pressure to be too great to resist? The normative nature of the excuse of duress is reflected in several criteria for its application. For example, the seriousness of the threat to the actor generally is assessed under an "objective" standard which requires that the threat must be sufficiently great that a person of reasonable firmness could not resist. Id. at 353-54. Also, as Robinson has observed, jurisdictions "commonly limit the permissible causes of the duress disability by requiring, as did the common law, threats of death or imminent bodily injury." Id. at 359. Finally, many jurisdictions limit the offenses for which duress qualifies as an excuse. Id. at 367-71.

Necessity can be viewed as a defense analogous to duress, with pressure exerted by natural, or at least nonhuman, circumstances, instead of a human agent. The analysis of the exculpatory effect of duress presented above also applies to this view of necessity. An alternative view, however, depicts necessity as a justification defense based on a choice of lesser evils. See G. FLETCHER, supra note 116, at 818-35 (tracing and discussing historical development in Europe and America of necessity defense defined as lesser of two evils, where expediency of situation is so great that it justifies breaking law); 1 P. ROBINSON, supra note 83, at 108-09 (discussing necessity as justification defense exculpating justified conduct no matter what the source).

n141 Some observers have maintained that the insanity defense (perhaps along with infancy) should not be regarded as a standard excuse defense, because, as a defense related to a defendant's status rather than the quality of his acts, it rests on principles different from those of other excuse defenses. See, e.g., H. FINGARETTE, supra note 41, at 128-42 (insanity defense means no responsibility status; defendant not a responsible agent, which is something more significant than concept found in defenses of compulsion or lack of knowledge); H. GROSS, supra note 2, at 136-58, 306 (no point in worrying about intention when actor not responsible agent); Moore, supra note 105, at 1098, 1137-39 (insane should be excused not because their mental illness causes them to commit crimes but because they lack status of moral agents). See generally M. MOORE, supra note 41, at 222; H. PACKER, supra note 2, at 134-35. This position finds theoretical support in Kadish's distinction between two categories of mens rea: mens rea as the defined mental element of a crime and mens rea as criminal responsibility. See infra note 188.

I agree that the kind of incapacity at issue in cases of mistake, accident, and duress differs from the kind at issue in cases of insanity and infancy. The latter two involve an underlying incapacity of an actor that can affect his response to legal and moral factors in all, or at least many, situations. In contrast, the first three excuses are situational excuses that arise when particular circumstances impair an actor's capacity to guide his conduct according to morally and legally relevant factors. This section of the article will show the common basis of the insanity defense and the other excuse defenses. The relevance of this common basis to the general role of mens rea is discussed infra notes 174-88 and accompanying text.

By offering an interpretive rationale for the insanity defense, I do not propose a new test for the defense. My purpose is to probe the theoretical underpinnings of the defense. While the rationale for the defense may be sufficiently broad to encompass a wide range of cognitive and volitional impairments as exculpatory, courts and legislatures nevertheless may legitimately take into account important policy and practical considerations in deciding to narrow the defense. See, e.g., supra notes 82-90 and accompanying text (discussion of new federal test and tests proposed by Bonnie and American Bar Association and American Psychiatric Association: defendant not responsible only if he was unable to appreciate wrongfulness of his conduct at time of offense; volition component excluded entirely).

n142 The interpretive theory of the insanity defense offers a resolution of the traditional debate between lawyers and psychiatrists over free will and determinism. According to standard legal theory, criminal responsibility rests on a premise that most people act with free will, while much psychiatric theory insists that human actions are predetermined by interactions of intrapsychic forces, by neurochemical reactions, or by both causal factors. See, e.g., State v. Sikora, 44 N.J. 453, 475-79, 210 A.2d 193, 205-07 (1965) (Weintraub, C.J., concurring) (absurd to accept psychiatric thesis that finds no basis for individual criminal liability); H.R. REP. NO. 577, supra note 85, at 11-12 (acknowledging that tenets of modern psychology would entirely undermine current justice system; rejecting defenses of psychology as inexact); J. GLOVER, supra note 110, at 43-45 (noting determinists believe no human behavior is in principle unpredictable); S. GLUECK, supra note 2, at 5-19 (noting conflict between theories; suggesting capacity for conscious, controlled action among alternatives differs among individuals); P. LOW, J. JEFFRIES & R. BONNIE, supra note 2, at 697-701 (noting 20th century psychiatric theory that behavior is unconsciously motivated); M. MOORE, supra note 41, at 350-65 (discussing and rejecting causation theory of unconscious motivation as inapplicable to legal inquiry about actor's responsibility); Moore, supra note 105, at 1091-1112 (criticizing entire causation theory of criminal law excuses, including unconscious intrapsychic motivation; rejecting theory as inadequate and unsound); Morse, supra note 127, at 983-91 (tracing from Freud development of the psychoanalytic theory and its central hypothesis that unconscious forces are primary determinants of behavior). For philosophical discussions of this issue, see DETERMINISM AND FREEDOM IN THE AGE OF MODERN SCIENCE (S. Hook ed. 1958). The interpretive theory avoids this dilemma by requiring only a limited concept of freedom. Criminal law need not make a broad claim for absolute freedom of choice unaffected by neurochemical or intrapsychic causal factors. Plainly, even if our will is free to some extent, it also is constrained in many ways. For example, our physical composition limits to a great extent our freedom to perceive, feel, and act. The theory needs to ascribe to most people only the freedom sufficient to permit others to interpret an actor's attitude toward legally protected interests. This degree of freedom is defined as the ability to incorporate relevant moral and legal factors in one's choice of conduct. That is, since criminal responsibility under the interpretive theory requires only the capacity to express respect for legally protected interests, which in turn requires only the capacity to guide conduct according to relevant moral and legal factors, criminal responsibility requires only the freedom necessary to respond to these factors. See infra note 160 (applying interpretive theory to unconscious and conscious motivations). In the world of philosophy, this position is compatible with "soft determinism." For similar views, see G. FLETCHER, supra note 116, at 801-02 (unless we abandon entire idea of criminal law, excuse must represent limited situation where actor's capacity for choice was distorted); S. GLUECK, supra note 3, at 92-95 (most persons can choose to act in more or less rational manner, unless something interferes with the volitional capacity); M. MOORE, supra note 41, at 360-65 (it is disturbance of practical reasoning that excuses, not fact that such disturbance was caused); Grano, supra note 68, at 870-72 (mental freedom entails opportunity and capacity to be influenced by rational persuasion); Moore, supra note 105, at 1128-39 (excuses valid only if involve interference with individual's capacity to follow normal dictates of reason). For an introduction to the concepts of hard and soft determinism, see J. GLOVER, supra note 110, at 62 (soft determinists believe a series of causal laws predicts all behavior yet still accept normal moral concepts of blame; hard determinists reject these concepts of blame); M. MOORE, supra note 41, at 350-65 (rejecting hard determinist views that unconscious mental motivations cause bad behavior and that causation excuses; suggesting soft determinism applicable where causation is irrelevant to criminal responsibility); Edwards, Hard and Soft Determinism, in DETERMINISM AND FREEDOM IN THE AGE OF MODERN SCIENCE, supra, at 104 (soft determinists see no antithesis between determinism and moral responsibility; hard determinists see one's whole character derived from inherited equipment and environmental influences, hence no moral responsibility).

