+ Guide Bar +
Legal Professional and Scholars

Laws and Regulations

Legal News
Law Study
About the Center


But the problem with the egalitarian justification runs much deeper. A procedural regime based on whether the government is a litigant makes sense only if we can identify certain characteristics, bearing significantly and systematically on the litigation, that distinguish the government qualitatively from all other types of litigants. What, then, makes litigating against the government uniquely different from litigating against a private party? The usual answer is that the government possesses numerous qualities that enable it to fare considerably better than any other party to litigation: it has vast resources and ample experience, and it is also the entity that sets the basic rules of the game. There is, however, growing reason to doubt the uniqueness of the government as a litigant. The argument is not that the government enjoys no advantage in litigation--it most certainly does--but that there are other classes of litigants besides the government who also enjoy such an advantage.
In his seminal work, "Why the 'Haves' Come Out Ahead," Professor Marc Galanter famously set forth an analysis of the basic architecture of the American legal system and its structural limitations. [FN77] Galanter began by dividing the litigant world into two types: one-shotters ("OSs") and repeat players ("RPs"). One-shotters are those who have only occasional recourse to the courts--in an automobile accident, divorce proceedings, or a quarrel with a neighbor. OSs are usually individuals; they tend to have fewer resources, and they litigate for immediate outcomes. Repeat players, *103 by contrast, are litigants involved in similar litigation over time. [FN78] Generally speaking, they differ from OSs in important respects. They are usually institutions, tend to be relatively wealthy, and, due to their size and resources, are able to pursue long-run interests. Consequently, RPs are less concerned with the outcome of a particular case. [FN79] Accordingly, Galanter claims, "[w]e should expect an RP to play the litigation game differently from an OS." [FN80] This ability to play differently affords RPs substantial advantages that enable them to fare better in litigation. First, having been through many similar litigations, RPs are able to structure their next transactions and build a record. Second, because of their long-term involvement in litigation, RPs become skilled in the process, helped along, in no small part, by their ability to access specialists. Third, RPs have at their disposal economies of scale and enjoy low start-up costs for any case. Fourth, as they frequently make use of the system, RPs are able to develop helpful informal relationships with insiders and establish "bargaining reputation" within the system. Fifth, RPs can "play the odds." Given their size and resources, the stakes at risk for RPs in any given litigation are likely to be relatively small. Therefore, they can adopt strategies calculated to maximize gains over a long series of cases, even when this involves the risk of maximum loss in some cases. Sixth, RPs are well positioned to play not just for immediate gains, but also for the very rules of the game in both the legislative arena (through lobbying efforts) and within the litigation framework itself. [FN81] In short, according to Galanter, the exceptional advantages that give a significant edge to a particular party are not unique to the government and are enjoyed by other RPs as well. Many private entities, including financial institutions, insurance companies, and large corporations, can operate in the courtroom much the way the government does.
Subsequent empirical studies of trial and appellate courts have confirmed Galanter's basic findings. [FN82] Generally, these studies indicate *104 that classes of litigants with the greatest resources and the lowest relative stakes at risk in litigation have the highest rates of success in both trial and appellate courts. Accordingly, all RPs fare substantially better than individuals, whose chances of winning a case against an RP both at trial and on appeal are quite slim. [FN83] A *105 recent study examining the differences between individual and organizational litigants in the disposition of federal civil cases (the rates of trial, settlement, and nontrial adjudication) revealed that the gap in relative success rates between OSs and RPs has not diminished over the years--in fact, the difference has increased. [FN84] Individual-plaintiff cases litigated against organizational entities are considerably more likely to end in a nonfinal termination (including voluntary dismissal, transfer, and remand). Individuals are also more likely to abandon their cases and substantially less likely to obtain default judgments. [FN85] Finally, individual-plaintiff cases are less likely to survive pretrial motions to dismiss or for summary judgment. [FN86] The OSs' unimpressive rates of success in litigation reflect also on their ability to succeed in out-of-court negotiations with RPs, for lack of a credible a priori threat to initiate litigation. To convince the RP of the sincerity of its intention to go to trial, an OS must invest large sums of money from the outset. As a result, cases brought by individuals are costlier to settle and thus more likely to be adjudicated than organizational-plaintiff cases. [FN87]
This last point leads us to our final observation in this context. Although trials are becoming an endangered species, [FN88] we should still care a great deal about the rules of procedure governing in-court litigation. These rules affect not only court proceedings but also the bargaining process that occurs outside the courtroom. *106 Bargaining does not transpire in a vacuum; it takes place "in the shadow of the law." [FN89] Legal rules and procedures structure the bargain, governing what each party can expect to gain in litigation and giving each party certain bargaining chips--an endowment of sorts. [FN90] A simple example may be illustrative. Assume that in a dispute between an individual plaintiff and Microsoft, the rules of civil procedure give the defendant massive leverage in deposition and discovery, enabling it to conduct endless pretrial interrogations. [FN91] Assume also that the grounds for summary judgment are extensive and that, under the relevant rules, the loser must pay for his opponent's *107 attorneys' fees. [FN92] In negotiations, the individual plaintiff subject to this regime would be willing to settle for much less than he actually deserves because the cost of litigation would be insurmountable and the stakes at risk in the event of defeat extremely high. Microsoft, for its part, would be either unwilling to settle at all or else willing to accept only a small settlement in the plaintiff's favor. Changing any of these legal rules would yield a significant effect on the bargaining outcomes by shifting the bargaining power from one party to the other. Thus, it is clear that to the extent that the inequality in power and resources between parties distorts the outcome of litigation, this distortion is not confined to the realm of litigation but, rather, spills over into the domain of settlement and other out-of-court agreements. [FN93] The advantages some classes of litigants enjoy in legal proceedings impact their ability to secure favorable settlements outside of the courtroom. Therefore, fair and just settlements are to a large extent dependent on fair and just rules of procedure.
In sum, any effort to construct an alternative to the traditional procedural regime must take into consideration not merely the types of cases (that is, civil versus criminal) but also who is bringing these cases and who is defending them. It is erroneous, however, to craft the rules of procedure based on a dichotomy between the government and private parties that fails to appreciate the complexity of the world of litigation. This world is comprised of different classes of litigants--individuals, small businesses, big businesses, and various branches of government--some of whom are OSs (and, therefore, "like individuals") and others RPs (and, therefore, "like government"). The public-private dichotomy should, accordingly, be replaced with a regime that is sensitive to *108 the significant differences between classes of litigants and their effects on litigation. Such a procedural regime would mould different sets of rules to govern the interactions between the various classes of litigants.

