Remorse, cooperation, and "acceptance of responsibility": The structure, implementation, and reform of section 3E1.1 of the federal sentencing guidelines

Northwestern University Law Review

Volumn 91, Issue 4, 1997, Pages 1507-1573

  O'Hear, Michael M ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0346045058     PISSN: 00293571     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (258)

1. Daniel J. Freed, Federal Sentencing in the Wake of the Guidelines: Unacceptable Limits on the Discretion of Sentencers, 101 YALE L.J. 1681, 1715 (1992).
2. U.S. SENTENCING GUIDELINES MANUAL § 3E1.1(a) (1994). Currently, the benefit for acceptance of responsibility is equal to a reduction of two offense levels, U.S.S.G. § 3E1.1(a), with a possibility of reduction by a third level under certain circumstances, U.S.S.G. § 3E1.1(b). All subsequent citations to the 1994 Guidelines Manual will be made using the abbreviated citation form "U.S.S.G." with no specific year denoted. Citations to the 1987 Guidelines Manual will also employ the "U.S.S.G." form, but with the date noted in parentheses.
3. 62 Fed. Reg. 179-81 (1997). The first of these proposed amendments (amendment 24) would represent the most substantial departure from the present structure of the guideline, clearly shifting section 3E1.1 in the direction of what I term the cooperation paradigm, yet not fully resolving the ambiguities in the guideline. Amendment 24 is addressed in greater detail below in subpart II.D. Amendment 25 addresses a circuit split over the narrow issue of whether the commission of a crime dissimilar to the offense of conviction pending sentencing may preclude the award of the section 3E1.1 adjustment. If adopted, amendment 25 will overrule the Sixth Circuit's decision that only similar or related criminal conduct may be considered in the acceptance inquiry. See United States v. Morrison, 983 F.2d 730, 735 (6th Cir. 1993).
4. U.S. SENTENCING COMM'N, STAFF DISCUSSION PAPER: CHAPTER THREE ADJUSTMENTS app. 2 (1995) [hereinafter STAFF DISCUSSION PAPER] (indicating that 84.4% of defendants receive the section 3E1.1 discount).
5. 33 F.3d 175 (2d Cir. 1994).
6. Id. at 177. Although Echevarria was not a licensed physician, he advertised and operated a practice in psychiatry and neurology, prescribed drugs under a false registration number and applied for reimbursement for his services from private and governmental insurers. Id.
7. Id. at 179.
8. Id. at 177 n.2.
9. Echevarria was sentenced to 70 months, near the middle of the recommended guideline range for his offense level (26) and criminal history category (I). Id. at 178. If his offense level has been reduced by two, and if the judge still would have placed him in the middle of the suggested range, Echevarria's sentence would have been 57 months.
10. 922 F.2d. 1026 (2d Cir. 1990), cert. denied, Tarsell v. United States, 500 U.S. 941 (1991).
11. Id. at 1029. Under 19 U.S.C. § 1955(b)(1)(i) (1988), operation of a gambling business is a federal offense only if it is proscribed by the law of the state in which it is conducted. Cook, 922 F.2d at 1035. The government maintained that the federal Indian Gaming Regulatory Act did not implicitly repeal § 1955. According to the government, the defendants were subject to state law, thought their gambling operations were conducted on Indian land. Id. at 1033-34.
12. Id. at 1029. Under 19 U.S.C. § 1955(b)(1)(i) (1988), operation of a gambling business is a federal offense only if it is proscribed by the law of the state in which it is conducted. Cook, 922 F.2d at 1035. The government maintained that the federal Indian Gaming Regulatory Act did not implicitly repeal § 1955. According to the government, the defendants were subject to state law, thought their gambling operations were conducted on Indian land. Id. at 1033-34.
13. Id. at 1039 (Lasker, J., dissenting) (quoting from a sentencing transcript).
14. Id. at 1037 (quoting United States v. Royer, 895 F.2d 28, 30 (1st Cir. 1990)).
15. See, e.g., Letter from Maryanne Trump Barry, Chair of the Committee on Criminal Law of the Judicial Conference of the United States, to Judge Richard P. Conaboy, Chairman of the United States Sentencing Commission 1 (Dec. 5, 1995) (copy on file with author) [hereinafter Barry Letter]. This letter, discussing proposed changes to section 3E1.1, observes, "The case law indicates continued confusion with the way the acceptance of responsibility guideline has come to be interpreted," and further notes, "[T]he various factors which comprise the current 'acceptance' adjustment interact with each other . . . to generate needless litigation." Id. Among Guidelines Issues, section 3E1.1 accounts for the third highest number of appeals by both defendants and the government. Id.
16. Id.
17. William W. Wilkins, Jr., Plea Negotiations, Acceptance of Responsibility, Role of the Offender, and Departures: Policy Decisions in the Promulgation of Federal Sentencing Guidelines, 23 WAKE FOREST L. REV. 181, 190 (1988).
18. Id. at 191.
19. See Bordenkircher v. Hayes, 434 U.S. 357, 364 (1978) (Although "confronting a defendant with the risk of more severe punishment clearly may have a 'discouraging effect on the defendant's assertion of his trial rights, the imposition of these choices [is] an inevitable' - and permissible - 'attribute of any legitimate system which tolerates and encourages the negotiations of pleas.'") (quoting Chaffin v. Stynchcombe, 412 U.S. 17, 31 (1972)).