Although I believe that criminal responsibility requires only this minimal degree of freedom, I wish to sketch an even stronger claim regarding human freedom. Determinists view human beings basically as complex billiard balls upon which varied neurochemical reactions or intrapsychic forces act. Even if an unbroken series of neurochemical or intrapsychic events "causes" a particular form of conduct, however, we should ask what steers these causal chains in particular directions? My answer, briefly, is that we are not just arbitrarily arranged collections of interacting atoms; rather we are intricately organized beings whose physical composition gives us the capacity -- albeit limited -- to find, create, and respond to meaning. This capacity to attend to meaning is a mediating one that gives human beings some freedom -- not full freedom -- to control their own conduct. See also supra notes 125-26 and accompanying text (reason for insanity defense is special human capacity to convey and interpret meaning from conduct).

n143 MODEL PENAL CODE § 4.01 commentary at 156 (Tent. Draft No. 4, 1955).

n144 See Boykin v. State, 149 Ga. App. 457, 254 S.E.2d 457 (1979) (despite delusion about alien force, defendant still knew right from wrong; conviction for aggravated assault with intent to rob affirmed); State v. Williams, 346 So.2d 181 (La. 1977) (despite suffering form of schizophrenia with impaired judgment, evidence suggested defendant was able to distinguish right from wrong); Commonwealth v. Brown, 386 Mass. 17, 434 N.E.2d 973 (1982) (despite latent chronic schizophrenia, defendant had capacity to conform his conduct to the law); Commonwealth v. Tempest, 496 Pa. 436, 437 A.2d 952 (1981) (despite chronic schizophrenia, defendant knew drowning her six-year-old son was wrong); Cowles v. State, 510 S.W.2d 608 (Tex. Crim. App. 1974) (mental illness alone does not establish legal insanity; a defendant is legally insane only if his illness prevents him from distinguishing between right and wrong); R. PERKINS & R. BOYCE, supra note 8, at 956-58 (question is whether defendant's mind was impaired to such a degree and in such a way as to preclude criminal responsibility); 2 P. ROBINSON, supra note 83, at 222-28, 291-93 (mental disorder by itself does not excuse; it excuses only if disorder renders him blameless for conduct constituting particular event).

The question naturally arises as to how severe a defendant's mental disorder must be to excuse his criminal conduct. Although I cannot quantify the requisite degree of illness, it seems that only serious disorder should be fully exculpatory. The vast majority of people -- even many people with psychiatric problems -- have sufficient cognitive and volitional capacity to express meaning by taking account of morally and legally relevant factors in their conduct. Only a very few defendants are so cognitively or volitionally impaired that they are unable to incorporate those relevant factors to an extent that we would not interpret disrespect from their conduct. For similar conclusions reached for different reasons, see G. FLETCHER, supra note 116, at 504-14 (moral accountability, i.e., the demands of social protection through justice system, requires limited defense of insanity); Wechsler, supra note 73, at 374-75 ("category of the irresponsible . . . must be so extreme that to the ordinary man, burdened by passion and beset by large temptations, the exculpation of the irresponsibles bespeaks no weakness in the law"). See infra notes 210-219 and accompanying text for a discussion of the related issue of diminished responsibility, a partial defense that might arise when a defendant's capacity to incorporate legally and morally relevant factors into conduct is lessened, but to a degree that only diminishes the meaning of conduct.

n145 The summary of this case history quoted in the text is reprinted from Bonnie, supra note 2, at 195-96. A more complete case history and analysis appears in P. LOW, J. JEFFRIES & R. BONNIE, supra note 2, at 660-83.

n146 Livermore & Meehl, supra note 55, at 844-45. For similar cases, see State v. Reidell, 14 Del. 470, 14 A. 550 (1888) (husband suffering from meloncholia kills his wife and injures himself); State v. Pouncey, 297 Me. 264, 465 A.2d 475 (1983) (defendant drowns five-year-old to prevent him from being captured by devil and taken to hell); W. SULLIVAN, CRIME AND INSANITY 91-100 (1924) (case studies of murderers diagnosed as suffering manic-depressive insanity); Diamond, Criminal Responsibility of the Mentally Ill, 14 STAN. L. REV. 59, 61-62 (1961) (woman strangles eight-week-old child believing it better off dead, declared legally insane due to schizophrenia).

n147 Livermore & Meehl, supra note 55, at 845-46.

n148 27 How. State Tr. 1281 (Eng. 1800), discussed supra notes 27-28 and accompanying text.

n149 Apparently Hadfield is not alone in psychiatric annals in his desire to secure self-sacrifice through execution. See W. SULLIVAN, supra note 146, at 95 (occasionally a melancholic commits murder, with no ill-feeling towards his victim, in order to be hanged and so achieve a suicidal purpose). Hadfield did not take into account the one circumstances that ultimately thwarted satisfaction of his yearning for execution: the effective advocacy of attorney Erskine.

n150 As will be discussed infra notes 189-90 and accompanying text, the interpretive theory proposed in this article also explains why even the allegedly good motives of a sane defendant are irrelevant to criminal responsibility.

n151 See supra notes 15-21 and 32-38 and accompanying text.

n152 Bonnie has suggested that a defendant whose volitional capacity has been so impaired by mental illness that he should be excused is also likely to have suffered from exculpatory cognitive impairment. Bonnie, supra note 2, at 196-97.

n153 51 Cal. 2d 716, 336 P.2d 492 (1959).

n154 See Diamond, supra note 146, at 77-80.

n155 Id. at 79.

n156 At the time of the trial in Gorshen, California employed only the M'Naghten test, rejecting the irresistible impulse test. Gorshen, 51 Cal. 2d at 726, 336 P.2d at 498-99. The defense in Gorshen shrewdly sought to submit evidence of Gorshen's uncontrollable impulse to kill through the rear door in the form of a diminished capacity partial defense. Se Arenella, The Diminished Capacity and Diminished Responsibility Defenses: Two Children of a Doomed Marriage, 77 COLUM. L. REV. 827, 836-49 (1977); Diamond, supra note 146, at 73-81.

n157 P. LOW, J. JEFFRIES & R. BONNIE, supra note 2, at 682.

n158 For other examples of cases involving prominent volitional impairment accompanied by or related to cognitive impairment, see Commonwealth v. Chester, 337 Mass. 702, 150 N.E.2d 914 (1958) (despite testimony that defendant viewed himself as worthless, was obsessed with guilt, and was given to violent fits of emotional expression outside of conscious control, conviction sustained) and the cases of"F.W." and "S.D." in W. SULLIVAN, supra note 146, at 109-15 (case studies of persons with strong homicidal impulses accompanied by losses of cognitive capacity).

n159 W. SULLIVAN, supra note 146, at 113-14.

