|
Article Excerpt
I. INTRODUCTION
Ever since the practice of plea bargaining took root in the Anglo-American legal world, the face of the criminal sphere has changed beyond recognition. The public model of criminal procedure, which places the procedural
aspects of the process beyond the reach of the defense and the prosecution, has given way to a new model: a semi-private paradigm that acknowledges the right of the parties to wield effective control over the procedural structure of the criminal trial. The adoption of plea bargains expresses a readiness to open the criminal arena to contractual ordering. (1) Many features of the criminal process have turned into default rules and "bargaining chips" (2) in the hands of the defense, including the Fifth Amendment right against self-incrimination, (3) the Sixth Amendment right to a jury trial, (4) and the right to appeal. (5) In exchange for deviation from and waiver of these rights, the defendant may receive various concessions from the prosecution, including mitigation in the charge or the sentence. This allows for the efficient resolution of the criminal case, and enables both prosecutor and defendant to avoid the costs of trial. (6) In addition to their attributed efficiency, plea bargaining practices can be normatively anchored in the defendant's autonomy of will, (7) and in his fight to effective control of his fate. (8)
This Article explores the boundaries of the negotiations under discussion and examines whether the current borderlines, set between alienable procedural rights (which the defendant may waive in exchange for sentence mitigation) and inalienable procedural rights (which may not be waived) can be justified. It makes a case for expanding the range of alienable procedural rights with regard to plea bargaining and for the extension of the boundaries of negotiation in the criminal arena. The test case on which I choose to focus, in order to examine the expansion of the negotiation borderlines, touches on the standard of proof. This Article examines the issue of changing the criminal standard of proof to a default variable from which the prosecution and defense can agree to deviate. According to the model under examination, the prosecutor would be able to "acquire" from the defendant a reduction of the standard of proof required during the criminal trial. The parties could agree that the standard of proof to which the prosecutor must adhere, in order to meet the burden of proof, will be less than "beyond a reasonable doubt." For example, the prosecution would be able to obtain a conviction if it established the case against the defendant in accordance with the civil standard of proof--preponderance of the evidence. The prosecution may also obtain "exemption" from the need to present corroborative evidence, or even reach an agreement for the reversal of the burden of proof between the parties, in which case the defendant takes it upon himself to prove his innocence beyond a reasonable doubt. In exchange for the greater risk of being convicted under a lower standard of proof, the defendant would face a lighter sentence in the event of conviction. The degree to which his sentence is reduced would reflect the extent to which the agreed-upon standard of proof deviated from the negotiable default constituted by the criminal standard of "beyond a reasonable doubt." It should be emphasized that the basis for stipulation and deviation from the standard of criminal proof would be consensual, expressing the mutual wishes of both parties to the judicial process. (9)
Under the current legal regime, the standard of proof is considered a constitutional safeguard of the criminal trial. (10) The fundamental principle, that the prosecution must bear the burden of proving all elements of guilt beyond a reasonable doubt as a prerequisite to conviction, dates back to the eighteenth century (11) and constitutes a "bedrock" principle of American criminal procedure. (12) The Supreme Court explicitly adopted this standard as a constitutional requirement in the 1970 case In re Winship. (13) The Court held that the Due Process Clauses of the Fifth and Fourteenth Amendments require the protection of the accused against conviction "except upon proof beyond a reasonable doubt of every fact necessary to constitute the crime with which he is charged." (14) Justice Frankfurter lauded the reasonable doubt rule as playing "a vital role in the American scheme of criminal procedure." (15) This quantum of proof is ubiquitous, and applies to every criminal trial. (16) Its significance is accentuated when contrasted with other procedural safeguards, such as the right to counsel or jury trial, which are not mandated in every criminal proceeding. (17)
In light of the fact that the standard of proof currently constitutes a nonnegotiable, fixed, and indivisible feature of the criminal process, the prosecution and defense have only two polar options to choose from when bargaining to settle the criminal case. The first is to conduct a full trial, placing the entire burden of proof upon the prosecution. The second is to enter a plea bargain, whereby the prosecution gains full exemption from having to prove its incriminating case. In other words, due to the indivisible, nonnegotiable nature of the criminal standard of proof, the existing model of plea bargaining is based upon self-incrimination by the defendant. (18) Such self-incrimination can relate to all areas of the criminal "dispute" (as occurs in plea bargains of the sentence bargaining (19) type) or, alternatively, it can apply on certain portions only (as occurs in bargains of the charge bargaining (20) type). In this sense, under current plea bargains the prosecution obtains a vertical exemption from the requirements of proof: a full exemption vis-a-vis some component of the factual case against the defendant, or the case in its entirety.