20. See Stephen Breyer, The Federal Sentencing Guidelines and the Key Compromises upon Which They Rest, 17 HOFSTRA L. REV. 1, 29 (1988) (noting that many witnesses argued to Commission that "plea bargaining was highly desirable and practically necessary").
21. Id. at 29-30.
22. Wilkins, supra note 16, at 190.
23. Breyer, supra note 19, at 28. The constitutional analysis of plea discounts will be discussed at greater length in Subpart V.G infra. The Commission's belief in 1986 and 1987 seems to have been that "[i]nvesting the Court with discretion to mitigate the sentence by a specified amount or amounts, rather than directing specified 'guilty plea credit' in all cases, would very much undercut any Constitutional objection to the plan." Wilkins, supra note 16, at 191 n.65 (quoting U.S. SENTENCING COMM'N, PUBLIC HEARINGS ON PLEA AGREEMENTS IN WASHINGTON, D.C. 3 (Sept. 23, 1986) (testimony of William F. Weld, Assistant Att'y Gen., Criminal Div., U.S. Dep't of Justice)) (emphasis added).
24. Wilkins, supra note 16, at 191.
25. U.S.S.G. § 3E1.1(a).
26. Breyer, supra note 19, at 29.
27. I cannot ascertain whether this functionalist maneuver was consciously made by the Commission. I offer this account of the genesis of section 3E1.1 less as history than as a helpful analytic device which is not inconsistent with the sources. Precisely how section 3E1.1 grew out of the plea discount debate - politically or conceptually - is not clear; that it grew out of the plea discount debate is more clear. Breyer and Wilkins, both framers of the Guidelines, strongly imply as much in their separate accounts of acceptance of responsibility. Breyer, supra note 19, at 28-29; Wilkins, supra note 16, at 190-92. Moreover, in the 1986 draft of the Guidelines, the commentary on the acceptance of responsibility provision includes a request for public comment on whether guilty pleas should receive an automatic sentence discount. U.S. SENTENCING COMM'N, PRELIMINARY DRAFT OF SENTENCING GUIDELINES § B321 commentary (Sept. 1986) [hereinafter 1986 DRAFT]. Finally, the 1987 report on the Guidelines discusses section 3E1.1 under the heading "Guilty Pleas, Applicable Guidelines and Policy Statements." U.S. SENTENCING COMM'N, SUPPLEMENTAL REPORT ON THE INITIAL SENTENCING GUIDELINES AND POLICY STATEMENTS 48, 50 (1987).
28. 1986 DRAFT, supra note 26, § B321.
29. id.
30. Id.; see also Brady v. United States, 397 U.S. 742, 753 (1969) (stating a defendant "demonstrates by his plea that he is ready and willing to admit his crime and to enter the correctional system in a frame of mind that affords hope for success in rehabilitation over a shorter period of time than might otherwise be necessary").
31. 1986 DRAFT, supra note 26, § B321, commentary.
32. U.S.S.G. § 3E1.1(a).
33. 1986 DRAFT, supra note 26, § B321.
34. See U.S. SENTENCING COMM'N, ACCEPTANCE OF RESPONSIBILITY WORKING GROUP 33 (1991) [hereinafter Working Group] ("The current guideline takes an 'all or nothing' approach. . . . An alternative would be to allow different defendants to qualify for a different level of reduction. This would allow courts to distinguish between defendants who admit their wrong-doing and those who do something more . . . .").
35. U.S.S.G. § 3E1.1(b). According to the Commentary to section 3E1.1, subsection (b) applies only to defendants with high offense levels because "At offense level 15 or lower, the reduction in the guideline range provided by a 2-level decrease in offense level under subsection (a) (which is a greater proportional reduction in the guideline range than at higher offense levels due to the structure of the Sentencing Table) is adequate for the court to take into account the factors set forth in subsection (b) within the applicable guideline range." Id commentary (backg'd.). No doubt, the Commission was partially motivated by concerns that the two-level reduction provided an insufficient incentive to plead guilty for defendants facing lengthy jail terms. See WORKING GROUP, supra note 33, at 2 (noting that one of the four areas of concern for the working group was "whether the guideline provides a significant enough offense level reduction, especially for defendants whose offense levels are relatively high"); id. at 4 (noting that "defendants convicted of more serious offenses tended to plead guilty less often than those convicted of less serious offenses"); id. at 32 (discussing possibility of adding a third point to section 3E1.1 adjustment in order to provide "a greater incentive, across the board, for defendants to accept responsibility"). Among defense counsel and prosecutors, and to some extent among probation officers and judges, there was a pervasive perception that the acceptance of responsibility discount was too low. . . . [W]here offenders faced terms that would keep them in prison with no hope of parole for fifteen to twenty years, the possibility of getting out three to five years earlier was thought to be an insignificant inducement. Stephen J. Schulhofer & Ilene H. Nagel, Negotiated Pleas Under the Federal Sentencing Guidelines: The First Fifteen Months, 21 AM. CRIM. L. REV. 231, 267 (1989).