n160 For discussions of cases of volitional incapacity possibly resulting from neurological disorders, see Bear, Freeman & Greenberg, Changes in Personality Associated with Neurologic Disease, in 1 PSYCHIATRY ch. 28 (R. Michels ed. 1985) (neurologic illnesses important causes of alternatives in emotion, behavior, and personality); Hecaen & Albert, Disorders of Mental Functioning Related to Frontal Lobe Puthology, in PSYCHIATRIC ASPECTS OF NEUROLOGIC DISEASE 137, 138-40 (D. Benson & D. Blumer eds. 1975) (correlating various behavior disorders with frontal lobe damage); Ratner & Shapiro, The Episodic Dyscontrol Syndrome and Criminal Responsibility, 7 BULL. AM. A. PSYCHIATRY & L. 422 (1979) (in certain circumstances the presence of a neurological deficit considered strong evidence of inability of individual to conform his conduct to requirements of law). Some psychiatrists have maintained that the psychodynamic operation of unconscious drives can also cause individuals to suffer irresistible impulses to commit criminal activity. See, e.g., United States v. Pollard, 282 F.2d 450 (6th Cir. 1960) (accepting psychiatric testimony that defendant had irresistible, unconscious desire to be apprehended and punished; reversing conviction); M. MOORE, supra note 41, at 350-81 (discussing issue of unconscious motivation and criminal responsibility as applied in Pollard). I take no position in this article regarding the debate over the particular causes of conduct that can be excused under the insanity defense. Compare State v. Sikora, 44 N.J. 453, 478-79, 210 A.2d 193, 204-06 (1965) (Weintraub, C.J., concurring) (rejecting as absurd psychiatric theory that can find no basis for individual criminal liability), Morse, supra note 127, at 991-1043 (challenging scientific legitimacy and relevance to criminal responsibility of psychodynamic explanations of unconscious motivation), and M. MOORE, supra note 41, at 350-83 (arguing that unconscious motivation can at best play only a narrow role as an exculpatory factor) with K. MENNINGER, THE CRIME OF PUNISHMENT 118, 180-86 (1968) (supporting an "utterly new" psychiatric understanding of human behavior based on unconscious motivations), Diamond, With Malice Aforethought, 2 ARCHIVES CRIM. PSYCHODYNAMICS 1, 26-27 (1957) (supporting case-by-case approach in analyzing the motivations behind the actions of individuals deviating from the legal concept of free will), and Hospers, Meaning and Free Will, 10 PHIL. & PHENOMENOLOGICAL RES. 307, 320-21 (1950) (advocating scientific legitimacy and relevance to criminal responsibility of psychodynamic explanations of unconscious motivation). Instead, for the purpose of argument, I assume the scientific validity of all types of causation and focus instead on exploring the exculpatory character of any criminal impulse experienced by the defendant as irresistible, regardless of its cause. In the debate over psychodynamic theory, for example, where it is claimed that unconscious drives operate in all people to cause behavior, the important question raised is why only some people should be excused because of these drives? See M. MOORE, supra note 41, at 351 (if all conscious actions are determined by unconscious mental states, "why is everyone not excused for all of his actions?"); Morse, supra note 127, at 1036-39 (no principled way for the unconscious motivation theory to apply to only certain cases). To answer this question, let us adopt the perspective of psychodynamics by dividing a person's conduct into two levels of meaning: conscious and unconscious. See M. MOORE, supra note 41, at 366-67 (discussing psychoanalysis at unconscious and conscious levels). In the case of an offender whom we deem to be sane, an "ordinary" thief, criminal conduct might be undertaken for unconscious reasons, but it is also readily understandable as an action undertaken for conscious reasons, e.g., greed. Here, the thief's conscious and unconscious motives to steal operate in harmony. Moreover, he has channelled his unconscious drives into conscious purposes. He consciously steals valuable items to increase his wealth and satisfy his material desires, while trying to minimize his risk of capture. He knows he is violating legal and moral norms and injuring his victim's property interests, but he consciously prefers to ignore them in order to further his own ends. Because he exercises a conscious preference to steal for known reasons, his congruent unconscious motives are irrelevant to criminal responsibility under the interpretive theory here proposed. His conscious preference conveys the thief's disrespect for the victim's property interests.

In contrast, we might excuse other unconsciously motivated conduct because it lacks this double layer of meaning. In such a case, the two levels of meaning -- conscious and unconscious -- are not in harmony; the actor appears to have no intelligible or adequate conscious reason for acting. See J. FEINBERG, supra note 109, at 284 (crimes with no apparent motive are symbolic of underlying mental illness). This conduct of often marked by its irrational and even self-defeating quality and the actor's lack of insight into his own conduct. Id. at 284-92; see M. MOORE, supra note 41, at 378 (discussing unknown but passionately felt emotions). The defense proposed this view in Pollard, in which a police officer had engaged in a series of robberies and attempted robberies -- several of them pathetically bungled -- arguably without a satisfactory explanation on the level of conscious motivation. In such a case, when unconscious drives erupt into conduct that the actor does not consciously prefer but that he cannot control, the conduct does not express disrespect towards the victim's interests. In short, the question from the standpoint of the interpretive theory is whether the actor's conduct is intelligible on the level of conscious motivation. If it is, then the unconscious motivations are legally irrelevant. If is not, then unconscious motivations might well be legally relevant.

n161 To apply Austin's observation quoted in the first sentence of this article, how does the interpretive theory of excuses throw light on the nature of the normal criminal act?

n162 Sayre, Mens Rea, 45 HARV. L. REV. 974 (1932).

n163 J. HALL, supra note 8, at 151.

N164 Id.; O. HOLMES, THE COMMON LAW 49-55 (1881).

n165 O. HOLMES, supra note 164, at 75.

n166 For discussions of the meanings of "objective" and "subjective" in regard to criminal liability, see G. FLETCHER, supra note 116, at 504-14; A. GOLDSTEIN, supra note 3, at 16-18, 191-94; see also J. HALL, supra note 8, at 146-70 (arguing that subjective factual mistakes and ignorance should be recognized in community's definition of penal wrongs). question of culpability while objectively maintaining community's definition of penal wrongs).

n167 G. FLETCHER, supra note 116, at 398-400, 492-504.

n168 2 J. STEPHEN, supra note 3, at 95, 112-13.

n169 Sayre, supra note 162, at 1025-26.

n170 Turner, The Mental Element in Crimes at Common Law, 6 CAMBRIDGE L.J. 31 (1936).

n171 G. WILLIAMS, supra note 53, at 31.

n172 G. FLETCHER, supra note 116, at 499-50 (footnote omitted).

n173 Williams has acknowledged that the legislature might well take into account the moral quality of conduct in deciding whether to prohibit such conduct. G. WILLIAMS, supra note 53, at 50. Thus, he regards the morality of prohibited conduct as relevant to legislative concerns, but not to the application of those concerns in judicial proceedings; accord H.L.A. HART, supra note 2, at 39.

n174 J. HALL, supra note 8, at 103. Kadish has made a similar point:

That the absence of the mens rea in this special sense of the required mental state, precludes liability in all of these cases is of course the merest tautology. This is the way these crimes are defined. But it is important to see that they are so defined because the special mens rea element is crucial to the description of the conduct we want to make criminal. And description is crucial in so far as it is regarded as important to exclude from the definition of criminality what we do not want to punish as criminal.

Kadish, supra note 2, at 274. Thus, in contrast to Williams' view that the moral quality of conduct is exclusively a legislative matter, Hall and Kadish contend that it also relates directly to the courtroom inquiry into a defendant's criminal liability; see also Packer, Mens Rea and the Supreme Court, 1962 SUP. CT. REV. 107, 109 (punishing conduct without reference to actor's state of mind is unjust and inefficient).

n175 See supra notes 130-36 and accompanying text.

n176 Consider the mens rea among these types of common law homicide, from first-degree murder -- requiring specific intent to kill, premediation, deliberation, and malice -- down through involuntary manslaughter, which may be based on criminal negligence. The differences in grades and punishments of these offenses reflect a judgment that higher grades of homicide represent greater disrespect towards legally protected interests. Of course, this analysis applies only to grading based on the mens rea element within a given category of offense. It does not apply to grading within a given category based on other, nonmental factors, such as the difference between grand and petit larceny. This analysis does apply to the Model Penal Code's sophisticated "element analysis" system of mens rea, which recognizes a maximum of four mental elements for all categories of crimes (purpose, knowledge, recklessness, and negligence) and applies them to discrete elements of conduct, circumstances, and consequences. See MODEL PENAL CODE § 2.02 (Proposed Official Draft 1962) (defining the four mental elements); MODEL PENAL CODE § 2.02 commentary at 123 (Tent. Draft No. 4, 1955) (the kind of culpability should be established for each material element). For a thoughtful study of the element analysis approach to mens rea, see Robinson & Grall, Element Analysis in Defining Criminal Liability: The Model Penal Code and Beyond, 35 STAN. L. REV. 681 (1983).

n177 For example, as Sayre noted, "[r]obbery and rape neccessitate a design; they can not possibly be committed through mischance." Sayre, supra note 162, at 988.

n178 The focus of the interpretive theory on the meaning of a defendant's conduct -- using mental elements as indicia of meaning rather than as mere subjective contents of a defendant's mind -- mirrors the focus of modern hermeneutic philosophy and psychology on the meaning of an actor's conduct. See supra note 132 (discussing emphasis of hermeneutic philosophy on an intersubjective approach, rejecting strict subjective or behaviorist theories). Fingarette and Hasse have proposed an analysis very similar to the one presented here. They also take note of the capacity to respond to legally and morally relevant influences, using the crime of murder to illustrate their point:

The universality of this condemnation of murder -- what shall here be referred to as the community perception of murder as a wrong of malum in se -- reflects a crucial fact. Among members of the community, in spite of their many differences, there is a certain nexus of basic feelings, customs, values, and social relationships, and this nexus constitutes the valuational background out of which emerges the common response to murder. Some individuals respond more to the pragmatic dimensions of this shared background-nexus, others more to the considerations of principle, others more to direct humane feeling, and still others respond very largely out of habit and custom. Some respond keenly and sensitively; others dully. In the upshot, however, without some such minimum shared background-nexus of basic perceptions and values, which provide the basic standards relevant to criminal mala in se, there would, of course, be no community. This is the background-nexus against which the community views and sees as wrong such acts as assault, rape, arson, and theft.