C. The Expressive Justification Reconsidered
The third line of argument for the civil-criminal divide in procedure relates to the expressive dimension of criminal law. The central distinctive aspect of criminal liability, it is argued, is the moral condemnation it engenders, which leads to the stigmatization of those found culpable. As civil law, however, generally deals with conduct devoid of fault, civil liability and sanctions do not communicate a similar message of blame, stigma, and moral condemnation.
The communicative function of criminal law has both negative and positive implications. On the negative side, the blame and stigma that accompany a criminal sanction are as much a threat to the liberty and autonomy of the individual as any deprivation of liberty or property. On the positive side, the condemnation of the offense and the offender expressed by criminal liability and punishment is a valuable function of criminal law and worthy of protection. Both of these aspects lead to the same conclusion: a special procedural regime is necessary to protect defendants from unjustified infliction of culpability and stigma and, at the same time, to prevent the dilution of criminal law's blaming function and maintain criminal punishment as an effective and powerful mechanism of social control.
This expressive justification, though very appealing, is equally misleading. In order to understand its erroneousness, two aspects of the argument, hitherto collapsed in the literature, should be distinguished: the expressivity of substantive law and its effect on the stigmatization of the offender, and the expressivity of the process itself. Accordingly, we divide our criticism into two parts. The first part argues that stigma, which epitomizes the communicative power of the law, is inconsistent with the civil-criminal divide and therefore expressive theories cannot descriptively account for the existing bifurcation. The second part points to the tautology inherent in the expressive justification, which emanates from the expressive power of procedure. These two lines of criticism, taken together, *109 lead to the inevitable conclusion that the expressive argument fails to justify the civil-criminal divide in procedure.
1. The Expressivity of Substantive Law and Its Effect on Procedure
It is commonly held by criminal law theorists that the substantive civil-criminal distinction is fuzzy and "muddled" and became increasingly so over the final decades of the twentieth century. [FN94] Two complementary developments have contributed to this phenomenon: the "criminalization" of civil law, reflected in the increased use of punitive damages as well as other previously criminal sanctions, and the "civilization" of criminal law, reflected in the propensity to criminalize behavior previously regarded as civil or regulatory in character. [FN95] Criminal law is no longer (if it ever was) unique in terms of what behavior it punishes, nor in terms of the types of deprivations it imposes on offenders. [FN96] Numerous types of conduct are violations of both criminal law and civil law, and deprivation of liberty and property characterizes both civil and criminal sanctions. [FN97] Thus, it became much more difficult to justify the distinction between the civil and criminal spheres. It should come as no surprise, given this background, that the expressive theory made such a big splash within legal academia, especially among criminalists, for it offered the only means of distinguishing criminal liability and sanctions from civil ones, when all other ways had proven unsuccessful. [FN98]
*110 Unfortunately, the expressive theory does not succeed at carrying its heavy load. Explaining the procedural divide as based on the stigmatizing aspect of criminal liability as opposed to the nonstigmatizing effect of civil liability is untenable. Not all conduct defined in the books as a crime in fact bears stigma, certainly not to the same extent. Similarly, numerous behaviors not defined as crimes by the legislature do carry stigma, sometimes quite acute. Consequently, we cannot count on stigma per se as a tool to distinguish civil from criminal law.
The interrelation between crime and stigma is complex and convoluted because the two concepts are of a distinctly different kind. Crime is a legal concept. Legislators have absolute discretion to decide which behaviors are classified as crimes and which are not. [FN99] Stigma, in contrast, is a sociological phenomenon. It emanates from social norms and involves the messy routines of social intercourse, which are much harder to anticipate, let alone control or manipulate. [FN100] Nevertheless, there is significant interaction between crime and stigma. Law has the power to influence tastes and alter preferences, [FN101] and criminal law plays a particularly important role in shaping and altering social norms. It is a source of moral authority, the forum where the community expresses its shared values and *111 beliefs as to what is condemnable. [FN102] Hence, the criminalization of a certain conduct signals to the relevant community that the conduct deserves moral condemnation and can set in motion a process that leads to the stigmatization of that conduct. Similarly, increasing the sanction imposed on a criminal activity (say, changing the sanction for air pollution from a fine to imprisonment) sends a message to the public that the conduct in question is morally condemnable to a higher degree than most people had thought until then.
This is, however, only half the story. The other side, arguably the more important one, is that social norms dictate, to an extent, criminal law and policy. [FN103] Criminal law rules do not, indeed cannot, create social norms out of thin air; they can do little more than contribute to existing normative forces. The government can generate a change in social norms only when society is in part open to that change. [FN104] Thus, "[p]assing a statute that criminalizes new conduct does not itself cause that conduct to be perceived as immoral" by the relevant community. [FN105] In fact, if the law "strays too far" from prevailing social norms, it will become self-defeating because the public will lose its respect for the law and those violating it will not be stigmatized. [FN106] Many understand the predicament of modern criminal law in this context. The rise of the administrative state and the strong trend toward criminalizing regulatory offenses has led to *112 the astonishing legislation of more than 4,000 federal crimes. [FN107] The problem with such overcriminalization from an expressive point of view is that the conducts defined as criminal have extended far beyond society's understanding of what is condemnable. As a result, "the meaning of 'criminal liability' becomes incrementally less tied to blameworthiness and incrementally less able to evoke condemnation." [FN108] In sum, there are two interrelated, yet distinguishable, layers to this predicament: First, certain conducts labeled as criminal do not in fact carry stigma since they are not perceived by the community as morally condemnable. Second, as a result of the addition of many such "crimes" to the criminal code, the meaning of criminal liability in general has been diluted.
Just as passing a statute criminalizing certain conduct does not create a social norm that stigmatizes the conduct, so failing to pass a law prohibiting a conduct that is morally repugnant to the community cannot make that conduct morally acceptable. [FN109] Abortion and adultery are conducts that many Americans find morally wrong despite the fact that criminal law does not prohibit them. [FN110] *113 Accordingly, the civil process is fraught with behaviors that impose stigma on the actors. One example is the termination of parental rights: parents whose rights are terminated by the state are subject to a severe stigma that "reflects the judgment of blame that universally underlies a decision to terminate parental rights." [FN111] Civil commitment and confinement under sexual predator laws are another example. Not only do the civilly committed "lose their right to be free from state-imposed confinement[,] but [they] also are stigmatized as mentally ill," which is no less damaging to them as individuals and to their autonomy than being labeled as a criminal. [FN112] Likewise, a lawyer's debarment or a physician's loss of license has a detrimental effect on his respective social status and reputation. The civil forfeiture of an individual's residence or other assets is obviously quite stigmatizing. [FN113] Lastly, losing one's residence due to bankruptcy is also a stigmatizing event. In a society in which the assessment of personal worth is grounded in the belief that "bad things happen to bad people" and in which being poor is *114 considered morally dubious, the emotional and social consequences of bankruptcy cannot be overstated. Bankruptcy is viewed by most as a signal of failure and irresponsibility, which brings with it a significant loss of esteem. [FN114]
Moreover, with the legislature's increasing use of civil penalties to augment social control, the public comes to associate these civil penalties with conduct that is morally wrong. [FN115] Punitive damages, sometimes awarded to the plaintiff in civil actions, are one case in point. Such damages are an overt way of deterring and condemning the injurer. [FN116] As argued by Joel Feinberg, "What more dramatic way of vindicating his violated right can be imagined than to have a court thus forcibly condemn its violation through the symbolic machinery of punishment?" [FN117]
Since the expressive justification for the civil-criminal procedural separation rests on the power of criminal law to condemn offenders, the more crimes on the books that fail to stigmatize their violators and the greater the incidence of stigmatization in the context of civil proceedings, the less compelling the expressive justification.
2. The Expressivity of Procedure
The expressive justification asserts a need for two separate sets of procedure not only to protect people from the wrongful infliction of stigma, but also to sustain the blaming function of a criminal *115 conviction. A procedural regime that treats civil and criminal cases alike would risk diluting the expressive value of the criminal conviction and impairing the state's ability to use criminal law as a mechanism to condemn the culpable. Criminal conviction, so the argument goes, constitutes a valuable social label precisely because a unique procedural regime sets it off as a special site at which criminal law's blaming function can be executed. [FN118]
The expressive justification, however, is undermined by circular reasoning. Beyond its role in determining criminal liability, the criminal process has a communicative function. Criminal procedure reflects society's normative ideals as to what is a fair and just process for the conviction of a criminal. The application of heightened procedural standards assists in the realization of criminal law's expressive objectives in two distinct ways: At the outset, the very application of these standards reflects the social preference for undeserved acquittals over wrongful convictions. [FN119] More importantly, criminal procedure preserves the stigmatizing effect of the criminal conviction; it guarantees maximal exactitude in conviction as reflecting de facto guilt and thus preserves the ability of the criminal trial mechanism to communicate moral blame and generate indignation toward a person and behavior defined as criminal. The expressive argument must, therefore, be reformulated as follows: a criminal conviction constitutes a valuable sign of social condemnation not only because it is set off by a separate procedure, but also because the procedure by which conviction is secured is more demanding and renders criminal branding more accurate. [FN120] The heightened standards of criminal procedure guarantee a high level of certitude in convictions, which is a prerequisite for the effective functioning of criminal law as the locus of blame and moral condemnation. The elimination of the civil-criminal divide would adversely affect the value of the institution of criminal convictions due to the lowering of the procedural bar. *116 This would necessarily entail a loss of legitimacy and, thus, diminish the expressive function of a criminal conviction.
This reformulation of the expressive argument exposes its circular nature: in order to protect the blaming function of criminal law and sustain the stigmatizing power of criminal conviction, a more demanding procedure is required, but, at the same time, the more demanding procedure is itself the basis for the heavier stigma costs embodied in the criminal branding. In other words, this line of argument seeks to justify the bifurcation of procedure into civil and criminal by referring to the expressive value of criminal conviction, while that very value is, in fact, contingent upon the level of accuracy (namely, the probability of wrongful convictions), which is derived from the type of procedure in use. [FN121]