36. U.S.S.G. § 3E1.1(b). According to the Commentary to section 3E1.1, subsection (b) applies only to defendants with high offense levels because "At offense level 15 or lower, the reduction in the guideline range provided by a 2-level decrease in offense level under subsection (a) (which is a greater proportional reduction in the guideline range than at higher offense levels due to the structure of the Sentencing Table) is adequate for the court to take into account the factors set forth in subsection (b) within the applicable guideline range." Id commentary (backg'd.). No doubt, the Commission was partially motivated by concerns that the two-level reduction provided an insufficient incentive to plead guilty for defendants facing lengthy jail terms. See WORKING GROUP, supra note 33, at 2 (noting that one of the four areas of concern for the working group was "whether the guideline provides a significant enough offense level reduction, especially for defendants whose offense levels are relatively high"); id. at 4 (noting that "defendants convicted of more serious offenses tended to plead guilty less often than those convicted of less serious offenses"); id. at 32 (discussing possibility of adding a third point to section 3E1.1 adjustment in order to provide "a greater incentive, across the board, for defendants to accept responsibility"). Among defense counsel and prosecutors, and to some extent among probation officers and judges, there was a pervasive perception that the acceptance of responsibility discount was too low. . . . [W]here offenders faced terms that would keep them in prison with no hope of parole for fifteen to twenty years, the possibility of getting out three to five years earlier was thought to be an insignificant inducement. Stephen J. Schulhofer & Ilene H. Nagel, Negotiated Pleas Under the Federal Sentencing Guidelines: The First Fifteen Months, 21 AM. CRIM. L. REV. 231, 267 (1989).
37. U.S.S.G. § 3E1.1(b).
38. Accordingly, some appellate courts have held that the subsection (b) reduction is much less discretionary than the subsection (a) reduction. E.g., United States v. Eyler, 67 F.3d 1386, 1390-91 (9th Cir. 1995); United States v. Covarrubias, 65 F.3d 1362, 1368 (7th Cir. 1995).
39. The Commission is presently considering two amendments that would alter subsection (b). See supra note 3. The more modest amendment would offer the third-point benefit to defendants with offense level below 16. United States Sentencing Commission, Sentencing Guidelines for United States Courts 62 Fed. Reg. 152, 180 (1997) (proposed Jan. 2, 1997) [hereinafter cited by amendment]. A more sweeping amendment would in a sense reverse the nature of the inquiries under subsections (a) and (b). See Amendment 24, 62 Fed. Reg. at 179. Under this amendment, the two-point discount under subsection (a) would turn more clearly on objective aspects of post-offense conduct, particularly pleading guilty, while the third point under subsection (b) would depend on a more open-ended, remorse-oriented inquiry. See infra subpart II.D.
40. See, e.g., U.S.S.G. § 3E1.1 application note 1 ("In determining whether a defendant qualifies under subsection (a), appropriate considerations include, but are not limited to, the following . . . ."); id. application note 5 (emphasis added) ("The sentencing judge is in a unique position to evaluate a defendant's acceptance of responsibility.").
41. Id. application note 2.
42. Id. Notwithstanding the strong language of note two, some sentencing judges seem quite generous in awarding the section 3E1.1 reduction to defendants who go to trial. See infra subpart IV.B (discussing differences between districts in the correlation between the mode of conviction and the award or denial of acceptance-of-responsibility credit).
43. U.S.S.G. § 3E1.1 application note 4.
44. Id. The Commission recently considered an amendment to application note 4 that would impose a similarly strong presumption accepting responsibility when a defendant commits a new offense while awaiting trial or sentencing. See Amendment 25, 62 Fed. Reg. at 180.
45. See THOMAS HUTCHISON & DAVID YELLEN, FEDERAL SENTENCING LAW AND PRACTICE 517 (2d ed. 1994) ("There is nothing inherently inconsistent with finding that a defendant who one point obstructed an investigation later regretted that action and accepted responsibility for the offense.").
46. U.S.S.G. § 3E1.1 application note 2.
47. An analysis of earlier versions of section 3E1.1 underscores the strength of the current presumption against giving the a/r adjustment to defendants who go to trial. The a/r provision of the 1986 Preliminary Draft of the Guidelines made no mention of particular inferences to be drawn from the decision to go to trial, and stated, "An offender may qualify for a reduction under this section without regard to whether the offender's conviction is based upon a guilty plea or a finding of guilty by a court or jury." 1986 DRAFT, supra note 26, § B322 (emphasis added). The 1987 Draft employed similar language, U.S.S.G. § 3E1.1(b) (1987), though more clearly recognized the evidentiary value of mode of conviction. The pertinent application note in 1987 stated, "Conviction by trial does not preclude a defendant from consideration under this section. A defendant may manifest sincere contrition even if he exercises his constitutional right to a trial." Id. application note 2 (emphasis added). Thus, not only were the early drafts more neutral in their language with respect to defendants who went to trial, but they expressly envisioned the very scenario precluded by the current draft: granting an a/r adjustment to the defendant who goes to trial, but who afterwards manifests sincere remorse. In this respect, the early versions of section 3E1.1 seem to have been closer than the current version to the remorse paradigm. Treatment of guilty pleas has undergone similar modification.
48. U.S.S.G. § 3E1.1 application note 1. These considerations are as follows: (a) truthfully admitting the conduct comprising the offense(s) of conviction, and truthfully admitting or not falsely denying any additional relevant conduct . . . (b) voluntary termination or withdrawal from criminal conduct or associations; (c) voluntary payment of restitution prior to adjudication of guilt; (d) voluntary surrender to authorities promptly after commission of the offense; (e) voluntary assistance to authorities in the recovery of the fruits and instrumentalities of the offense; (f) voluntary resignation from the office or position held during the commission of the offense; (g) post-offense rehabilitative efforts . . . and (h) the timeliness of the defendant's conduct in manifesting the acceptance of responsibility. Id.