The basic factual presumption, then, that underlies our expectation that people should not deliberately kill, is that each individual in the community shares in a practical awareness of this background-nexus of basic perceptions and basic values; and against this background, killing a human being is perceived as, broadly speaking, gravely criminal unless certain kinds of excuse or justification are present.

H. FINGARETTE & A. HASSE, supra note 95, at 224-25. Fingarette and Hasse also maintain that it is against that background-nexus that the various mental elements of crime operate to express a defendant's attitudes. For example, with regard to the significance of the element of "deliberation" in first-degree murder, they write:

In broadest terms, the presumption is that a deliberate, premeditated killing is the fully conscious affirmation of the act by one who has, in some fundamental way, the ability to perceive its criminal significance. In purposefulness arising out of such awareness lies the highest degree of culpability in regard to law prohibiting killing. Therein lies the essential ground for proclaiming that deliberation typically signals the gravest form of murder.

Id. at 226-27.

Fingarette and Hasse also state that excuses, such as the insanity defense, arise when a defendant lacks the capacity to respond to that background-nexus of influences. For example, in discussing a homicide committed by a person suffering from paranoid delusions, they note:

The paranoid knows that he is killing a person, knows that it is contrary to law, and still he deliberately and intentionally does it. Such knowledge, deliberation, and intent have lost their full significance as "mental elements" because the usual presumption of fact about the background of skills and attitudes is radically inapplicable here. These elements no longer reflect the ability to respond to legal norms that we usually presume; they do not reflect response-ability.

Id. at 229. They summarize their position on mens rea and excuses as follows:

In broadest terms, then, there is a deep and tacit presumption that an individual shares, at least to some minimal practical extent, in a background of basic concepts, perceptions, values, skills, and attitudes common to the members of the community. This is not at all the same as presuming that any individual agrees with, or responds in the same ways as, everyone else or even anyone else, or that all individuals always act in the same way or support the same specific policies or institutions. When these presumptions are valid, then we can simply assess conduct by asking: Did the person do that voluntarily? Was it done intentionally? Did the person know the particular facts material to that particular act as relevant to standards of law? A few basic questions tell all -- or at least enough for the purpose of the serious matters contemplated by law. For the intention is seen against a (presumed) background of shared understanding and skills and attitudes.

But when there is real doubt as to the existence or integrity of this background of skills, attitudes, and perceptions, then we cannot take for granted the usual crucial significance that these particular features of conduct have for assessing responsibility. It works the other way around as well: when, after reviewing the circumstances, we come to realize that we have serious doubt whether we can take a person's intent or knowledge or deliberation to be criminally significant in the usual way, that is the sign that we have come to see irrationality, come to see that somehow the background skills and attitudes are gravely awry. The discovery is typically not one of temporal sequence but of an emerging awareness simultaneously that the "mental elements" don't have their usual significance, and that the background mental processes are grossly flawed or deranged.

Id. at 231-32 (emphasis in original).

Finally, Fingarette and Hasse observe that in an insanity defense case tried in our criminal justice system, the jury makes a practical judgment about whether, given the possible discrepancies between a particular defendant's capacities to respond to legally and morally relevant factors and the capacities most people are presumed to have, it is reasonable to expect the defendant to have adhered to those factors:

The discrepancy between a defendant's actual capacities and those normally presumed must be assessed in the light of the specific circumstances of the particular act. The jury must decide whether the discrepancy is enough, and of suitable kind, to ascribe nonresponsibility. This requires getting a good grasp of the facts of the particular case and having a sense of the larger context. Against that background the policy aspect of the decision takes the form: Is it still reasonable, in the context of crime and punishment, for society to expect responsible conduct of such a person in such circumstances? -- Or is it more reasonable in this particular case to give up such otherwise normal expectations and deal with the problem in a different perspective?

Id. at 233.

n179 As Stephen observed:

This account of the nature of intention explains the common maxim which is sometimes stated as if it were a positive rule of law, that a man must be held to intend the natural consequences of his act. I do not think the rule in question is really a rule of law, further or otherwise than it is a rule of common sense. The only possible way of discovering a man's intention is by looking at what he actually did, and by considering what must have appeared to him at the time the natural consequences of his conduct.

2 J. STEPHEN, supra note 3, at 111. Although due process prohibits a court from phrasing this second-order rule as a presumption that shifts the burden of persuasion to the defendant, the rule may be used as an inference and possibly as a presumption that shifts only the burden of production to the defendant. Francis v. Franklin, 105 S. Ct. 1965, 1971 (1985); Sandstrom v. Montana, 442 U.S. 510, 517-19 (1979); United States v. Love, 767 F.2d 1052, 1059-60 (4th Cir. 1985). This result confirms the jury's role as primary interpreter of a defendant's conduct. Sandstrom, 442 U.S. at 523 (quoting United States v. United States Gypsum Co., 438 U.S. 422, 446 (1978)).

n180 As in the case of the standard jury instruction about intent, due process prohibits a court from phrasing this second-order rule about malice as a presumption that shifts the burden of persuasion to the defendant, but the rule may be used as an inference and possibly as a presumption that shifts only the burden of production to the defendant. Rose v. Clark, 106 S. Ct. 3101 (1986); Davis v. Allsbrooks, 778 F.2d 168 (4th Cir. 1985); State v. Reynolds, 307 N.C. 184, 297 S.E.2d 532 (1982); Hankerson v. North Carolina, 288 N.C. 632, 220 S.E.2d 575 (1975), rev'd on other grounds, 432 U.S. 233 (1977).

The view presented here of inferred malice as a second-order interpretive rule may well be controversial because of the peculiar historical development of the concept of malice in homicide cases. The problem stems from the fact that over the course of centuries, common law judges have shaped malice in homicide cases into a concept often identified by its absence rather than by its presence. Fletcher, in tracing this history, has written:

The critical step in the conceptual evolution of malice is MacKally's Case. That early 17th century decision, as reported and interpreted by Coke, stands for the principle that the prosecution need not prove the element of malice to convict of murder. The judges realized that malice does not lend itself to affirmative proof; by and large, the malicious killing is defined by reference to what it is not, not by what it is. As agreed by all, one type that was not malicious was a killing provoked by a sudden quarrel. Thus, to have a triable issue of malice, one had to have a triable claim that the defendant killed in the course of a sudden quarrel.

fletcher, Two Kinds of Legal Rules: A Comparative Study of Burden-of-Persuasion Practices in Criminal Cases, 77 YALE L.J. 880, 905 (1968) (citing 9 Co. Rep. 656, 77 Eng. Rep. 828 (1611)); see also R. PERKINS & R. BOYCE, supra note 8, at 75-78 (describing development of rebuttable presumption that every homicide is committed with malice aforethought as prosecution-serving device).