The proposed model would enhance the possibilities available to both parties by adding the option of a horizontal exemption from the requirements of proof. In accordance with this alternative option, the defendant would not make a full confession to any single part of the dispute, but would make it possible for the prosecution to reduce the standard of proof necessary for criminal conviction across the board. This new version of plea bargains would enable the prosecution to obtain partial exemption from the burden of proof for the entire dispute. To illustrate a possible implication of the proposed model, let us take a hypothetical rape case and assume significant disparities between the parties' subjective evaluations of its outcome in court. Let us assume the prosecution is of the opinion that under the "beyond a reasonable doubt" rule it has a 40% chance of obtaining a criminal conviction. Let us further assume that the prosecutor estimates that, if convicted, the defendant will be sentenced to twenty years in prison. The defendant, for his part, assesses his chances of being convicted completely differently. In his opinion, the probability of his being convicted under the criminal standard of proof is only 20%, due to an alibi claim he wishes to raise in court. He estimates that, if convicted, he will be sentenced to fifteen years in prison. In order to simplify the argument at this stage, I will focus only on the expected sanction, as perceived by each side, disregarding additional parameters such as the decline or increase in the marginal costs associated with each year of imprisonment over successive years. (21) I will also assume reasonable costs of trial. These issues will be discussed in detail later in the Article. (22)
Under the abovementioned set of assumptions, the lack of congruity between each party's assessment (23) of the expected punishment can stand in the way of reaching a full plea bargain. (24) The prosecutor would offer the defendant a prison term of slightly less than eight years (the length of the expected sentence times the probability of conviction, minus the costs of trial) in exchange for his self-incriminating guilty plea. The defendant, for his part, would demand that, if convicted, he be sentenced to a maximum of slightly more than three years (the expected punishment, in his opinion, plus costs of trial). Thus, from the point of view of the prosecutor, the reduction in years demanded by the defendant, in exchange for his self-incriminating guilty plea, is too large. But from the defendant's point of view, the prosecution's demand of a full guilty plea is too high a price to pay for saving trial costs. Therefore, in the current situation of a nonnegotiable standard of proof ("beyond a reasonable doubt") the parties will most likely seek the trial option. For such cases, the proposed model can open the door to settlements that would benefit both parties. Let us now introduce a new variable, according to which the defendant can "exchange" concessions in the standard of proof for mitigation of the sanction in case of conviction. Let us further assume that, following the move to a civil standard of proof ("preponderance of the evidence"), the expected sentence envisioned by each of the parties changes as follows: The prosecution calculates that the move to a civil standard of proof will result in a 90% probability of conviction. This is because from the prosecution's point of view the difficulties of proof relate mostly to the elimination of any reasonable doubt. The defendant, for his part, believes that the move to a "preponderance of the evidence" standard of proof will not dramatically aggravate his situation because of the alibi claim that he is keeping under his hat. In his opinion, the likelihood of conviction through the move to a "civil" standard of proof will only increase by 5% (from a 20% to a 25% chance of conviction). In this scenario, the prosecution's proposal to request a ten-year sentence upon conviction, in exchange for moving to a "civil" standard of proof will enable both parties to increase their expected utility. (25) The prosecution's expected sanction will rise to nine years (an improvement on the original option of eight years). The defendant will face an expected sanction of only two-and-a-half years (an improvement on the original option of three years).