D. The State-Centered Justifications Reconsidered
The final type of rationale for the civil-criminal distinction focuses on the state and its unique powers and obligations. As we saw, state-centered justifications can be divided into two basic types of arguments: the one revolves around the exceptional dangerousness of the state if left unchecked and the other around the need to legitimize the liberal state's extraordinary actions in the criminal sphere.
1. The Political Oppression Argument
The political oppression argument, advocated most powerfully by Professor Donald Dripps, asserts that the fact that the state has the sole authority to impose criminal punishment constitutes a tremendous political threat to liberty. Since governments have, in the past, tended to exploit their vast criminal law powers to disable opposition and target certain identifiable groups, [FN122] special procedural safeguards are necessary to curb these powers and prevent their abuse.
*117 The fallacy of this claim, along with the analogous public choice argument, [FN123] lies in the fact that the criminal apparatus is not the sole mechanism that can be used for rent-seeking purposes or in order to oppress and disable political adversaries. A regime that seeks to squash opposition or express malice toward a given group can just as easily resort to civil sanctions that are materially indistinguishable from, and sometimes even more onerous than, criminal sanctions. Moral condemnation and stigma accompany not only criminal sanctions but civil sanctions as well. Deportation, commitment to a mental institution, administrative detention, and termination of parental rights are all civil sanctions that are both painful and stigmatizing. [FN124] It is the ability to apply these mechanisms easily through the civil system, and thus stay below the public radar, that makes them attractive to bad governments. The claim that some governments are sadistic and would therefore find satisfaction only in inflicting pain through the criminal process [FN125] is unconvincing and unsupported by historical evidence. To the contrary, as Dripps himself admits, totalitarian regimes have often resorted to civil mechanisms to silence their political adversaries, confining them to mental institutions or placing them in administrative detention. [FN126] Indeed, a government interested in removing opposition would likely use any tool at its disposal, civil or criminal.
Dripps's argument, after some of the rhetoric has been peeled away, is no more than a combination of the utilitarian and expressive arguments. It is the severity of the sanction accompanied by the power of blaming that makes criminal sanctions tempting to the bad government and hence Dripps himself concedes that "if a great many examples of civil proceedings that would tempt a bad government can be identified, the civil-criminal distinction may have to be surrendered." [FN127] Since governments make use of both the criminal and civil apparatuses for political oppression, procedural protections need to be applied not only in the criminal sphere *118 but also in civil proceedings to which the state is a party and that may cause severe harm to the individual.
2. The Liberal State Argument
Under the liberal state argument, the state's use of its criminal apparatus to mete out punishment is something that does not sit well with the concept of a liberal democracy. According to conventional wisdom, criminal liability is unique in that it "reflect[s] moral blameworthiness deserving condemnation and punishment," something that civil liability does not necessarily convey. [FN128] By engaging in moral condemnation, the state encroaches on the autonomy of its subjects and tries to reform them, a task that the liberal state, with its ideological commitment to value neutrality, is not supposed to do. Since blaming is an essential element of criminal law and doing away with it would, therefore, be not only impossible but also undesirable, [FN129] more rigorous and exacting procedure is necessary to legitimize instances of the exercise of these extraordinary powers. It is noteworthy that, unlike the other rationales presented, the liberal state argument for the civil-criminal divide is nonconsequential. Indeed, it does not focus on--is not even interested in--the harm caused to the individual by criminal conviction or punishment (which could be trivial or nonexistent) but, rather, focuses only on the need to justify the state's actions.
There is, however, a deep sense in which the liberal state argument fails: the idea that the state uses its coercive power to impose certain values and promote a specific conception of good only in criminal law, and not in civil law, is simply and plainly wrong. The origins of this erroneous conception lie in the liberal ideology's deep commitment to the notion of value subjectivity. Value subjectivity understands values (beliefs as to what goals are worthy of pursuit) to be no more than arbitrary tastes and preferences. [FN130] A *119 commitment to value subjectivity implies a "facilitative" end-neutral state that does not seek to promote any particular vision of the good life but simply facilitates people in pursuing their preferences and desires. [FN131] The fundamental problem with such a political theory is that it places at odds two crucial components of any democratic regime: freedom and order. If freedom allows all individuals to strive for their subjectively chosen ends, it must permit the pursuit of unconstrained selfish desires, which inevitably conflict with those of others. One person's desire to rape necessarily clashes with another's preference not to be raped. In a world in which values are regarded as nothing more than subjective preferences, there would be no ground rule favoring one preference over another. Such a regime would yield, of course, a "Hobbesian war of all against all." [FN132]
The compromise liberals reach is to divide the world into two spheres, the public and the private. They do not deny state coercion a place in this world entirely but claim that it is limited to the public sphere, which can be clearly distinguished from the private sphere. In the public sphere, the state can legitimately place certain constraints on the individual's conduct. Criminal law is, of course, a prime example. The state has a legitimate interest in preventing conduct that harms other individuals and infringes on their rights, and it may use criminal law for that purpose. [FN133] The private sphere, in contrast, is where the commitment to value subjectivity and the idea of the state's end-neutrality is most pronounced. The state must refrain from passing value judgments or otherwise interfering in the affairs of the private sphere. It is essential, therefore, for liberals to maintain a strict boundary between the public and the private. [FN134]
Ever since the formulation of legal realism, critical scholars as well as contemporary liberal scholars have argued against the tenability of such a separation. [FN135] The state, they posit, most certainly *120 uses its coercive powers to impose certain values and promote specific conceptions of good not only in the public sphere but also in the private sphere. [FN136] In fact, coercive state action in the private realm is inevitable. Even contract law, in which the liberal commitment to the idea of freedom and subjective end-seeking is most pronounced, is not immune to such criticism. The belief that there is a private domain, dominated by consensual contracts, is contingent on ignoring the extent to which the state regulates the ground rules that hinder one's ability to induce others to contract. Express contracts become "express" only when a court (an agent of the state) has made a political and moral decision to treat certain parties' expressions of intention as binding. [FN137] A decision by the state to give public force to promises made by unmarried cohabiters transforms legally empty words of commitment into binding contracts. It is, thus, a political decision imbued with collective values. There is no way to conceive such a decision other than as a coercive state action informed by collective attitudes about the proper nature of the particular relationship. [FN138]
The reason that the role of values and morality in the private sphere is obfuscated is that we are used to perceiving only rules of prohibition, and not rules of permission, as ground rules. [FN139] In reality, private law sets the rules of the game by which people seek their goals. It influences behavior not by telling people what to do and what to refrain from doing, as criminal law does, but by providing the background for the realm of possibility--that is, prescribing what people can and cannot get away with in their dealings with others. Thus, the ostensibly nondirective rules of property, *121 tort, and contract impact the social order even though they neither compel nor forbid. [FN140] Consequently, the coercive role of the state in private law and its intervention in individual autonomy undercut the liberal state argument for the civil-criminal divide.

E. Summary
None of the four types of justifications for the civil-criminal divide in procedure can withstand the criticism directed at it. The conclusion that arises from our analysis in this Part is that the differences between the civil and the criminal paradigms are of degree rather than kind: there are more criminal cases that impose high costs and stigma on the defendant than civil cases that do so. In criminal trials, the state is invariably the moving party, while, in civil litigation, this is only sometimes the case. A bad government will find more criminal mechanisms to serve its malicious ends than civil mechanisms, and so on. The number of civil cases that fit into the criminal paradigm, and vice versa, is, however, large enough to give rise to significant doubts about the dichotomous approach. [FN141] Therefore, if we are to remain true to the goals of equality, due process, and efficiency, we must either find new ways to justify the civil-criminal divide or, alternatively, abandon it altogether and seek new analytical foundations for procedure that are more in line with its goals and values.