The history of malice in homicide cases raises the question of whether the concept has any positive content or whether it is strictly a negative concept denoting a set of exceptions to charges of first- or second-degree murder. For example, some observers have contended that over time, English law stripped malice of positive meaning. See, e.g., Michael & Wechsler, A Rationale of the Law of Homicide, 37 COLUM. L. REV. 701, 707 & n.21 (1937) (term reduced to signifying neither malice nor forethought in the popular sense). That position is an example of H.L.A. Hart's concept of mens rea as a negative, defeasible principle: a person who voluntarily violates the criminal prohibition against homicide is guilty of murder unless he presents evidence of certain excusing or mitigating circumstances, such as absence of malice. Hart, Ascription of Responsibility and Rights, 49 ARISTOTELIAN SOC'Y PROC. 171 (1949), reprinted in FREEDOM AND RESPONSIBILITY, supra note 1, at 143, 145. Other writers, however, contend that malice still has positive content. For example, Perkins and Boyce regard malice as a "person-endangering-state-of-mind" that includes intent to kill, intent to inflict great bodily injury, intent to do an act in wanton and wilful disregard of an unreasonable human risk, and intent to perpetrate a dangerous felony. R. PERKINS & R. BOYCE, supra note 8, at 73. Each of these four types of "person-endangering-states-of-mind" represents disrespect for the victim's interests in life and safety, though to a degree greater than the disrespect involved in either voluntary or involuntary manslaughter. See also State v. Reynolds, 307 N.C. 184, 191, 297 S.E.2d 532, 536 (1982) (dividing malice in homicide cases into express and implied forms).

To the extent that malice has such positive content, the familiar inference in jury instructions about malice, cited in the text, can be viewed as a second-order interpretive rule that identifies common cases in which malice is likely to be present. Indeed, even the familiar concept that malice may be negated by evidence that the act was committed in the heat of passion can be regarded as a second-order interpretive rule, one that identifies common situations in which malice may be absent, assisting a jury resolve the issue in a particular case. In law as in everyday life, the articulation and identification of a given concept can be derived from either positive or negative reasoning. See, e.g., United States v. Frady, 456 U.S. 152, 171-74 (1982) (upholding a jury's finding of malice through both positive and negative analysis). As Austin observed in the passage quoted in the text accompanying note 1 supra, there are circumstances in which a concept, such as malice, can best be understood by highlighting what it does not include. The concept of malice is such an example. Thus, malice can best be seen as a concept with genuine meaning that a jury can assess with the aid of positive or negative rules of interpetation.

n181 See generally W. LaFAVE & A. SCOTT, CRIMINAL LAW 179-81 (1972) (discussing voluntary requirement for criminal liability in case law); R. PERKINS & R. BOYCE, supra note 8, at 605-11 (analyzing various distinctions raised about sufficiency of attempted acts); MODEL PENAL CODE § 2.01(1) (Proposed Official Draft 1962) (basis for liability requires voluntary act); MODEL PENAL CODE § 2.01(1) commentary at 119-20 (Tent. Draft No. 4, 1955) (law cannot hope to deter involuntary conduct). In this article I will not address the issue of criminal liability for omissions. On that issue, see generally W. LaFAVE & A. SCOTT, supra, at 182-91 (survey of actual requirements and trends in creation of duties to act); R. PERKINS & R. BOYCE, supra note 8, at 658-80 (extensive analysis of potential duties to act recognized by cases, statutes, and scholars); MODEL PENAL CODE § 2.01(1), (3) (Proposed Official Draft 1962) (alternate basis for liability is omission of legally required act one is physically capable of performing); MODEL PENAL CODE § 2.01(3) commentary at 123 (Tent. Draft No. 4, 1955) (duty to perform must be imposed by law before omission will count as voluntary act).

n182 R. PERKINS & R. BOYCE, supra note 8, at 610; accord W. LaFAVE & A. SCOTT, supra note 181, at 180.

n183 O. HOLMES, supra note 164, at 54.

n184 W. LaFAVE & A. SCOTT, supra note 181, at 179-81; R. PERKINS & R. BOYCE, supra note 8, at 609-10; MODEL PENAL CODE § 2.01(2) (Proposed Official Draft 1962).

n185 H.L.A. HART, supra note 2, at 107.

n186 W. LaFAVE & A. SCOTT, supra note 181, at 179-81; R. PERKINS & R. BOYCE, supra note 8, at 609-11; 2 P. ROBINSON, supra note 8o, at 262; MODEL PENAL CODE § 2.01(2) (Proposed Official Draft 1962). We might find a defendant guilty if we conclude that he wrongly put himself in a situation in which his known disability was likely to cause harm. See, e.g., People v. Decina, 2 N.Y.2d 133, 139, 138 N.E.2d 799, 803, 157 N.Y.S.2d 558, 564-65 (1956) (in case of criminally negligent vehicular homicide, epileptic defendant's decision to drive alone, while aware that he was subject to epileptic seizures at any time, manifests disregard for consequences and indifference to others).

n187 For a discussion of the differences involved in treating involuntariness as the absence of an essential element of an offense, rather than as a defense (including the impact on burdens of production and persuasion), see A. GOLDSTEIN, supra note 3, at 203-04 (treating involuntariness as lack of essential element raises mental illness issue without risking commitment); 2 P. ROBINSON, supra note 83, at 263-74 (voluntary act "almost universally treated as required element of every offense"); see also State v. Caddell, 287 N.C. 266, 289, 215 S.E.2d 348, 363 (1975) (burden of proof on defendant to prove insanity or "temporary mental lapse due to concussion, somnolentia, epilepsy or the like" as affirmative defense).

n188 The form of analysis used here, beginning with the typical case of expressive conduct and then addressing the exceptional cases, corresponds to the "structured" form of legal reasoning recently articulated by Fletcher. See Fletcher, The Right and the Reasonable, 98 HARV. L. REV. 949, 950-64 (1985) ("Structural legal discourse proceeds in two stages: first, an absolute norm is asserted; and second, qualifications enter to restrict the scope of the supposedly dispositive norm."); see also Gadamer, supra note 114, at 141:

[E]very law admits of a certain internal tension with respect to the concrete possibilities of action: a law is always general and can never address itself to all the concrete complexities of a particular case. (Let us note in passing that this is the original problem of juridical hermeneutics.) A law is always insufficient, not by reason of any intrinsic fault, but because the practical world as the field of our actions is always imperfect in comparison to the ideal order envisioned by laws.

The interpretive theory of mental elements of crime and excuses also offers a synthesis of the two categories of mens rea distinguished by Kadish, by giving them a common theoretical basis. See supra note 129. Mental elements functioning as general rules of interpretation constitute Kadish's first, "special sense" of mens rea. The excuses of mistake and accident, which can negate these elements in some cases, operate within this special sense of mens rea. Other excuses, such as insanity, duress, and infancy, function as exceptions to these general rules and constitute Kadish's second category of mens rea, that of legal responsibility. There are differences between these two senses: one sense serves as the set of general rules, while the other serves as the set of exceptions to those rules. Nevertheless, they are conceptually linked in that they operate together as the criminal law's system for normative interpretation of behavior. Moreover, the excuses that fall within these separate categories of mens rea are conceptually linked in that they represent cases in which we do not interpret a defendant's conduct to manifest the disrespect condemned by criminal law. The interpretive theory also rebuts Hart's contention that excuses are not intrinsic factors in guilt determination. See H.L.A. HART, supra note 2, at 35-53; Hart, supra note 180. Although excuses function as exceptions to the general interpretive rules of mens rea elements of crime, they play a vital role in the same purpose served by mens rea elements: interpreting conduct.

The nature of the insanity defense as an exception to the general interpretive rules of mens rea bears directly on the allocation of the burden of production on insanity. Because the defense functions as an exception to the general rules, it would be appropriate under the interpretive theory to impose the burden of production on the defendant.