49. Id. I see greater emphasis in these provisions on the remorse vision than the cooperation vision; I recognize, though, that all eight enumerated considerations are, to a greater or lesser extent, consistent with both visions. The "truthful admission" factor has been among the more controversial aspects of § 3E1.1, due to concerns that it may lead to infringements on the privilege against self-incrimination. Consequently, the language of this provision has been modified several times since 1987. See infra section III.C.1.
50. U.S.S.G. § 3E1.1 application note 2. Curiously, notwithstanding the apparent disavowal of remorse and contrition in the guideline itself, appellate courts have generally seen these principles as the primary focus of section 3E1.1. See infra subpart III.A.
51. Id. application note 1. While the Commission has not provided clear guidance as to how the eight factors are to be weighted, appellate case law generally holds that no single factor is either necessary or sufficient for a demonstration of accepting responsibility. See, e.g., United States v. Russell, 913 F.2d 1288, 1295 (8th Cir. 1990); United States v. Knight, 905 F.2d 189 (8th Cir. 1990). But cf. United States v. Austin, 17 F.3d 27, 30 (2d Cir. 1994) (implying that acceptance of responsibility may only be denied based on one of the eight factors listed in note one). For a discussion of the Austin case, see infra section III.C.1.
52. U.S.S.G. § 3E1.1 application note 5.
53. 1986 DRAFT, supra note 26, § B321 commentary (emphasis added).
54. This remains the predominant attitude among appellate courts. See, e.g., United States v. Zanni, 64 F.3d 654, 654 (1st Cir. 1995) (upholding denial of a/r adjustment when district court "was not persuaded that [defendant] was genuinely contrite"); United States v. Boothe, 994 F.2d 63, 71 (2d Cir. 1993) (upholding denial of adjustment when district court had relied on the fact that defendant "failed to prove the sincerity of his acknowledgment of guilt"). See also infra subpart III.A.
55. The First Circuit, in an opinion emphasizing the remorse paradigm, has captured the spirit of this observation: The inquiry into acceptance of responsibility is necessarily factbound. In deciding whether a defendant is entitled to a reduction on this score, a district court must weigh a multitude of factors, some objective, some subjective. Credibility and demeanor evidence play a crucial role in determining whether a person is genuinely contrite. United States v. Royer, 895 F.2d 28, 30 (1st Cir. 1990). For similar sentiments, see also United States v. Spires, 79 F.3d 464, 467 (5th Cir. 1996) and United States v. Zanni, 64 F.3d 654, 654 (1st Cir. 1995).
56. In addition to the deletion of the "merely self-serving" provision discussed supra, the current version of section 3E1.1 has also changed the basic mechanism from a state-of-mind-oriented inquiry, 1986 DRAFT, supra note 26, § B321 (providing adjustment for offender who "recognizes and sincerely accepts responsibility for the offense(s)") (emphasis added), to a more objective inquiry, see U.S.S.G. § 3E1.1(a) (1994) (providing adjustment for defendant who "clearly demonstrates acceptance of responsibility for his offense").
57. U.S.S.G. § 3E1.1 commentary (backg'd).
58. See, e.g., Barry Letter, supra note 14.
59. Amendment 24, 62 Fed. Reg. at 179-80.
60. Id. (proposed application note 1).
61. Although the defendant's admission of guilt must be "timely" under amendment 24, the drafters of the amendment do not envision the timeliness requirement as particularly demanding: "[A] defendant who pleads guilty one day before his scheduled trial date may qualify under subsection (a)." Id. (note to proposed application note 2(b)).
62. Id. (proposed application note 1(b)).
63. Compare U.S.S.G. § 3E1.1(a) with Proposed § 3E1.1(a).
64. Id. (proposed application note 3).
65. Compare U.S.S.G. § 3E1.1(b) with proposed § 3E1.1(b).
66. Proposed application note 2.
67. Id.
68. STAFF DISCUSSION PAPER, supra note 4, at 8 (noting that 11% of all sentencing-related appeal issues involve acceptance of responsibility).
69. My conclusions in this Part about appellate court treatment of section 3E1.1 are drawn primarily from an analysis of three subsets of the enormous body of section 3E1.1 case law: (1) all published section 3E1.1 cases decided between March 1995 and March 1996 (approximately 50 cases), (2) all published Second Circuit cases on section 3E1.1 (approximately 50 cases), and (3) an additional 20 to 30 cases frequently cited in the secondary literature or in other cases.
70. In a particularly insightful opinion, Judge Easterbrook wrote: "Acceptance of responsibility" is in most cases a thinly disguised reduction for pleading guilty, a lure the prosecutor and the court may dangle for saving them the time and risk of trial. Perhaps it would have been simpler had the Guidelines said this. They do not, the reduction is not automatic, and the vague standard creates a possibility of confusion. United States v. Escobar-Mejia, 915 F.2d 1152, 1153 (7th Cir. 1990) (citations omitted).
71. United States v. DeLeon Ruiz, 47 F.3d 452, 455 (1st Cir. 1995).
72. United States v. Cruz, 977 F.2d 732, 734 (2d Cir. 1992) ("The Sentencing Guidelines reflect this [plea] 'discount' approach by affording a defendant a two-level reduction in the otherwise applicable offense level in recognition of the defendant's 'acceptance of responsibility.'").