III. The Civil-Criminal Puzzle: Comments on the Literature
The conceptual and normative pitfalls inherent in the existing bifurcated procedural regime have not gone unnoticed in the literature. To maintain this dichotomous system, it is essential to be able to clearly distinguish the civil from the criminal. Over the years, however, as the number of civil statutes imposing sanctions traditionally thought of as criminal has risen and the regulatory function of criminal law has expanded, this has become an increasingly difficult, if not impossible, task. Litigation over the application of *122 criminal procedural protections to actions labeled by Congress as civil began to intensify in the mid-twentieth century and further accelerated toward the end of the century. The Supreme Court was called upon time and time again to determine the appropriate procedure for such diverse matters as deportation, loss of citizenship, loss of license, debarment, termination of parental custody, juvenile delinquency, civil commitment, pretrial detention, forfeiture, civil fines, increased tax assessment, and civil contempt. [FN142] With the breakdown of the civil-criminal boundaries, growing normative concerns have arisen regarding both efficiency and deprivation of due process. If, on the one hand, the parameters of the criminal sphere are too broad, then the costly and time-consuming criminal procedure will be brought into a wide range of proceedings, placing an impossible burden on the state budget and impairing the state's ability to bring wrongdoers to justice. If, on the other hand, it is defined too narrowly, then the important values underlying criminal procedural protections will be sacrificed and, again, justice will be obstructed. [FN143] Many attempts have been made, by both jurists and legal scholars, to offer a fair, efficient, and workable test for distinguishing civil from criminal cases that would clearly indicate when it is appropriate to apply enhanced procedural safeguards. In what follows, we describe in brief the laudable, yet largely unsuccessful, attempts at resolving this civil-criminal puzzle.

A. Identifying Punitiveness: An Exercise in Futility
The solution most frequently raised for the civil-criminal procedural challenge is to reframe the question: instead of asking whether a given law or sanction has been formally labeled "criminal" or "civil," many suggest examining its purpose. Conventional wisdom tells us that the purpose of the civil system is compensatory or remedial (that is, to redress concrete losses), whereas the purpose of the criminal system is to punish past wrongdoings. [FN144] Under this approach, sanctions that serve remedial purposes are civil in nature and thus warrant no special procedural protections, whereas those aimed at punishing are criminal in nature and therefore merit *123 heightened procedural safeguards, regardless of how Congress chooses to label them. [FN145] Had the legislature indeed shaped substantive criminal law in line with its punitive purpose, we would not have had any problem to contend with. Under such a scenario, criminal substantive law would have dealt only with conduct and people deserving of punishment, and criminal procedure could have been applied mechanically to all cases labeled criminal.
Difficulties do arise, however, because prevailing substantive criminal law does not hold exclusive reign over the realm of punishment. Due to the state's propensity for civil avenues to address criminal conduct and tendency to use criminal law to regulate (rather than punish) behavior, it is impossible to rely on a sanction's legislatively assigned label to determine the appropriate procedure to apply. The Supreme Court had already reached this conclusion by the late-nineteenth century. [FN146] In the 1886 case of Boyd v. United States, [FN147] the government had subpoenaed personal papers and business records of a merchant who had been found liable for fraudulently importing certain goods in violation of the custom laws, seeking to subject the goods to forfeiture. [FN148] The Court rejected the government's petition, holding that seizing books in order to facilitate forfeiture serves the same purpose as forcing a defendant to testify against himself in a criminal trial. [FN149] To the claim that the proceedings were civil and therefore the protection against self-incrimination did not apply, the Court responded that "[w]e are . . . clearly of opinion that proceedings instituted for the purpose of declaring the forfeiture of a man's property by reason of offences committed by him, though they may be civil in form, are in their nature criminal." [FN150] In a number of cases since Boyd, the Court has held that the label chosen by Congress is not a definitive indicator of the real nature of the proceedings. Accordingly, a legislative *124 decision to label a certain sanction "civil" does not automatically entail denial of constitutionally guaranteed criminal procedural protections. Such protections are triggered, instead, by the "punitive purpose" of the sanction, a concept that captures the sanction's function rather than its form. [FN151]
1. Practical Obstacles
Ever since Boyd, the Supreme Court, assisted by leading scholars, has been trying to formulate a clear notion of the meaning of "punishment" that would make the bifurcated procedural system workable. These efforts to distinguish criminal sanctions from civil ones by referring to their punitive purpose have been unavailing. The reason for this failure derives from the methods by which courts isolate the purpose served by a given law or sanction and the factors that must be weighed in deciding what makes them punitive. There are basically two methods courts can and do employ in order to determine the purpose of a sanction. One method is to inquire into its legislative history in an attempt to uncover the legislature's intention. The other method is to infer the nature of the sanction from its effects, namely, by probing into whether it demonstrates certain or all of the indicia of punishment.
Many academic commentators have rejected examining the legislative history to uncover the purpose of a given law or sanction as "inconclusive" and "unseemly." [FN152] As Professor John Hart Ely has argued, inquiries into the legislature's motives is a hazardous enterprise that gives rise to issues of "[a]scertainability, [f]utility and [d]isutility." [FN153] Applied to our case, the Court's inquiries into the legislature's intent in instituting a sanction have been anything but consistent and conclusive. These inquiries often produce contradictory conclusions, as the Court remains uncertain about which test *125 should be used to determine motivation. [FN154] It is therefore difficult, if not impossible, to ascertain the purpose of a sanction through the legislative history, and it is certainly normatively dubious.
Furthermore, the impact of the sanction is identical regardless of the legislature's motivation in legislating it into law. Accordingly, if enhanced procedural protections are designed to protect individuals from being subjected to state-sponsored coercion without due process, the legislature's motivation should not be of concern. [FN155] In addition, the legislative motivation is open to the negative incentives that the Court pointed to in Boyd as preventing it from blindly accepting its label. Therefore, relying on the legislature's intent is not only useless but can also be dangerous.
The second method for determining the nature of a sanction, namely, based on its punitive effects, is equally imprecise. Any attempt to identify "criminal" by defining "punitive" does nothing more than substitute one obscure term (criminal) with another (punitive). For the punitive-purpose method to work, a prior determination as to the definition of "punishment" is required; this turns out to be a daunting task. The Court's most elaborate endeavor to grapple with the meaning of punishment was in its 1962 decision in Kennedy v. Mendoza-Martinez. [FN156] This case dealt with the question of whether expatriation for draft evasion constitutes punishment and therefore requires "a prior criminal trial and all its incidents." [FN157] Rather than deferring to the labels assigned by Congress, the Court adopted a multifaceted test to distinguish between civil sanctions that are punitive and thus criminal in nature and those that are not. The seven factors identified by the Court were as follows:
[1] Whether the sanction involves an affirmative disability or restraint, [2] whether it has historically been regarded as a punishment [3] whether it comes into play only on a finding of scienter, [4] whether its operation will promote the traditional aims of *126 punishment-- retribution and deterrence, [5] whether the behavior to which it applies is already a crime, [6] whether an alternative purpose to which it may rationally be connected is assignable for it, and [7] whether it appears excessive in relation to the alternative purpose assigned . . . . [FN158]
Although the multifaceted test has been invoked frequently, [FN159] it has likely never determined an outcome, not even in the Mendoza-Martinez case itself, [FN160] and for good reason: the Mendoza-Martinez test is unable to yield a principled and predictable answer. In every case of its application, the judge ends up with a mixture of yes and no answers to each of the seven factors and must ultimately determine whether the sanction at hand is civil or criminal based upon her own valuation of each factor and its relative weight. [FN161] Inevitably, the Court resorts to tautological reasoning: it purports to define as criminal, and thus order heightened procedural safeguards for sanctions that serve primarily to punish, when, as a matter of fact, the punitive purpose ascribed to the sanction rests upon some intuition regarding the procedural safeguards that the sanction merits. In other words, the purported test does no more than restate the underlying issue, rather than lead to the answer. [FN162] It *127 therefore comes as no surprise that the Court's application of the punitive purpose test in specific cases "has proved to be highly unpredictable and confusing." [FN163]
Efforts to classify legal actions as civil and criminal according to whether or not they have a punitive purpose are doomed to failure for yet another reason: many of these actions are a hybrid in the sense that they serve more than one purpose. The regulatory state we live in is built on a complex system of rewards and penalties, and many of the measures developed in this system combine features from both the civil and the criminal sides of the divide. For instance, does the state revoke a physician's license in order to punish and deter her or to ensure the provision of adequate medical services? [FN164] Is the imposition of double tax assessment on a person found guilty of tax evasion an additional punishment or a way to compensate the government for its efforts and expenses? [FN165] Does the forfeiture of an asset to the government serve solely remedial purposes or retributive and deterrent purposes as well? [FN166] Is the indefinite *128 detention of dangerous people not convicted of any crime a punishment or simply a way of ensuring public safety? [FN167] These measures, like many others, are neither wholly criminal nor entirely civil, but rather, as Professor Susan Klein has aptly observed, they are "more like an old style Chinese menu, where the patron selects one entrée from column A and two from column B." [FN168] An *129 approach that classifies all proceedings that include some so-called punitive component as criminal, no matter how slight, would compel the state to afford the cumbersome procedural protections to a vast array of measures, thereby burdening the legal system with additional expense and resulting in a decrease in efficiency. [FN169] It is therefore hardly surprising that by the late 1970s, the Supreme Court had abandoned all attempts to identify punishment independently and in all but the most extreme cases it "docilely accepted at face value the 'civil' label attached to a proceeding: if the legislature said the sanction had a non-punitive purpose, the Court agreed." [FN170]