The outcome is not as clear regarding allocation of the burden of persuasion on insanity. Considered apart from other concerns, the interpretive theory suggests that the prosecution should bear that burden. Even though the insanity defense functions only as an exception to the general interpretive rules of mens rea, once it is raised in a case it is as important as mens rea states of mind in assessing the meaningfulness of a defendant's conduct. To the extent that the prosecution should carry the burden of persuasion on all issues essential to the determination of guilt, see In re Winship, 397 U.S. 358, 364 (1970) (proof of all essential facts beyond reasonable doubt is constitutionally required for criminal conviction); A.B.A. REPORT, supra note 39, at 381 ("prosecution should generally bear the burden of persuading the factfinder beyond a reasonable doubt of all facts which are regarded as necessary to establish an ethically adequate predicate for criminal liability"), so too should the burden of persuasion for the insanity defense -- as for all mens rea issues -- be imposed on the prosecution. This view is more consistent with due process than the approach taken in Patterson v. New York, 432 U.S. 197 (1977), where the burden of persuasion was mechanically allocated according to legislative manipulation of elements. In Patterson, the Court held preponderance of the evidence, the affirmative defense of extreme emotional disturbance. Id. at 210. (The Court did acknowledge, in dictum, that due process concerns could preclude excessive legislative manipulation of the burden of persuasion. Id. at 210.)

Significant practical concerns might legitimately outweigh the argument made above, however, warranting imposition of the burden of persuasion on the defendant. These concerns include the difficulty of assessing the insanity defense in a particular case, the greater difficulty of proving sanity than insanity, the danger of fabrication, and the defendant's advantage in gaining access to relevant evidence. Bonnie, supra note 2, at 197; Morse, supra note 39, at 824-25; A.B.A. REPORT, supra note 39, at 382-83. The Supreme Court has thus far indicated that imposition of the burden of persuasion on the defendant concerning insanity does not violate due process. Leland v. Oregon, 343 U.S. 790 (1952) (Oregon statute requiring defendant to prove insanity beyond a reasonable doubt does not violate due process); Rivera v. State, 351 A.2d 561 (Del. 1975) (Delaware statute classifying mental illness as affirmative defense that must be proved by preponderance of the evidence does not violate due process), appeal dismissed, 429 U.S. 877 (1976). For a probing discussion of constitutional issues in allocating the burden of persuasion, see Jeffries & Stephan, Defenses, Presumptions, and Burden of Proof in Criminal Law, 88 YALE L.J. 1325 (1979); see also 1 P. ROBINSON, supra note 83, at 41-54, 138-48.

n189 E.g., H. GROSS, supra note 2, at 103-13; W. LaFAVE & A. SCOTT, supra note 181, at 204-07; R. PERKINS & R. BOYCE, supra note 8, at 926-32; 2 J. STEPHEN, supra note 3, at 120-21; G. WILLIAMS, supra note 53, at 31.

n190 R. PERKINS & R. BOYCE, supra note 8, at 928.

n191 Difficult cases can arise that appear initially to raise the issue of good motive as a defense, but on closer analysis may not. For example, we might say that a man who steals a loaf of bread to feed his starving children should be acquitted of theft because he acted with good motive. This example, however, also could be analyzed as a case of necessity in which the urgent need to feed his children made the father so desperate that he could not resist the impulse to steal. The emergency nature of his circumstances operates in a manner similar to the threat in a case of duress, imposing emotional pressure on the father that we do not expect him to resist. For discussions of whether to treat necessity as an excuse or as a justification based on a choice of the lesser of two evils, see United States v. Bailey, 585 F.2d 1087, 1097-98 (D.C. Cir. 1978), rev'd, 444 U.S. 394 (1980); G. FLETCHER, supra note 116, at 818-29; W. LaFAVE & A. SCOTT, supra note 181, at 381-88; R. PERKINS & R. BOYCE, supra note 8, at 1065-72; Perkins, Impelled Perpetration Restated, 33 HASTINGS L.J. 403, 416-25 (1981).

n192 E.g., H.L.A. HART, supra note 2, at 132-57; Fletcher, The Theory of Criminal Negligence: A Comparative Analysis, 119 U. PA. L. REV. 401 (1971); Hall, Negligent Behavior Should Be Excluded from Penal Liability, 63 COLUM L. REV. 632 (1963); Turner, supra note 170, at 38-48.

n193 As Hart has written:

What is crucial is that those whom we punish should have had, when they acted, the normal capacities, physical and mental, for doing what the law requires and abstaining from what it forbids, and a fair opportunity to exercise these capacities. Where these capacities and opportunities are absent, as they are in different ways in the varied cases of accident, mistake, paralysis, reflex action, coercion, insanity, etc., the moral protest is that it is morally wrong to punish because "he could not have helped it" or "he could not have done otherwise" or "he had no real choice." But, as we have seen, there is no reason (unless we are to reject the whole business of responsibility and punishment) always to make this protest when someone "who just didn't think" is punished for carelessness. For in some cases at least we may say "he could have thought about what he was doing" with just as much rational confidence as one can say of any intentional wrongdoing "he could have done otherwise."

H.L.A. HART, supra note 2, at 152 (emphasis in original); accord H. GROSS, supra note 2, at 419-23 (negligent defendant liable because intention to do harm is not necessary element of criminal liability and he should be held responsible for consequences of conscious course of conduct); Fletcher, supra note 192 (criminal law should not hesitate to punish negligence).

n194 As discussed supra note 132 and accompanying text, an actor's understanding of the meaning of his own conduct is relevant to, but not necessarily determinative of, its meaning in the eyes of others.

n195 Fletcher has written:

Now it is not clear why someone would insist that there is no accountability for a breach of duty unless the actor chooses the breach. Jerome Hall argues that the breach is not "voluntary" unless it is consciously chosen. But this cannot be right. Voluntariness is not synonymous with choice. An act or failure to act is voluntary only if the actor could have done otherwise under the circumstances. An intentional act is voluntary if the actor could have abstained. A mistake is voluntary if the actor could have avoided the mistake. Inadvertent risk-taking is voluntary if, under the circumstances, the actor could have realized the risks implicit in his conduct. It is true that the conscious and intentional actor realizes at every moment that he might desist, whereas the mistaken or inadvertent actor does not repeatedly forgo the opportunity of correcting himself. This might explain why intentional wrongdoing is regarded as more culpable than negligent wrongdoing, but it does not follow that blaming inadvertent risk-taking is never permissible.

G. FLETCHER, supra note 116, at 710-11.

n196 B. WOOTTON, CRIME AND CRIMINAL LAW (1963).

n197 J. FEINBERG, supra note 109, at 110-13; H.L.A. HART, supra note 2, at 193-209; H. PACKER, supra note 2, at 121-31; Kadish, supra note 2, at 285-90; Packer, supra note 174, at 109. But see H. GROSS, supra note 2, at 357-74 (proposing "mild" form of strict liability applicable in limited circumstances).

n198 See generally G. FLETCHER, supra note 116, at 713-36; H. PACKER, supra note 2, at 125-31. The Supreme Court has not decisively addressed the constitutionality of strict criminal liability. Compare United States v. Dotterweich, 320 U.S. 277 (1943) (upholding conviction for strict liability regulatory offense without considering constitutionality of strict liability) with Lambert v. California, 355 U.S. 225 (1957) (reversing conviction for violating ordinance prohibiting convicted felons from entering city without registering with police, on ground that inadequate notice violates due process) and Morissette v. United States, 342 U.S. 246 (1952) (reversing conviction for theft of government property on ground that criminal intent is essential element of crime even though not included in statute by Congress). Regarding the meager history of Supreme Court analysis of the constitutionality of strict liability offenses, see Packer, supra note 174, at 110-37 (Supreme Court has yet to deal adequately with mens rea in strict liability criminal cases).

n199 Note also that as a practical matter, Lady Wootton's proposal would not remove mens rea from all consideration in criminal proceedings. Rather, it would merely shift the occasion for its consideration from trial to disposition. B. WOOTTON, supra note 196. A sentencing judge would encounter the same difficulties now faced by a jury sifting through evidence about a defendant's mental state.

n200 As Hart has observed, strict liability treats people as manipulable objects rather than as human subjects who interact in a rich world of meaning. H.L.A. HART, supra note 2, at 182-83.

n201 See generally H. GROSS, supra note 2, at 385-400; H.L.A. HART, supra note 2, at 1-27; P. LOW, J. JEFFRIES & R. BONNIE, supra note 2, at 1-30; J. MICHAEL & H. WECHSLER, CRIMINAL LAW AND ITS ADMINISTRATION 4-17 (1940); H. PACKER, supra note 2, at 35-61; SENTENCING (H. Gross & A. Von Hirsch eds. 1981).