73. For another opinion sharing this spirit, see United States v. Bradford, 78 F.3d 1216, 1226 (7th Cir. 1996) ("When the United States Attorney's office and the district court expend the vast amount of time and resources necessary to properly prepare for and hold a trial, the defendant will rarely be eligible for an acceptance of responsibility reduction . . . .").
74. 78 F.3d 610 (D.C. Cir. 1996).
75. Id. at 618 (citations omitted). I appreciate the tendency to reach this result under the guideline as presently drafted (i.e., without clear principles) but, as suggested in the previous Part, I believe that there is a plausible reading of section 3E1.1 - the cooperation paradigm - in which remorse is not a central element of the provision. One advantage of the proposal I will make in the final Part of this Article would be to clarify the availability of a departure for remorse. The Ninth Circuit recently also engaged in some semantic hair-splitting in a section 3E1.1 case, but reached a somewhat different conclusion, suggesting that lack of contrition was not a valid basis on which to deny an acceptance-of-responsibility adjustment. See United States v. Vance, 62 F.3d 1152, 1158 (9th Cir. 1995). For further discussion of Vance, see infra section III.C.2.
76. 883 F.2d 1010 (11th Cir. 1989).
77. Id. at 1012 (quoting Brady v. United States, 397 U.S. 742, 753 (1969)). Though interesting as an articulation of the remorse paradigm, the Henry court's analysis here is quite obviously wrong: even without section 3E1.1, remorseful defendants could, in fact, be rewarded at sentencing. First, such defendants could be sentenced at the lower end of the applicable sentencing range. Second, remorse might be used as a basis for departing downward in certain cases. The greater ambivalence of the Dyce opinion, decided seven years later, with respect to the remorse paradigm might be in part a reflection of the decreasing emphasis in section 3E1.1 itself on remorse. See supra subparts II.C. and II.D.
78. See, e.g., United States v. Spires, 79 F.3d 464, 467 (5th Cir. 1996) (referring to the § 3E1.1 inquiry as "the trial court's assessment of a defendant's contrition"); United States v. Eyler, 67 F.3d 1386, 1390-91 (9th Cir. 1995) ("[T]he key inquiry for purposes of section (a) is whether the defendant has demonstrated contrition . . . ."); United States v. Echevarria, 33 F.3d 175, 179 (2d Cir. 1994) (holding that defendant may be denied the adjustment for failing "to acknowledge the wrongfulness of his acts"); United States v. Cousineau, 929 F.2d 64, 69 (2d Cir. 1991) ("[O]ne who is without remorse and fails to acknowledge that his behavior was wrong clearly is not entitled to a reduction for acceptance of responsibility . . . ."); United States v. Royer, 895 F.2d 28, 30 (1st Cir. 1990) ("In deciding whether a defendant is entitled to a reduction . . . . a district court must weigh a multitude of factors . . . . [to] determin[e] whether [the] person is genuinely remorseful."); United States v. Cook, 922 F.2d 1026, 1037 (2d Cir. 1990) (noting that "acceptance of responsibility necessitates candor and remorse" (citing Royer, 895 F.2d 28)). See also HUTCHINSON & YELLEN, supra note 43, at 513 ("What the Commission seems to have intended is that the defendant be sincerely contrite. The few reported decisions are consistent with this analysis.").
79. See, e.g., United States v. Zanni, 64 F.3d 654, 654 (1st Cir. 1995) (upholding a denial of adjustment when district court "was not persuaded that [defendant] was genuinely contrite"); United States v. Boothe, 994 F.2d 63, 71 (2d Cir. 1993) (upholding denial of adjustment when district court had relied on the fact that defendant "failed to prove the sincerity of his acknowledgment of guilt"); United States v. Austin, 948 F.2d 783, 787-88 (1st Cir. 1991) (upholding a denial of adjustment when district court "remained unconvinced the [defendant] had any remorse whatever"); United States v. Whitehead, 912 F.2d 448, 451 (10th Cir. 1990) (upholding denial of adjustment based in part on finding that defendant "did not accept fully that his actions were morally and legally improper"); United States v. Hill, 911 F.2d 129, 131 (8th Cir. 1990) (upholding denial of adjustment based on fact that defendant "did not express sorrow or state that he wished he had not committed the crime").
80. See supra note 53 and cases cited therein.
81. See, e.g., United States v. Spires, 79 F.3d 464, 467 (5th Cir. 1996) ("[T]he standard of review is even more deferential than clear error. . . . [b]ecause the trial court's assessment of a defendant's contrition will depend heavily on credibility assessments . . . .") (citation omitted); United States v. Myers, 66 F.3d 1364, 1372 (4th Cir. 1995); United States v. Luciano-Mosquera, 63 F.3d 1142, 1157 (1st Cir. 1995); United States v. Kerr, 13 F.3d 203, 205 (7th Cir. 1993) ("[T]he trial judge is in a unique position to decide whether the acceptance is sincere . . . ."); United States v. Sanchez-Estrada, 62 F.3d 981, 986 (7th Cir. 1995); United States v. Bonds, 48 F.3d 184, 187 (8th Cir. 1995); United States v. Cousineau, 929 F.2d 64, 69 (2d. Cir. 1991); United States v. Whitehead, 912 F.2d 448, 451 (10th Cir. 1990) ("Because of our deference to the trial court's assessment of credibility . . . .").