As will be demonstrated later, the proposed model can serve as an attractive alternative not only to a full trial, but also to the existing form of plea bargaining. (26) I will also demonstrate that when variables such as litigation costs and attitude toward risk are taken into account, the proposed model can improve both parties' expected utility (in relation to the current plea bargain or to trial according to the criminal standard of proof) even under conditions of full information. (27) At this point, I will only make an intuitive claim for the model: the proposed model offers each party the option to "insure" itself partially against the outcome of the criminal trial. Because of the move to a lower standard of proof, the "danger of acquittal" to which the prosecution is exposed is reduced, due to a somewhat increased probability of conviction. However, this risk is not entirely eliminated because the prosecution does not obtain a full guilty plea, as would be the case under the current full plea bargain practice. The defendant, too, is partially insured, because the extent of the sanction he faces upon conviction has diminished, but not to the "maximal" penalty reduction level, as in the current full plea bargain. Of course, the main reason for choosing "partial insurance" (the proposed model) over "full insurance" (the current plea bargain) stems from the costs of each of these policies: Obtaining full insurance against the risk of a criminal conviction entails a high price for each party. In the event that the bargain is accepted, the defendant must relinquish all possibility of acquittal. The prosecution, for its part, must "pay" in maximal sanction reduction. In other words, according to the existing model of plea bargaining (which is subject to the nonnegotiable standard of proof), the degree of risk involved in a criminal proceeding is limited to polarized solutions: carrying the full risk of a legal process under a "beyond a reasonable doubt" standard of proof or comprehensive cancellation of risk by eliminating the need for a fact-finding process. In contrast, the proposed model allows far greater flexibility in the apportionment of risk within the framework of the judicial process, enabling more optimal solutions. It sets the stage for partial diminution of the risks involved in adjudication. Thus, it enables the parties to regulate the degree of risk to which they are willing to expose themselves, on the basis of its inherent cost. Accordingly, the parties can concoct their own mix of optimal risk and cost, unfettered by extreme solutions. (28)
Similar claims can be made for the reduction of the costs of trial. The proposed model does not completely eliminate trial costs, as does the existing plea bargain, but it enables the parties to reallocate costs and lower the overall cost of trial. This point will be analyzed more thoroughly in upcoming sections, which will also be devoted to the question of why partial reduction in trial costs may be more beneficial to the parties, under certain circumstances, than their full elimination. (29) At this point, I will briefly explain the underlying intuition: One can posit a situation where the task of proving the final X percent (30) of the prosecution's case requires a vast investment in resources on its part, such as the monetary cost of obtaining evidence from out-of-state witnesses, the emotional price paid by child witnesses, or the cost of revealing evidence where the prosecution wants to preserve the cover of police agents. The prosecution may regard this evidence as crucial for proving its case "beyond a reasonable doubt" but find it unnecessary when a move to a lesser standard of proof has been made. The reason for opting for a cheaper judicial process over a full plea bargain (which cuts down on the full costs of trial involved) stems from the differential prices of the two alternatives, in terms of sentence reduction. (31)
The partial reduction of risk and costs vis-a-vis the criminal trial is not unique to the proposed model and is not a novel phenomenon on the criminal procedure scene. There are other cases in which the parties opt to reduce the risks and costs of the judicial process rather than completely eliminate them. One example is the flourishing market of deals to "turn State's evidence" whereby the state's witness submits information incriminating others in exchange for mitigation of his punishment. (32) From the prosecution's point of view, the deal struck with the state's witness is a partial reduction of the risks and costs of the criminal proceeding against the chief defendant. In theory, the prosecution could hammer out a full plea bargain with the chief defendant, thus precluding the risks and costs of a trial. However, from the prosecution's perspective, the price involved (a lenient sentence for the chief defendant) would be too high. Instead, the prosecution prefers to partially reduce its risks and costs by recruiting witnesses on its behalf. In exchange it pays a lower price in the form of sentence mitigation or other benefits bestowed upon the state's witness. Another clear example of reducing the risks and costs of a criminal trial, as distinguished from their complete abrogation, is reflected in agreements whereby the defendant waives his right to appeal his sentence in exchange for concessions from the prosecution. (33) The fact that such "partial insurance" mechanisms against the outcome of the trial exist indicates the viability of interim solutions for contending with the risks and costs of criminal proceedings. The changes proposed in this model aim to expand the spectrum of options available to both parties by removing an additional stumbling block from their way--the fixed, indivisible standard of proof.
Concomitant to this discussion, an important point must be clarified: The discussion thus far may have created the false impression that the proposal to change the standard of proof to a negotiable variable is aimed at opening the parties to an endless, sequential continuum of proof standards from among which they can choose--i.e., that in accordance with the proposed model, the parties would be able to contract for a precise degree of probability for conviction, such as 85%, 78%, or 70%, and that this statistical rate could be validated by the courts. This is not the case, for there is indeed room to claim that accurate arithmetic evidentiary standards may not be practically viable. Judicial decisions are not statistical by nature and cannot be quantified accordingly. Courts typically lack the necessary tools and statistical information for precise calculation of the probability of guilt. Instead, courts draw their conclusions from categorical generalizations that transform legal decisions into rough estimates not subject to accurate statistical measurement. Support for this concept can be found in the fact that no exact statistical quantification exists for "reasonable doubt." (34) Since from a practical point of view the exact statistical measure has no concrete significance that can guide the courts (and neither does the transition from "beyond a reasonable doubt" to an 80% standard of proof and thence to a 75% standard of proof), the parties must formulate a standard of evidence that will be widely applicable. This requires the transition to categories such as "preponderance of the evidence" or "clear and convincing evidence." We can assume that under the proposed system, typical agreements will not be based upon shifts of single percentages one way or another in the standard of proof. Rather, these agreements will have the nature of calculable conversions that do not lend themselves to precise quantification, such as the move from the criminal to the civil standard of proof, an agreed-upon waiver of the need for corroborative evidence, or reversal of the burden of proof between defendant and prosecution.