[FN77]. See Marc Galanter, Why the "Haves" Come out Ahead: Speculations on the Limits of Legal Change, 9 Law & Soc'y Rev. 95 (1974).

[FN78]. Id. at 97-98.

[FN79]. Id. at 98.

[FN80]. Id.

[FN81]. Id. at 98-101.

[FN82]. Since the publication of Galanter's analysis, his theoretical insights have spawned numerous studies examining the advantages of RPs over OSs in a wide variety of trial and appellate courts. Generally speaking, these studies have confirmed Galanter's insights. A study of federal civil cases between the years 1971 and 1991 revealed that big business ("Fortune 2000" companies) had a success rate of 71% as plaintiff and 61% as defendant when facing all types of litigants in court, whereas nonbusiness litigants won only 64% of the time as plaintiff and a mere 28% of the time as defendant. See Terence Dunworth & Joel Rogers, Corporations in Court: Big Business Litigation in U.S. Federal Courts, 1971-1991, 21 Law & Soc. Inquiry 497, 558 (1996). Similarly, a study of diversity cases in federal courts found that in instances where litigants are of the same type (individual versus individual or corporate versus corporate), the plaintiff prevails 72% to 75% of the time; however, when corporate plaintiffs sue individuals, they win 91% of the time, and when individuals sue corporate plaintiffs, they win only 50% of the time. Theodore Eisenberg & Henry S. Farber, The Litigious Plaintiff Hypothesis: Case Selection and Resolution, 28 RAND J. Econ. (Special Issue) S92, S103 (1997). More recent empirical work assessing Galanter's theory has focused on appellate courts at the state and federal levels. See Paul Brace & Melinda Gann Hall, "Haves" Versus "Have Nots" in State Supreme Courts: Allocating Docket Space and Wins in Power Asymmetric Cases, 35 Law & Soc'y Rev. 393 (2001); Kevin M. Clermont & Theodore Eisenberg, Plaintiphobia in the Appellate Courts: Civil Rights Really Do Differ from Negotiable Instruments, 2002 U. Ill. L. Rev. 947; Theodore Eisenberg, Appeal Rates and Outcomes in Tried and Nontried Cases: Further Exploration of Anti-Plaintiff Appellate Outcomes, 1 J. Empirical Legal Stud. 659 (2004). A recently published book was devoted to exploring the continued validity of Galanter's theory in contemporary civil litigation. See generally In Litigation: Do the "Haves" Still Come out Ahead? (Herbert M. Kritzer & Susan S. Silbey eds., 2003) [hereinafter In Litigation].

[FN83]. See Donald R. Songer et al., Do the "Haves" Come out Ahead over Time: Applying Galanter's Framework to Decisions of the U.S. Courts of Appeals, 1925-1988, in In Litigation, supra note 82, at 93 (describing a study of decisions from all circuits in the U.S. Courts of Appeals for a sixty-four-year period between 1925 and 1988 and finding that the federal government had a net advantage of 25.6%, state and local governments had a net advantage of 15.6%, businesses had a net advantage of negative 2.8%, and, at the bottom, individuals, with a net advantage of negative 12.6%); see also Donald R. Songer & Reginald S. Sheehan, Who Wins on Appeal? Upperdogs and Underdogs in the United States Courts of Appeals, 36 Am. J. Pol. Sci. 235, 241 (1992) (analyzing both the published and unpublished decisions of the U.S. Courts of Appeals and finding that the overall success rate of the government was roughly four times higher than the success rate of individuals and two and a half times the success rate of businesses). Several studies applying Galanter's theory to court decisions in other common law countries have generated similar findings. See, e.g., Burton M. Atkins, Party Capability Theory as an Explanation for Intervention Behavior in the English Court of Appeal, 35 Am. J. Pol. Sci. 881, 894-95 (1991) (surveying the English Court of Appeal and finding that the government enjoyed a 25% advantage over corporate litigants and corporations enjoyed a 14% advantage over individuals); Peter McCormick, Party Capability Theory and Appellate Success in the Supreme Court of Canada, 1949-1992, 26 Can. J. Pol. Sci. 523, 532 (1993) (describing a study of the Canadian Supreme Court revealing that government's net advantage was approximately 5% higher than that enjoyed by big business, 26% higher than the net advantage found for other businesses, and 30% higher than the success rate for individuals).

[FN84]. Gillian K. Hadfield, Exploring Economic and Democratic Theories of Civil Litigation: Differences Between Individual and Organizational Litigants in the Disposition of Federal Civil Cases, 57 Stan. L. Rev. 1275, 1322 (2005).

[FN85]. Id. at 1314-15.

[FN86]. Id.

[FN87]. See id. at 1317.

[FN88]. See generally Marc Galanter, The Vanishing Trial: An Examination of Trials and Related Matters in Federal and State Courts, 1 J. Empirical Legal Stud. 459 (2004) (discussing declining trial frequencies). But see Gillian K. Hadfield, Where Have All the Trials Gone? Settlements, Nontrial Adjudications, and Statistical Artifacts in the Changing Disposition of Federal Civil Cases, 1 J. Empirical Legal Stud. 705 (2004) (arguing that the primary shift in litigation over the past three decades has not been from trial to settlements but from bench trials to nontrial adjudications (that is, motions to dismiss and summary judgments)).

[FN89]. See Robert H. Mnookin & Lewis Kornhauser, Bargaining in the Shadow of the Law: The Case of Divorce, 88 Yale L.J. 950 (1979); see also Steven Shavell, The Fundamental Divergence Between the Private and the Social Motive to Use the Legal System, 26 J. Legal Stud. 575, 607 (1997) (arguing that the social utility of trials is to "provide victims with the threat necessary to induce settlements").

[FN90]. See Duncan Kennedy, The Stakes of Law, or Hale and Foucault!, in Sexy Dressing Etc. 83, 87-89 (1993).