82. See, e.g., United States v. Lunsford, 1996 WL 67919, at *2 (6th Cir. Feb. 15, 1996); United States v. Morris, 76 F.3d 171, 175 (7th Cir. 1996); United States v. Myers, 66 F.3d 1364, 1372 (4th Cir. 1995); United States v. Bonds, 48 F.3d 184, 187 (8th Cir. 1995); United States v. Whitehead, 912 F.2d 448, 451 (10th Cir. 1990); see also Jon M. Sands & Cynthia A. Coates, The Mikado's Object: The Tension Between Relevant Conduct and Acceptance of Responsibility in the Federal Sentencing Guidelines, 23 ARIZ. ST. L.J. 61, 78 (1991).
83. United States v. Spires, 79 F.3d 464, 467 (5th Cir. 1996).
84. United States v. Boothe, 994 F.2d 63, 70 (2d Cir. 1993) (citations omitted).
85. Among 47 published section 3E1.1 cases from all circuits decided between early 1995 and early 1996, only five resulted in a remand for a fresh acceptance-of-responsibility determination. Among 53 Second Circuit cases decided between 1988 and 1996, only seven resulted in a remand on the section 3E1.1 issue.
86. Generally, the cases involve an appeal by a defendant who was denied the adjustment, rather than an appeal by the government over an improperly granted adjustment.
87. See, e.g., United States v. Spires, 79 F.3d 464, 467 (5th Cir. 1996) (defendant admitted factual elements of guilt, but went to trial to raise entrapment and duress claims); United States v. Lunsford, 1996 WL 67919, at *2 (6th Cir. Feb. 15, 1996) (defendant received obstruction enhancement); United States v. Luciano-Mosquera, 63 F.3d 1142, 1157 (1st Cir. 1995) (defendant went to trial after government took a hard line in plea negotiations); United States v. DeLeon Ruiz, 47 F.3d 452, 455 (1st Cir. 1995) (defendant went to trial after government took a hard line in plea negotiations).
88. See, e.g., United States v. Vital, 68 F.3d 114, 120-21 (5th Cir. 1995) (granting no adjustment when defendant tested positive for drugs prior to sentencing); United States v. Gordon, 64 F.3d 281, 285 (7th Cir. 1995) (denying adjustment because of bank robbery prior to sentencing). With respect to evidence of post-offense drug use, some - but not all - courts distinguish cases based on whether the offense of conviction was drug-related. If the offense was not drug related, then subsequent drug use may be said to be irrelevant to the question of whether the defendant accepts responsibility for the offense. Compare United States v. Morrison, 983 F.2d 730, 735 (6th Cir. 1993) (holding that subsequent drug use is not relevant to acceptance of responsibility in a firearms violation) with United States v. Watkins, 911 F.2d 983, 985 (5th Cir. 1990) (stating that subsequent drug use may be considered in acceptance-of-responsibility analysis in case of possession of stolen checks).
89. In United States v. Drapeau, the Eighth Circuit upheld a denial of acceptance of responsibility to a sex offender who pled guilty, but who was drunk at the time of the offense, claimed an inability to recall the details of his conduct and stated that he "had difficulty in believing that he had committed the offense. 943 F.2d 27, 28-29 (8th Cir. 1991). And in United States v. Whitehead, a fraud case, the Tenth Circuit upheld - albeit reluctantly - a denial based primarily on the defendant's honest admission to a probation officer that, while he regretted causing harm to individuals, he "felt no remorse over cheating large businesses." 912 F.2d 448, 450-51 (10th Cir. 1990).
90. In United States v. Cousineau, the Second Circuit upheld a denial of acceptance of responsibility based solely on the district court's finding that the defendant "had not shown remorse or acknowledged the wrongfulness of the conduct for which he was convicted," but cited no evidence to support this conclusion. 929 F.2d 64, 69 (2d Cir. 1991). And in United States v. Friedman, the Second Circuit upheld a denial based on a similarly conclusory finding by the district court that the defendant "had refused to 'come forward' and, by his actions, show 'clear acceptance of responsibility.'" 998 F.2d 53, 60 (2d Cir. 1993).
91. Dan Freed & Marc Miller, Editors' Observations: Plea Bargained Sentences, Disparity and "Guideline Justice," 3 FED. SENTENCING REP. 175, 176 (1991) (citations omitted).
92. See WORKING GROUP, supra note 33, at 16.
93. HUTCHINSON & YELLEN, supra note 43, at 515. The Commission characterized this amendment as merely a "clarification," although the new language seemed to carry substantively different connotations. See id.
94. WORKING GROUP, supra note 33, at 17 (citing decisions by the Third, Fourth, Fifth, and Eleventh Circuits).
95. HUTCHINSON & YELLEN, supra note 43, at 516.
96. 873 F.2d 455 (1st Cir. 1989).
97. Id. at 457-58 (quoting transcript of sentencing hearing).
98. Id. at 463. For an argument against the reasoning of the Perez-Franco court, see Bradford C. Mank, Truth in Sentencing: Accepting Responsibility Under the United States Sentencing Guidelines, 25 GONZ. L. REV. 183 (1990). For a discussion of various ways that the 1988 version of section 3E1.1 might have been reconciled with the Fifth Amendment, see Luke T. Dokla, Note, Section 3E1.1 Contrition and Fifth Amendment Incrimination: Is There an Iron Fist Beneath the Sentencing Guidelines' Velvet Glove?, 65 ST. JOHN'S L. REV. 1077 (1991).