n202 Hampton, supra note 128, at 211-12.

n203 Id. at 212 (emphasis in original).

n204 Id. at 226.

n205 Id. Hampton unduly limits the range of possible criminal conduct by characterizing it as acts committed in order to satisfy the offender's own needs and interests. This view appears to exclude negligence as a ground of criminal liability and unnecessarily suggests a concern about a defendant's motive. This article proposes instead that criminal conduct consists of acts that express disrespect for legally protected interests.

n206 We could develop this analysis further by regarding the jury not only as the community's interpreter, but also as an expressive agent of the community. The jury plays an expressive role by declaring publicly, through a verdict, its position on the defendant's blameworthiness. Thus, the jury both interprets meaning in a defendant's conduct and expresses its view of whether the defendant acted with appropriate respect for legally protected interests. The sentencing judge concludes the jury's "statement" to a guilty defendant by imposing a sentence.

n207 It is appropriate at this point to express my indebtedness to Jean Hampton's article in developing the interpretive theory of excuses and mental elements of crime proposed in this article. By attuning me to the expressive role of punishment, Hampton's essay prompted me to consider the object of that communication. That is, if punishment is a form of expression that conveys particular meaning to an offender, then it might be an expressive response to meaning conveyed by and interpreted in the offender's conduct. For other helpful discussions of the expressive role of punishment, see J. FEINBERG, supra note 109, at 95-118; P. LOW, J. JEFFRIES & R. BONNIE, supra note 2, at 3-4; W. MOBERLY, THE ETHICS OF PUNISHMENT 201-25 (1968); R. NOZICK, PHILOSOPHICAL EXPLANATIONS 363-97 (1981); Hart, supra note 116, at 404; Morris, A Paternalistic Theory of Punishment, 18 AM. PHIL. Q. 263 (1981).

n208 Punishment is a form of communication to other people as well as to an offender. It can serve the purpose of general deterrence by conveying to others that certain conduct is wrongful and by warning them of the consequences of crime. It can also express to a victim both its recognition that the criminal has shown disrespect for the victim's interests and the community's commitment to preserve those interests. Finally, punishment expresses to other members of the community its self-image as a society that places great value on the preservation of designated interests.

n209 The interpretive view of the insanity defense and the expressive function of punishment together suggest that the proper verdict for a defendant to whom the insanity defense applies would be an outright acquittal. Any disposition other than complete freedom would be treated as a separate issue considered during a separate involuntary commitment proceeding. Although some jurisdictions provide for automatic commitment after an insanity acquittal, others require a separate, posttrial proceeding on the question of commitment -- the same type of proceeding required for all involuntary commitments. See generally 2 P. ROBINSON, supra note 83, at 305-09; Note, Commitment Following an Insanity Acquittal, 94 HARV. L. REV. 605 (1981).

Use of a separate commitment proceeding meets the criticism made by some observers that the real purpose of the insanity defense is not to carve out a just exception to criminal liability, but rather to legitimize the indeterminate restraint of a special class of people who otherwise would walk free after an acquittal. See Goldstein & Katz, Abolish the "Insanity Defense" -- Why Not?, 72 YALE L.J. 853, 865 (1963) (real problem is how to impose sanctions on insane defendants acquitted under general principles of criminal law). As the Supreme Court observed in Jones v. United States, 463 U.S. 354 (1983), however, automatic commitment procedures rest on the reasonable inference that a defendant acquitted by reason of insanity, because of a mental disorder he suffered at the time of his criminal conduct, probably remains mentally ill and dangerous at the time of verdict and disposition. Thus, while guilt determination and disposition theoretically are distinct inquiries directed to the defendant's mental state at two different times, they can be combined as a practical matter in one proceeding. Addressing the problem of potentially different burdens of persuasion for acquittal by reason of insanity and involuntary commitment, the Court concluded that insanity acquittees are distinguishable from those involuntarily committed because they have committed criminal acts and have already raised their own mental disorder as a defense. Id. at 366. Robinson has written a helpful summary of the controversy:

In evaluating whether a particular commitment scheme strikes the best, or even a reasonable balance between the interests of an insanity acquittee and those of society, one must evaluate the significance of an acquittal by reason of insanity. All insanity defenses require the fact-finder to be sufficiently persuaded that the defendant suffered from a mental disease or defect at the time of the offense. A successful insanity defense does not necessarily ential a finding as to the defendant's present mental state.

As long as the criminal justice system is premised upon the principle that punishment is justified only where there is guilt, no future restraint of the acquitted defendant can be justified as punishment. It follows that this verdict does not justify special treatment of those acquitted by reason of insanity unless there are other implications of such a verdict that are relevant to the civil commitment standard of "present dangerousness and insanity." There are. The defendant's demonstrated dangerousness, as established by his satisfying the objective elements of an offense, is significant in evaluating his present danger to society. The fact that a jury sufficiently believed or found that the defendant suffered from a mental disability that played a substantial role in bringing about the harm he caused, gives rise to legitimate concern as to whether that condition continues.

Jurisdictions employing mandatory commitment procedures apparently give conclusive weight to findings of past dangerousness and insanity, and ignore the fact that the verdict of not guilty by reason of insanity says nothing about the defendant's present mental state. They risk the appearance of imposing punishment despite a verdict of not guilty. On the other hand, jurisdictions that judge insanity acquittees under standards identical to those used for all others in civil commitment, apparently give no special weight to the fact that one acquitted by reason of insanity has manifested an abnormality that posed and may continue to pose a real threat to societal interests.

2 P. ROBINSON, supra note 83, at 308; see also A. GOLDSTEIN, supra note 3, at 143-46 (defendant is regarded as mentally ill and in need of treatment after acquittal by reason of insanity); W. LaFAVE & A. SCOTT, supra note 181, at 317-20 (most jurisdictions commit acquitted defendant following trial court's determination that defendant's insanity continues or he is dangerous).

n210 See A. GOLDSTEIN, supra note 3, at 194-202; 1 P. ROBINSON, supra note 83, at 272-85, 474-78; Arenella, supra note 156, at 828-29; Bonnie & Slobogin, supra note 2, at 446; Mandiberg, Protecting Society and Defendants Too: The Constitutional Dilemma of Mental Abnormality and Intoxication Defenses, 53 FORDHAM L. REV. 221, 221-28 (1984); A.B.A. REPORT, supra note 39, at 340-45; MODEL PENAL CODE § 4.02 (Proposed Official Draft 1962); MODEL PENAL CODE § 4.02 commentary at 193 (Tent. Draft No. 4, 1955).

n211 United States v. Brawner, 471 F.2d 969, 1002 (D.C. Cir. 1972) (en banc); 1 P. ROBINSON, supra note 83, at 276; Bonnie & Slobogin, supra note 2, at 446, 479; Mandiberg, supra note 210, at 223, 233; A.B.A. REPORT, supra note 39, at 341; MODEL PENAL CODE § 4.02 commentary at 193 (Tent. Draft No. 4, 1955).

n212 Battalino v. People, 118 Colo. 587, 199 P.2d 897 (1948); 1 P. ROBINSON, supra note 83, at 280-83; A.B.A. REPORT, supra note 39, at 342.

n213 1 P. ROBINSON, supra note 83, at 475-78; Arenella, supra note 156, at 829-30. No American jurisdiction currently recognizes diminished responsibility as a defense. England adopted it as a partial defense to murder in 1957. See Homicide Act of 1957, 5 & 6 Eliz. 2, ch. 11, pt. I, § 2. The California courts once expanded the partial defense of diminished capacity to include a form of diminished responsibility as a partial defense. E.g., People v. Conley, 64 Cal. 2d 310, 322, 411 P.2d 911, 918, 49 Cal. Rptr. 815, 822 (1966); People v. Wolff, 61 Cal. 2d 795, 821, 394 P.2d 959, 975, 40 Cal. Rptr. 271, 287 (1964); People v. Gorshen, 51 Cal. 2d 716, 735-36, 336 P.2d 492, 504 (1959). The California legislature, however, abolished the defense in 1983. CAL. PENAL CODE § 189 (West Supp. 1983).