[FN91]. Unfortunately, this is not a figment of our imagination but the reality of our adversary system. In a well-known article, Robert Rabin looked at the adversarial techniques employed by the tobacco industry in suits brought against it by individuals who had contracted smoke-related illnesses and their families (a paradigmatic case of OS versus RP). The tobacco industry has never offered to settle in a single case. Instead, tobacco companies retain counsel from the most prestigious law firms and spare no cost in their attempt to exhaust their adversaries' resources short of the courthouse doors. They take good advantage of the large arsenal of easily manipulated procedural mechanisms that our adversarial system offers:
They have done this by resisting all discovery aimed at them, thus requiring a court hearing and order before plaintiffs can obtain even the most rudimentary discovery. They have done it by getting confidentiality orders attached to the discovery materials they finally produce, thus preventing plaintiffs' counsel from sharing the fruits of discovery and forcing each plaintiff to reinvent the wheel. They have done it by taking exceedingly lengthy oral depositions of plaintiffs and by gathering, through written deposition, every scrap of paper ever generated about a plaintiff, from cradle to grave. And they have done it by taking endless depositions of plaintiffs, expert witnesses, and by naming multiple experts of their own for each specialty, such as pathology, thereby putting plaintiffs' counsel in the dilemma of taking numerous expensive depositions or else not knowing what the witness intends to testify to at trial. And they have done it by taking dozens and dozens of oral depositions, all across the country, of trivial fact witnesses, particularly in the final days before trial.
William E. Townsley & Dale K. Hanks, The Trial Court's Responsibility to Make Cigarette Disease Litigation Affordable and Fair, 25 Cal. W. L. Rev. 275, 277 (1989). It is therefore not surprising that, over roughly thirty-five years, plaintiffs "failed to gain a single clear-cut victory." Robert L. Rabin, A Sociolegal History of the Tobacco Tort Litigation, 44 Stan. L. Rev. 853, 854 (1992).

[FN92]. As is the English rule.

[FN93]. Owen Fiss argues that the disparity in resources between the parties influences the settlement process in three ways:
First, the poorer party may be less able to amass and analyze the information needed to predict the outcome of the litigation, and thus be disadvantaged in the bargaining process. Second, he may need the damages he seeks immediately and thus be induced to settle as a way of accelerating payment.... Third, the poorer party might be forced to settle because he does not have the resources to finance the litigation.... The indigent plaintiff is a victim of the costs of litigation even if he settles.
Owen M. Fiss, Against Settlement, 93 Yale L.J. 1073, 1076 (1984).

[FN94]. Robinson & Darley, supra note 47, at 479.

[FN95]. Id.; see also John C. Coffee, Does "Unlawful" Mean "Criminal" ?: Reflections on the Disappearing Tort/Crime Distinction in American Law, 71 B.U. L. Rev. 193, 216 (1991).

[FN96]. See, e.g., Marc Galanter & David Luban, Poetic Justice: Punitive Damages and Legal Pluralism, 42 Am. U. L. Rev. 1393, 1394 (1993) (arguing that the conventional wisdom that civil law provides victims with compensation while criminal law inflicts punishments on wrongdoers is simply erroneous and that there are many "forms of legally recognized noncriminal or 'civil style' punishments that are as basic to social and legal life as criminal punishment").

[FN97]. Paul H. Robinson, The Criminal-Civil Distinction and Dangerous Blameless Offenders, 83 J. Crim. L. & Criminology 693, 693-94 (1993).

[FN98]. See, e.g., Sanford Kadish, The Use of Criminal Sanctions in Enforcing Economic Regulations, 30 U. Chi. L. Rev. 423 (1963), reprinted in Blame and Punishment: Essays in the Criminal Law 40, 51 (1987) ("The central distinguishing aspect of the criminal sanction appears to be the stigmatization of the morally culpable.").

[FN99]. Professor Bill Stuntz suggests constitutionally restricting the legislature's authority to define new crimes. Stuntz, supra note 10, at 2.

[FN100]. The best exposé of stigma to date has been offered by sociologist Erving Goffman in his well-known book, Stigma: Notes on the Management of Spoiled Identity. Goffman, supra note 52. Goffman defines stigma as "an attribute that is deeply discrediting," but does not envision it as a stable and fixed attribute. Id. at 3. Rather, for him, stigmatization is contextual and relational, arising out of a particular configuration of attributes and social expectations. It is possible, therefore, that a characteristic would be stigmatizing in one context or to a certain group but not in a different context or to a different group. Goffman provides the following example: "the shoulder patches that prison officials require escape-prone prisoners to wear can come to mean one thing to guards, in general negative, while being a mark of pride for the wearer relative to his fellow prisoners." Id. at 46. Moreover, the capacity of a certain attribute to stigmatize fluctuates over time. Being an upper-middle-class divorcee in nineteenth-century America carried a stigma at the time, whereas today the same status is a benign social fact. Id. at 32.

[FN101]. Tracey L. Mears et al., Updating the Study of Punishment, 56 Stan. L. Rev. 1171, 1179 (2004); see also Kenneth G. Dau-Schmidt, An Economic Analysis of the Criminal Law as a Preference-Shaping Policy, 1990 Duke L.J. 1, 1 (arguing that the criminal law assists in reducing crime by "shaping the individual's preferences by increasing her taste for desired behavior"); Cass R. Sunstein, Legal Interference with Private Preferences, 53 U. Chi. L. Rev. 1129, 1146 (1986).

[FN102]. Robinson & Darley, supra note 47, at 474.

[FN103]. Paul Robinson has stressed this point in many of his books and articles. See, e.g., Paul H. Robinson & John M. Darley, Justice, Liability, and Blame: Community Views and the Criminal Law (1995); Robinson & Darley, supra note 47, at 473-74; Paul H. Robinson, Why Does the Criminal Law Care What the Layperson Thinks Is Just? Coercive Versus Normative Crime Control, 86 Va. L. Rev. 1839, 1839 (2000).

[FN104]. Lawrence Lessig, The Regulation of Social Meaning, 62 U. Chi. L. Rev. 943, 1030 (1995).

[FN105]. Robinson & Darley, supra note 47, at 481; see also Susan R. Klein, Redrawing the Criminal-Civil Boundary, 2 Buff. Crim. L. Rev. 679, 718 (1999) ("Some criminal punishments, such as the Texas misdemeanor of driving with an open beer bottle, are considered petty by most people and are thus not particularly stigmatizing."); Stuntz, supra note 10, at 26 ("In a state that criminalizes the riding of bell-less bicycles, the criminal label will soon lose its punch.").

[FN106]. William J. Stuntz, Self-Defeating Crimes, 86 Va. L. Rev. 1871, 1872-73 (2000). Stuntz offers examples of three kinds of crimes that might prove self-defeating. The first is vice crimes such as prostitution, gambling, and drug use, with Prohibition the most notable instance. The second example is contemporary white-collar crimes. The third example is abortion. Id.

[FN107]. Baker, supra note 8, at 3; see also Coffee, supra note 95, at 216 (citing a study by Stanley Arkin according to which there are over 300,000 federal regulations that may be enforced criminally).

[FN108]. Robinson & Darley, supra note 47, at 481; see also John L. Diamond, The Myth of Morality and Fault in Criminal Law Doctrine, 34 Am. Crim. L. Rev. 111, 112-13 (1996) (arguing that contemporary criminal law cannot be fairly characterized as a moral system that condemns blameworthy choices, but, rather, to a substantial degree, as punishing transgressions without reference to personal culpability); Renfrew, supra note 71, at 603-04. Judge Charles Renfrew interviewed a group of business people and asked them to reply to several questions regarding violation of antitrust laws, including whether antitrust violators should be imprisoned. Representative answers included the following: "They broke the law but they are not 'criminals' in the true sense of that word;" "I think the laws and the judges should concentrate more on putting criminals in prison and keeping them there, than sending taxpayers to prison;" and "[Imprisoning antitrust violators would be a] gross miscarriage of justice." Id. But see Coffee, supra note 55, at 1889 ("[T]he limited empirical evidence on public attitudes toward white-collar crimes suggests that the public learns what is criminal from what is punished, not vice versa.").

[FN109]. Robinson & Darley, supra note 47, at 473.

[FN110]. It is possible that passing a law criminalizing these types of conduct would contribute to their increased repugnancy in the eyes of the community. If we take Stuntz's argument about self-defeating crimes seriously, however, criminalizing these conducts could have the countereffect of galvanizing opposition to such crimes and thus would dilute their stigmatizing force. "Sometimes," Stuntz argues, "the best way for the legal system to advance or reinforce norms may be to ignore them." Stuntz, supra note 106, at 1873.