99. HUTCHINSON & YELLEN, supra note 43, at 516.
100. U.S.S.G. § 3E1.1 application note 1(a) (emphasis added).
101. Id.
102. 17 F.3d 27 (2d Cir. 1994).
103. See id. at 30.
104. Id. at 31 ("The language of Application Note 1(a) is unconditional.").
105. Id. at 32.
106. Id. at 30.
107. See, e.g., United States v. Eyler, 67 F.3d 1386, 1391-92 (9th Cir. 1995); United States v. Vance, 62 F.3d 1152, 1159-60 (9th Cir. 1995); United States v. Gonzalez, 16 F.3d 985, 990-91 (9th Cir. 1993).
108. 62 F.3d at 1157. Vance refused to discuss the offense with his probation officer on the advice of counsel because of a concern that statements to the probation officer could be used against him if his appeal of the suppression motion proved successful. Id.
109. United States v. LaPierre, 998 F.2d 1460 (9th Cir. 1993).
110. 62 F.3d at 1157.
111. Id. at 1158.
112. Id. at 1158-59 (citations omitted).
113. Id. at 1157 ("[W]e are compelled to find that the district court clearly erred in this case . . . because of the absence of significant evidence that Vance did not accept responsibility.") (citation omitted).
114. Id. at 1160.
115. For other examples of the Ninth Circuit's aggressive approach to section 3E1.1 review, see United States v. Gonzalez, 16 F.3d 985, 991 (9th Cir. 1993) (holding that a defendant's untruthful statement to a probation officer concerning motive for offense cannot overcome presumption in favor of adjustment when defendant has pled guilty, admitted facts of offense, and stated to probation officer that he "feels bad for what he has done") and United States v. Eyler, 67 F.3d 1386, 1391-92 (9th Cir. 1995) (holding that a district court must grant one-point section 3E1.1(b) reduction when a defendant receives subsection (a) reduction and admits involvement in one charged offense at time of arrest, notwithstanding defendant's alleged misrepresentations of his background and decision to go to trial to contest second charge, of which defendant was acquitted).
116. This account is based on interviews conducted in New Haven, Connecticut, in April of 1996, with three senior probation officers, three senior assistant United States attorneys, and an assistant federal public defender. These seven individuals were in virtually complete agreement as to the manner in which section 3E1.1 is implemented in the district. The author's subsequent experience as a law clerk in the district has served to confirm the interviewees' observations. Nothing in this account, however, should be understood as a description of the practices or opinions of Judge Arterton specifically. The author is grateful to the seven interviewees for their thoughtful participation. The interviewees will be identified herein by organization but not by name.
117. The defender elaborated by discussing an instance in which one of his clients had, in the hope of getting the reduction, provided an exceptionally complete statement of his involvement in a series of illegal gun sales. However, the judge used the defendant's admission that the sales had largely been to gang members as a basis for upward departure, reasoning that gun sales to gang members were far more dangerous than the "heartland" cases of such transactions. "After you get burned like that," the defender concluded, "you learn to be very defensive."
118. This observation, of course, primarily applies to defendants who plead guilty (the vast majority); defendants who go to trial rarely make a serious bid for acceptance of responsibility, in part to preserve an uncompromised position through the appeals process and in part because such defendants have adopted a mentality of adversity that is incompatible with a request for reduction. It is to be observed, however, that only approximately 2% of cases go to trial in Connecticut, meaning that such situations are rare. U.S. SENTENCING COMM'N 1995 ANN. REP. 53 [hereinafter ANNUAL REPORT].
119. In order to underscore the importance of these functions, a public defender observed that the attorneys in his office "make a large investment" of time and resources in presentence interviews. Attending the interviews often involves lengthy car trips in addition to the time consumed by the interview itself, which is usually at least two hours. Moreover, defenders may spend several hours preparing defendants for the interviews.
120. Indeed, probation officers claim that they try to protect the interests of defendants whose lawyers do not "know how to play the game." For instance, if a defense lawyer fails to provide the standard written stipulation of culpability, the officer may refer the lawyer to the public defender's office for advice on how to represent his client properly during the sentencing process. Notwithstanding such assertions, however, the federal public defender I spoke with remains concerned that defendants without good lawyers may be disadvantaged in the acceptance-of-responsibility process by not having safeguards against freewheeling conversations with probation officers.
121. One probation officer noted, however, that some of her colleagues are less generous than others with recommending the adjustment. She believes that this may be because the less generous officers require some evidence of genuine remorse before recommending it. She tends to think that this is an unrealistic expectation, and believes that it may diminish with experience. In some cases, supervisors can change recommendations in presentence reports which they believe to be unfair, but acceptance of responsibility is considered to be enough of a close judgment call that recommendations are rarely altered.
122. Indeed, if a defense lawyer is unable to attend a presentence interview, probation officers say that they will often abide by requests from defense lawyers to avoid certain topics in the interview.
123. One probation officer noted an exception to this general tendency: if an offense involves drugs, guns, or money, and if the police have not recovered all of its fruits or instrumentalities, some judges are reluctant to give the reduction if the defendant has not assisted with recovery. The officer cited an instance in which one judge expressed displeasure with the probation office for recommending a reduction for an embezzlement in which the defendant had not helped the police recover all of the stolen money. The judge's statement had an effect on probation office practices, but, of course, the judge's approach to section 3E1.1 cannot be pushed very far in the wake of Austin, which was a Second Circuit case that, in fact, originated in Connecticut.