The discussion of this Article will be divided into two main parts. The first part will explore the practical viability of the proposed model. The discussion will be devoted to identifying some of the situations where deals to provide concessions in the standard of proof might emerge or are likely to be formulated. The second part of this Article will address the normative desirability of such agreements. The legitimacy of the proposed model will be explored in reference to existing plea bargaining practices. This requires a brief explanation. The adoption of plea bargaining, and the legal regulation to which it is subject, (35) reflect a normative choice, the basis of which will not be re-opened for discussion here. Rather, this choice will be treated as an axiomatic starting point for my argument. The normative examination of the proposed model will focus on locating the points where the model deviates, either qualitatively or quantitatively, from plea bargains in their present form, and on exploring the ramifications of these dissimilarities regarding the proposed model's normative desirability.
II. VIABILITY OF THE PROPOSED MODEL
This Part is dedicated to the claim that the proposed model has practical significance: that, under certain circumstances, the negotiating parties will prefer the interim solutions available due to the implementation of such a model over the existing end results of plea bargains in their present form or full implementation of the judicial apparatus. When examining the viability of the proposed model, one must first ask why the parties choose to engage in plea bargaining in the first place. What utility do they derive from such bargains, and what aims do they seek to advance through them? Against this background one can define those situations in which the proposed model will further each party's attainment of its goals, as compared with the binary end solutions available at present. Therefore, at the outset I will discuss the advantages of plea bargaining, as viewed by both the defendant and the prosecution.
A. ADVANTAGES OF PLEA BARGAINING FOR THE DEFENDANT
The defendant's main motivation for engaging in plea bargains with the prosecution is aimed at reducing the overall costs he faces for allegedly committing an offense. These costs include both the criminal punishment and its accompanying trial costs. Punishment comprises the formal legal sanction imposed on the defendant (e.g., years of imprisonment and/or fines) as well as reputation and opportunity costs (such as loss of income). (36) The trial costs encompass monetary and emotional resources, the time spent in conducting a full trial, and the cost of facing uncertainty (for risk-averse defendants). (37) By negotiating a plea bargain, the defendant can acquire a "discount" in the criminal sanction (conviction for a lesser crime or a lighter sentence), and also avoid the accompanying trial costs.
B. ADVANTAGES OF PLEA BARGAINING FOR THE PROSECUTION
The prosecution is restricted from a budgetary viewpoint. (38) The scope of its resources does not allow it to conduct a full trial with regard to every suspect against whom sufficient evidence has been amassed. (39) Plea bargains enable prosecutors to maintain control over their caseloads by reducing enforcement costs per case (40) and by minimizing the risk of acquittal. (41) Such bargains pave the way for increasing the overall number of offenders prosecuted, (42) thus enabling the prosecution to further its goals of deterrence, incapacitation, and retribution. (43) In addition, plea bargains shorten the period of time between the criminal incident and the act of punishment. (44)
In light of the attempt to increase general enforcement, the prosecution's motives in plea bargaining with a particular defendant can best be described as directed toward maximizing the expected punishment or "cost of offense" borne by the specific defendant (within the limits of the formal legal sanction). The prosecution will agree to enter into a plea bargain if the overall expected "price of the offense" paid by the particular defendant exceeds the expected price from alternative defendants through equal investment of resources (45) and also exceeds the expected sentence if the particular defendant is tried, after deduction of the prosecution's trial costs.
C. DEVIATION FROM THE CRIMINAL STANDARD OF PROOF
After defining the goals of both the prosecution and the defendant in the resolution of the criminal case, I will now identify the circumstances under which one can expect the proposed bargain--a horizontal stipulation of the criminal standard of proof--to further these goals (i.e., to offer the parties better options than the ones currently available to them: full plea bargain or full trial according to the "beyond a reasonable doubt" standard of proof). As discussed above, one can define the benefits accruing to the prosecution and the defendant on the basis of two variables: the cost of the offense for the defendant (as defined above) and the effective evidentiary requirements imposed on the prosecution to uphold the burden of persuasion. It may be claimed that the benefits to the prosecution are positively influenced by the cost of the offense and negatively influenced by the effective evidentiary requirements for conviction. Benefits to the defendant, on the other hand, are negatively influenced by the cost of the offense and positively influenced by the proof requirements imposed on the prosecution.