[FN111]. David D. Meyer, Family Ties: Solving the Constitutional Dilemma of the Faultless Father, 41 Ariz. L. Rev. 753, 782-83 (1999); see also Rachel Mallory Leitzë, In re: Samantha C.: Civilizing Civil Proceedings Through Full Incorporation of the Fifth Amendment Privilege Against Self-Incrimination, 19 Quinnipiac Prob. L.J. 111, 125 (2005) ("[A] hearing to terminate parental rights, while classified as a 'civil proceeding,' carries stigma akin to that of a criminal conviction[.]"); Colleen McMahon, Due Process: Constitutional Rights and the Stigma of Sexual Abuse Allegations in Child Custody Proceedings, 39 Cath. Law. 153, 160 (1999) ("[P]arents facing loss of custody based on sexual abuse accusations face similar risks, particularly with respect to the stigma which immediately attaches."). The Supreme Court acknowledged the heavy stigma visited upon parents in termination of parental rights cases and therefore enhanced the procedural safeguards in such cases. See M.L.B. v. S.L.J., 519 U.S. 102, 118, 124 (1996) (acknowledging the stigmatizing effect of termination of parental rights in holding that due process requires a waiver of fees if necessary to enable an indigent parent to appeal); Santosky v. Kramer, 455 U.S. 745, 756, 768-70 (1982) (acknowledging the stigmatizing effect in holding that due process requires that grounds be proved by clear and convincing evidence); Lassiter v. Dep't of Soc. Servs., 452 U.S. 18, 31-32 (1981) (acknowledging the stigmatizing effect in holding that due process may require appointment of counsel in some cases).

[FN112]. Brian J. Pollock, Note, Kansas v. Hendricks: A Workable Standard for "Mental Illness" or a Push down the Slippery Slope Toward State Abuse of Civil Commitment?, 40 Ariz. L. Rev. 319, 320 (1998); see also Dripps, supra note 61, at 215 ("[T]he stigma associated with lunacy is comparable to that associated with criminality."). The Supreme Court has recognized the severe stigma that attaches to civil commitment. See, e.g., Vitek v. Jones, 445 U.S. 480, 492 (1980) (noting that the stigma imposed by civil commitment can have "a very significant impact on the individual").

[FN113]. Klein, supra note 105, at 718.

[FN114]. See Rafael Efrat, The Evolution of Bankruptcy Stigma, 7 Theoretical Inquiries L. 365, 377-80 (2006) (citing studies from the late 1990s and early 2000s suggesting that stigma remains an important factor in preventing many individuals from filing for bankruptcy). Probably the most dramatic manifestation of the severe stigma that bankruptcy carries can be gleaned from one of the findings of a series of surveys conducted during the last thirty years of the twentieth century that a significant number of respondents (8%) believed that bankruptcy is a valid reason for committing suicide. Library Index, Public Opinion About Life and Death--Suicide, http:// www.libraryindex.com/pages/610/Public-Opinion-About-Life-Death-SUICIDE.html (last visited Jan. 21, 2008) (citing 2002 National Opinion Research Center survey). Respondents were asked to respond to the following question: "Do you think a person has the right to end his or her own life if this person has gone bankrupt?"

[FN115]. Clark, supra note 50, at 408.

[FN116]. State Farm Mut. Auto. Ins. v. Campbell, 538 U.S. 408, 426 (2003) (noting that a central role of punitive damages is to condemn conduct leading to outrage and humiliation).

[FN117]. Feinberg, supra note 49, at 408; see also Galanter & Luban, supra note 96, at 1428 ("[Punitive damages are] the most important instrument in the legal repertoire for pronouncing moral disapproval of economically formidable offenders.").

[FN118]. Steiker, supra note 7, at 808.

[FN119]. See generally Barbara A. Babcock, Fair Play: Evidence Favorable to an Accused and Effective Assistance of Counsel, 34 Stan. L. Rev. 1133, 1140 (1982) (claiming that criminal trial symbolizes community values).

[FN120]. In fact, Carol Steiker agrees that it is not enough that the civil and criminal procedural regimes would be separated but also that criminal procedure should be more demanding than civil procedure. Steiker, supra note 7, at 808.

[FN121]. See Talia Fisher, The Boundaries of Plea Bargaining: Negotiating the Standard of Proof, 97 J. Crim. L. & Criminology (forthcoming 2008).

[FN122]. Dripps, supra note 61, at 204-05; see also Galanter & Luban, supra note 96, at 1457-58.

[FN123]. See generally Hylton & Khanna, supra note 60.

[FN124]. See supra text accompanying notes 109-17.

[FN125]. Dripps, supra note 61, at 209.

[FN126]. Id. at 215-16.

[FN127]. Id. at 215. Dripps's argument could provide a justification for a distinction between litigation to which the government is a party and litigation to which it is not, but not for the civil-criminal divide.

[FN128]. Robinson, supra note 97, at 694.

[FN129]. For discussion of the expressive justification, see supra text accompanying notes 47-59; see also Robinson, supra note 97, at 697-98 (arguing that the Supreme Court's tendency to impose constitutional limitations on civil commitment reflected in Foucha v. Louisiana, 504 U.S. 71 (1992), is dangerous because allowing the conviction of blameless persons undercuts the criminal law's moral credibility, and without moral credibility, criminal law would lose much of its power as a mechanism of compliance).

[FN130]. See Mark Kelman, A Guide to Critical Legal Studies 64-65 (1987).

[FN131]. Id. at 66.

[FN132]. Id.

[FN133]. Duff, supra note 65, at 36-37.

[FN134]. See Kelman, supra note 130, at 102-03.

[FN135]. We refer here to such scholars as Joseph Raz, Willam Galston, and Stephen Macedo. See Stephen Macedo, Liberal Virtues: Citizenship, Virtue, and Community in Liberal Constitutionalism 263-65 (1990); Joseph Raz, The Morality of Freedom (1986); William Galston, Defending Liberalism, 76 Am. Pol. Sci. Rev. 621, 627-29 (1982).

[FN136]. See, e.g., Macedo, supra note 135, at 50-65. Macedo criticizes Rawls's argument that the greatest moral question can be ignored in politics and that liberals can avoid assessing the truth and falsity of deeply held personal views. He argues instead that
[t]he personal moral convictions of citizens and other political actors should be engaged, as features of our public moral framework are worked out. Public justification involves not a rigid segmentation of public and private spaces of value, but a process of negotiation between shared public values and each person's entire set of values.
Id. at 63.

[FN137]. See Cohen, supra note 23, at 839-40.

[FN138]. Kelman, supra note 130, at 105.

[FN139]. Kennedy, supra note 90, at 90.

[FN140]. Id. at 119.

[FN141]. See Mann, supra note 17, at 1804 ("[P]unitive sanctions...are paradigmatically associated with the criminal law, but now characterize so much of the civil law that punishment no longer seems a distinctive attribute of the criminal law.").

[FN142]. Steiker, supra note 7, at 779.

[FN143]. Cheh, supra note 9, at 1330.

[FN144]. See, e.g., State Farm Mut. Auto. Ins. v. Campbell, 538 U.S. 408, 416 (2003).

[FN145]. See Fellmeth, supra note 4, at 17-19.

[FN146]. The first case to question the legislative label was probably United States v. Chouteau, 102 U.S. 603 (1880). There, the Court asserted that "[t]he term 'penalty' involves the idea of punishment, and its character is not changed by the mode in which it is inflicted, whether by a civil action or a criminal prosecution.... To hold otherwise would be to sacrifice a great principle to the mere form of procedure...." Id. at 611.

[FN147]. 116 U.S. 616 (1886).

[FN148]. Id. at 617-19.

[FN149]. Id. at 634.

[FN150]. Id. at 633-34.

[FN151]. See, e.g., Stephen J. Schulhofer, Two Systems of Social Protection: Comments on the Civil-Criminal Distinction, with Particular Reference to Sexually Violent Predator Laws, 7 J. Contemp. Legal Issues 69, 79 (1996) ("The Supreme Court's juvenile court jurisprudence illustrates the possibilities for this sort of a functional approach, in which the civil and criminal labels ultimately play no role in determining doctrinal outcomes.").

[FN152]. Clark, supra note 50, at 438.

[FN153]. John Hart Ely, Legislative and Administrative Motivation in Constitutional Law, 79 Yale L.J. 1205, 1212-17 (1970).

[FN154]. Clark, supra note 50, at 438. Despite these difficulties, Clark still posits that inquiries into the punitive motivation of the legislature are appropriate. Id.