n214 See sources cited supra note 210.

n215 Arenella, supra note 156, at 849, 856-57.

n216 Bethea v. United States, 365 A.2d 64, 87-91 (D.C. 1976), cert. denied, 433 U.S. 911 (1977); Steele v. State, 97 Wis. 2d 72, 96-97, 294 N.W.2d 2, 13 (1980); Arenella, supra note 156, at 857-58; Mandiberg, supra note 210, at 237-50.

n217 Wechsler wrote: "The category [of nonresponsible defendants] must be so extreme that to the ordinary man, burdened by passion and beset by large temptations, the exculpation of irresponsibles bespeaks no weakness in the law." Wechsler, supra note 73, at 374; see also Bethea, 365 A.2d at 90 (D.C. 1976) (diminished capacity rule would eliminate means "whereby society is enabled to protect itself from those who cannot or will not conform their conduct to the requirements of the law"); G. FLETCHER, supra note 116, at 514 ("law functions by setting a minimum threshold of accountability"); A. GOLDSTEIN, supra note 3, at 206-07 (use of defense would require new methods of controlling dangerous person after acquittal); Arenella, supra note 156, at 850-53 (defense raises fear of releasing into society acquitted defendants who continue to be dangerous); Mandiberg, supra note 210, at 250-52 (diminished responsibility defense reduces defendant's punishment solely on grounds of decreased culpability, regardless of danger he may pose to society).

n218 See supra note 176 and accompanying text.

n219 Note that I am not advocating recognition of these partial defenses on the basis of the interpretive theory. Instead, I am simply explaining how these defenses would fare under the interpretive theory if all considerations of policy and practicality mentioned above were ignored. Courts and legislatures may legitimately decide that those considerations outweigh the appeal of the interpretive theory on this issue.

n220 See supra notes 196-200 and accompanying text.

n221 Id. Courts in three cases early in this century held that abolition of the insanity defense is unconstitutional, even without elimination of mens rea elements. State v. Lange, 168 La. 958, 123 So. 639 (1929); Sinclair v. State, 161 Miss. 142, 132 So. 581 (1931) (per curiam); State v. Strasbourg, 60 Wash. 106, 110 P. 1020 (1910). However, a court in a recent case held that abolition of the insanity defense is constitutional where a defendant could still present psychiatric evidence to negate a required mental state. State v. Korell, 690 P.2d 992 (Mont. 1984).

n222 See N. MORRIS, MADNESS AND THE CRIMINAL LAW 53-76 (1982). Morris has based this proposal on several grounds: (1) the total absence of freedom of choice purportedly required for applying the insanity defense rarely occurs; (2) the law should not give mental illness special exculpatory effect while ignoring the effects of other pressures, such as poverty, on human conduct; and (3) the insanity defense ignores many defendants who need psychiatric treatment, but who do not qualify for the defense. Id. This limited abolitionist approach has been adopted in Idaho, Montana, and Utah. See IDAHO CODE § 18-207 (Supp. 1985); MONT. CODE ANN. § 46-14-102 (1985); UTAH CODE ANN. § 76-2-305(1) (Supp. 1985). As mentioned supra note 221, at least one court has held this limited form of abolition to be constitutional. State v. Korell, 690 P.2d 992 (Mont. 1984).

n223 See the discussion of Hadfield's Case, 27 How. State Tr. 1281 (Eng. 1800), supra notes 27-28 and accompanying text.

n224 Bonnie, supra note 2, at 195-96; Morse, supra note 39, at 801-02; A.B.A. REPORT, supra note 39, at 327.

n225 A.B.A. REPORT, supra note 39, at 327 (emphasis in original).

n226 See supra note 175-88 and accompanying text.

n227 As a matter of intellectual pedigree, the position of limited abolitionists like Morris might be traced to the conventional view that mens rea, as denoting the mental elements of crime, is conceptually distinct from the issue of responsibility invoked by the excuse defenses of insanity, duress, and infancy. See supra note 129. For example, in reference to Hadfield's Case, Morris has written: "There was no doubt of Hadfield's brain damage nor of his gross psychological disturbance. He did, however, clearly intend to kill the king. He had the insane mens rea for murder and treason. I do not regard the phrase 'insane mens rea' as a contradiction in terms." Morris, Psychiatry and the Dangerous Criminal, 41 S. CAL. L. REV. 514, 521 (1968).

Once we label negativing defenses as "intrinsic" to mens rea and the excuse defenses of insanity, duress, and infancy as "extrinsic" to mens rea, it requires only a small analytic step to eliminate the extrinsic defense of insanity while retaining the intrinsic negativing defenses. Such a position might appear to preserve intact the basic framework of mens rea, while eliminating the troublesome issue of insanity. This limited surgery, however, ignores the common theoretical root of both types of defenses in the meaning-fulness of conduct and the important role of extrinsic defenses -- including insanity -- in supplementing the general interpretive rules embodied in mens rea elements of crime.

n228 For a list of these jurisdictions, see 2 P. ROBINSON, supra note 83, at 309-10 n.93.

n229 See Pouncey v. State, 297 Md. 264, 269, 465 A.2d 475, 478 (1983); Langworthy v. State, 282 Md. 588, 593-94, 399 A.2d 578, 581-82 (1979); 2 P. ROBINSON, supra note 83, at 310.

n230 See, e.g., Hart v. State, 702 P.2d 651 (Alaska Ct. App. 1985); ALASKA STAT. §§ 12.47.010, 12.47.030, 12.47.040 (1984); ILL. ANN. STAT. ch. 38, § 6-2 (Smith-Hurd Supp. 1985); UTAH CODE ANN. § 77-35-21(a) (Supp. 1985). There are strong arguments, however, against this type of verdict: the risk of encouraging morally muddled, compromise verdicts, and the problem of delegating the issue of disposition to a jury. See 2 P. ROBINSON, supra note 83, at 311-13; Arenella, Reflections on Current Proposals to Abolish or Reform the Insanity Defense, 8 AM. J.L. & MED. 271, 277 (1982); Bonnie, supra note 2, at 194; Morse, supra note 39, at 803-04.

n231 See 2 P. ROBINSON, supra note 83, at 310 n.95.

n232 Note that historically, neither criminal law nor tort law recognized strict liability. In both bodies of law, a defendant generally may not be held liable unless he has some capacity to avoid injuring legally protected interests. Of course, there are differences in the nature of fault prohibited in the two areas of law. Moreover, the threshold of prohibited fault is lower in tort law (ordinary negligence) than in criminal law (gross negligence). See W. PROSSER & W. KEETON, THE LAW OF TORTS 534-38 (5th ed. 1984). Infancy and insanity do not uniformly function as defenses to tort liability. Id. Moreover, the scope of strict tort liability has increased markedly in recent years. Id.

A more basic difference between tort law and criminal law is the purpose for which they require fault as a condition of liability. As this article has discussed, criminal law requires fault as a condition for the community's imposition of blame and punishment. In contrast, tort law (except in cases involving punitive damages) requires fault as the criterion for deciding whether a defendant or a victim should bear the financial burden of injury to legally protected interests. Id. at 7-15. Moreover, the degree of fault directly affects the degree of a criminal defendant's liability and punishment, while the degree of fault above ordinary negligence has no bearing on a tortfeasor's financial liability (again, except in cases involving punitive damages). Despite these differences, it remains true that tort law generally conditions liability on some degree of fault. Although a full study of the relevance of the interpretive theory to tort law is beyond the scope of this article, the importance of fault in tort law suggests that to some extent, our society regards the commission of torts and the public judicial procedure for private redress of resulting injury as forms of normative communication.