Within the framework of the existing plea bargain, the defendant and prosecutor exchange elimination of proof requirements for maximal punishment mitigation: the plea bargain will be formed when such evidentiary waiver is more valuable, in terms of units of punishment, for the prosecution than it is for the defendant. The proposed model, too, depends on both parties exchanging mitigation in proof requirements for a lighter punishment. The difference between this model and current plea bargains is that the model under discussion is not limited to the exchange of these variables as one indivisible block, but provides for partial and gradual exchange of units of standard of proof for units of punishment. In light of the feasibility of separating the "array of punishment" and the "array of proof standard" into units, one can thus refine the definition of the utility function of both parties in the following manner: the greater the degree of overall punishment and the lower the standard of proof needed to uphold the burden of persuasion, the more benefit accrues to the prosecution--and the reverse applies with regards to the defendant.
Because benefits to the parties are influenced by the degree of punishment and the standard of proof, and because the groupings of "punishment mitigation" and "proof standards" can be separated, it thus becomes possible to define specific degrees of exchange--the Marginal Rate of Substitution ("MRS") (46)--for each party between units of punishment and "standard of proof" units. The MRS for the prosecutor measures the number of additional units of punishment which he is prepared to relinquish in exchange for an additional unit of reduction in the standard of proof, in such a way that utility will not be affected. The MRS for the defendant describes the number of units of reduction in the standard of proof that he is willing to forgo in exchange for the reduction of an additional unit of punishment, while maintaining a fixed level of utility. The proposed model is based on the assumption that in some situations the MRS of the parties between units of punishment and units of standard of proof will equalize at an intermediate point set between a full trial (under a "beyond a reasonable doubt" standard) and a full plea bargain. As will be demonstrated in the following Part, under such circumstances both prosecutor and defendant would opt for partial reduction of the standard of proof, in exchange for a lesser sentence, rather than engage in a full plea bargain or conducting a full trial.
D. MARGINAL RATE OF SUBSTITUTION BETWEEN UNITS OF PUNISHMENT AND STANDARDS OF PROOF
The parties' MRS between units of punishment and units of standard of proof are influenced by a range of factors and variables. One such variable, briefly mentioned in the example of the rape case, is each party's subjective assessment of its chances of success at trial under various standards of proof. Although this factor has many practical implications, the proposed model is not limited to conditions of asymmetric information and differential subjective evaluations. Even under conditions of full information, the MRS of both parties may equalize at interim points due to the effect of additional variables, such as the prosecution's marginal cost of gathering evidence and managing the trial, the defendant's attitude toward risk, and the signaling effects of engaging in the proposed bargain. In order to illustrate paradigmatic situations where the proposed model would be viable, I will examine the possible impact of each of these factors on the MRS for the parties, and on the willingness of the prosecutor and the defense to enter into the proposed interim deals.
1. The Combination Effect of Trial Costs and the Defendant's Attitude Toward Risk
a. Trial Costs
The prosecution must invest resources in order to win its case in court. Evidence does not appear of itself. Witnesses must be located and coached before they testify. Expert witnesses and capable attorneys must be recruited to conduct the prosecution. Due to the fact that the gathering of evidence bears a price tag for the prosecution, the costs of trial are not exogenous to the proof requirements: The higher the standard of proof required to establish criminal culpability, the more resources must be invested by the prosecution in order to meet the evidentiary demands. (47) In other words, from the prosecution's point of view, the allocation of resources for the management of the trial and for the gathering of evidence is contingent upon the standard of proof necessary for the establishment of a criminal conviction. The move to a lower standard of proof, such as "preponderance of evidence," will reduce the prosecution's cost of trial, but only partially. The "interim" trial under this lower standard of proof will still be more expensive than the full plea bargain, but cheaper than the full trial according to the "beyond a reasonable doubt" standard of proof. The move to a full plea bargain, on the other hand, enables the prosecution to save maximally on case management and evidence-gathering as the cost of trial drops to the minimum, due to the fact that under such bargains the prosecution is effectively exempted from the need to prove its incriminating case. Therefore, maximum savings in trial costs will create an incentive, on...
|