[FN155]. George Fletcher, Comment, The Concept of Punitive Legislation and the Sixth Amendment: A New Look at Kennedy v. Mendoza-Martinez, 32 U. Chi. L. Rev. 290, 299 (1965).

[FN156]. 372 U.S. 144 (1963).

[FN157]. Id. at 167.

[FN158]. Id. at 168-69.

[FN159]. See, e.g., Smith v. Doe, 538 U.S. 84, 97 (2003); Seling v. Young, 531 U.S. 250, 268 (2001); Hudson v. United States, 522 U.S. 93, 99- 100 (1997); Allen v. Illinois, 478 U.S. 364, 370 (1986).

[FN160]. After listing these factors, the Court declined to apply them, and, instead, its determination that the loss of citizenship is a punishment that may not be imposed without all of the criminal procedural safeguards was based on the "conclusive evidence of congressional intent as to the penal nature of the statute." Mendoza-Martinez, 372 U.S. at 169.

[FN161]. Klein, supra note 105, at 719. For elaborate and insightful criticism of the Mendoza-Martinez test, see Fellmeth, supra note 4, at 36-41.

[FN162]. Attempts by legal scholars to define "punitive" sanctions have proven to be equally unsuccessful. One example is the test proposed by Carol Steiker for determining when an action is sufficiently punitive to merit criminal procedural safeguards. Based on her notion of punishment as blaming, Steiker
defines "punishment" using a four part test: (1) state intent to cause unpleasantness in an individual, that is not merely incidental to another goal; (2) the sanction is for a past offense; (3) the sanction is imposed by the state; and (4) the sanction expresses blame by the community. The fourth part of this test has a three-part subtest: blame occurs when (1) society resents the bad act; (2) the sanction is designed to tell the offender he misbehaved; and (3) the victim and society feels [sic] vindicated. Once it is decided that sanction constitutes punishment, the Court must answer three additional questions: (1) does the state intend to punish; (2) what is the effect of the state action on the defendant; and (3) how does the community view the state action?
Klein, supra note 105, at 719. However, as Klein convincingly argues, Steiker's test is indeterminate, overly complex, and replete with all the shortcomings of the Mendoza-Martinez test. See id. at 719-20. For another perspective on the sanction distinction, see Fellmeth, supra note 4.

[FN163]. Clark, supra note 50, at 384; see also Fellmeth, supra note 4, at 5 ("[T]he current position of the Supreme Court on the distinction between civil and criminal law is a hodgepodge of multifactor tests and genuflection to federal and state legislatures on questions of constitutional interpretation.").

[FN164]. See, e.g., Hawker v. New York, 170 U.S. 189 (1898) (holding that a statute preventing previously convicted felons from practicing medicine is regulatory, not punitive).

[FN165]. Compare Helvering v. Mitchell, 303 U.S. 391 (1938) (holding that the imposition of an increased tax assessment, amounting to fifty percent of the alleged deficiency, after criminal prosecution for tax evasion does not trigger the double jeopardy clause because it was not punitive), with Lipke v. Lederer, 259 U.S. 557 (1922) (holding that a double tax assessment, payable by anyone who manufactured illegal beverages without paying taxes, cannot be enforced in civil proceedings because it constitutes punishment).

[FN166]. Compare United States v. Halper, 490 U.S. 435 (1989) (holding that a civil fine for Medicaid fraud that was 220 times the government's actual loss was punitive because it served retributive and deterrent, rather than solely remedial, purposes), with Hudson v. United States, 522 U.S. 93 (1997) (holding that the double jeopardy clause does not prevent the federal government from bringing parallel criminal and civil forfeiture proceedings based on the same underlying events, because it only protects against the imposition of multiple criminal punishments, and civil forfeiture does not constitute a punishment), and United States v. Ursery, 518 U.S. 267 (1996) (holding that in rem civil forfeitures were neither punitive nor criminal for the purposes of the Double Jeopardy Clause). For a useful analysis of civil forfeiture, see Kevin Cole, Civilizing Civil Forfeiture, 7 J. Contemp. Legal Issues 249 (1996).

[FN167]. See infra notes 171-72.

[FN168]. Klein, supra note 105, at 680. An insightful solution to the civil-criminal puzzle, resting on the understanding that many sanctions in the regulatory state are hybrid, has been put forth by Kenneth Mann in his influential article, "Punitive Civil Sanctions: The Middleground Between Criminal and Civil Law." Mann, supra note 17. Mann proposes introducing middleground jurisprudence for the intermediate category of punitive civil sanctions, which could be either privately invoked or imposed by the state. Id. at 1799. Middleground jurisprudence draws on the paradigms of both civil and criminal law to form a hybrid jurisprudence, mixing the characteristics of both paradigms in new and innovative ways. This jurisprudence has a functional orientation, deriving from the punitive approach--an approach that seeks out the punitive aspects of any sanction and considers their necessary procedural implications. The courts should therefore develop tests and measures for "punitiveness" that would uncover the punitive purpose of a sanction and determine at what point heightened procedural protections are required. That is, the procedure warranted depends not just on whether a sanction is punitive, but also on the degree of its punitiveness. Id. at 1837-38. According to Mann, the Supreme Court's analysis in United States v. Halper, in which it concluded that a civil in personam fine for Medicaid fraud that was 220 times the government's actual loss served punitive rather than solely remedial purposes, is an example of a good functional analysis that "properly focus[es] on the fundamental jurisprudential issue: what is the degree of punitiveness required for state-invoked punitive civil sanctions to trigger heightened due process, and what are the indicators of such punitiveness?" Mann, supra note 17, at 1842-43 (construing Halper, 490 U.S. 435).
Much criticism has been leveled at the middleground approach, most of which is, in our opinion, without merit. John Coffee argues that middleground jurisprudence exacerbates the blurring of the line between civil and criminal law and contributes to a further encroachment of criminal law on civil law, which he considers the curse of modern criminal jurisprudence. See Coffee, supra note 55, at 1879-80. From a very different angle, Carol Steiker argues that middleground jurisprudence is more dangerous than advantageous for two reasons: first, it allows blaming individuals subject to only some (and it is hardly clear which) of the constraints that traditionally have accompanied criminal punishment; second, it undermines the usefulness of having a separate process as a forum of blaming. See Steiker, supra note 7, at 815-16. We take issue with both critiques for reasons elaborated upon earlier and discussed further in the final Part of this Article. There is, however, one line of criticism that successfully undermines the middleground approach. As Franklin Zimring insightfully observes, a number of key terms in Mann's article, including its most central term, "punitive civil sanction," are left undefined. As a result, the due process calculus offered by Mann as a substitute for the current civil-criminal jurisprudence is unspecified and, therefore, not workable. See Franklin E. Zimring, The Multiple Middlegrounds Between Civil and Criminal Law, 101 Yale L.J. 1901, 1901-03 (1992). In particular, Mann's failure (or, rather, inability) to define the term "punitive" exposes his argument to the entire arsenal of criticisms discussed above.

[FN169]. See Ursery, 518 U.S. at 285 n.2 ("It is hard to imagine a sanction that has no punitive aspect whatsoever.... [Such an] interpretation of Halper is both contrary to the decision itself and would create an unworkable rule inconsistent with well-established precedent."); see also Cheh, supra note 9, at 1356.

[FN170]. Klein, supra note 105, at 681. See also Hudson, 522 U.S. at 95- 96, 98, where the Court addressed the question of whether a criminal trial that followed debarment proceedings placed the defendant in double jeopardy. In concluding that it did not, the Court dismissed the notion that the constitutional protections apply independent of Congress's intent, unless the Court determines that the Mendoza-Martinez tests strongly and clearly dictate otherwise (which rarely, if ever, happens). Id. at 99-100, 104. Professor Mary Cheh endorses the Court's attitude, arguing that "a matter can only be criminal if formally intended to be and denominated as such: following the form of a criminal trial and calling a person to account for action clearly labeled as criminal by the legislature." Cheh, supra note 9, at 1360.

中国人民大学民商事法律科学研究中心| 民法学研究会 | 中国人民大学法学院 | 佟柔民商法发展基金 | 明德民商法研习社
征稿启事 投稿信箱:civillawruc@163.com