124. A defender asked rhetorically, "What is a probation officer going to do if the entire federal bar - prosecutors and defenders - supports the practice?"
125. If a defendant goes to trial, the probation office rarely recommends the reduction. According to a probation officer, this is due to the language of application note 2 counsel's desire to protect the defendant's position in appeals precludes even a minimal statement of culpability.
126. Observing the shared approach to section 3E1.1 by prosecutors, defenders, and probation officers, one prosecutor estimated that section 3E1.1 "is probably just about the least litigated Guidelines issue in this district."
127. A copy of this standard form has been provided to the author. Recommendation of the adjustment is conditioned on truthful disclosures to the probation office and may be withdrawn if the defendants maintain their criminal conduct or associations, obstruct justice, or violate a condition of their release. Assistant United States Attorneys state that they do not have discretion to deviate from this standard agreement. By contrast, outside of the plea context, prosecutors "have a hard time imagining" a situation in which acceptance of responsibility should be recommended. An example of such a case, according to one interviewee, is if the defendant goes to trial only to raise an entrapment defense.
128. One prosecutor noted that in a handful of guilty plea cases each year, prosecutors will change their position and oppose acceptance of responsibility as a result of post-offense conduct (e.g., failed drug tests), but that the court usually awards acceptance of responsibility anyway. A defender suggested that, more generally, the prosecutor's recommendation carries little weight with the court. He believes that this tendency reflects the continuing force of pre-Guidelines practices in Connecticut: traditionally, judges in the district "have not wanted to hear from prosecutors when it comes to sentencing" because "judges believe they can get everything they want to know from the probation office."
129. Indeed, one prosecutor suggested that, in his experience, recidivism rates may be higher among defendants pleading guilty than among defendants going to trial.
130. For instance, a probation officer recalled one instance in which a judge sentenced a defendant at the top of the recommended sentencing range due to lack of remorse, which may have resulted in an extra five years of incarceration.
131. A defender observed, "It is no secret in our district that judges think the Guidelines are too harsh. They are very results-oriented and will manipulate the Guidelines to obtain a result they believe is just."
132. A probation officer cited an instance in which one judge wished to deny the adjustment to a defendant who failed to assist police in recovering money he had embezzled. The judge ultimately awarded the adjustment, however, because everyone else in the room - prosecution, defense, and probation officer - supported it.
133. One prosecutor recalled an instance in which a new judge attempted to elicit mitigating factors from a defendant at a sentencing hearing in the hope of justifying a departure, but instead uncovered a great deal of relevant conduct. The judge was able to cover these admissions up only with much effort and embarrassment. Perhaps as a result of such experiences, most judges pose few, if any, questions to defendants at sentencing hearings.
134. STAFF DISCUSSION PAPER, supra note 4, app. 2 (citing 1994 data).
135. WORKING GROUP, supra note 33, at 5.
136. United States v. Knight, 905 F.2d 189, 192 (8th Cir. 1990). The Knight case also provides an interesting piece of anecdotal evidence for the commonness of equating acceptance of responsibility with pleading guilty: the trial court evidently caught the defense counsel completely off-guard by asking for evidence of acceptance of responsibility. Believing that the guilty plea would be sufficient, the lawyer was not prepared to answer the question. Id. at 190.
137. WORKING GROUP, supra note 33, at 7-8, 11.
138. Id. at 8. In ten of the cases, defendants maintained an outright denial of guilt, despite pleading guilty. Id.
139. Id.
140. Id. at 11. Unfortunately, the working group's study does not indicate what percentage of these defendants went to trial to raise purely legal issues, what percentage raised affirmative defenses (e.g, entrapment, duress, and insanity), and what percentage fully contested the factual bases of the charges against them. Application note 2 suggests that defendants in the third category, and perhaps defendants in the second as well, should not be permitted to receive the reduction.
141. One may hypothesize that these cases - and perhaps many others in which acceptance of responsibility is awarded notwithstanding a trial - represent efforts by judges to subvert Guidelines sentences that were perceived as overly harsh. Alternatively, these cases may be a result of carelessness or of unusual interpretations of section 3E1.1. Unfortunately, the records of the working group do not provide any basis for choosing among these explanations.
142. Id. at 6. Interestingly, all five of these districts that are the least generous with section 3E1.1 are located in the South: Western Texas, Western Arkansas, Northern Oklahoma, Northern Georgia, and Southern Alabama. Id.
143. Id. at 6-7 (Maryland, Middle North Carolina, Eastern Louisiana, and Southern Ohio).
144. Id. at 7. The working group's conclusion might be more persuasive if it provided more data. For instance, the working group noted that five districts awarded the section 3E1.1 adjustment to fewer than 75% of defendants who pled guilty, id. at 6, but did not indicate either how far below 75% these districts were, or how many districts were at the opposite extreme, say, above 90%.
145. Indeed, based on an extensive series of interviews conducted in twelve different judicial districts, the working group itself observed, "[S]everal statements suggest that acceptance of responsibility is applied differently across the nation and within a given district." Id. at 24.
146. Gerald W. Heaney, Revisiting Disparity: Debating Guidelines Sentencing, 29 AM. CRIM. L. REV. 771, 775 (1992). The percentage of defendants going to trial who received the benefit varied from 2% to 33%, with an average of 16%. Id. These numbers, of course, need not necessarily represent unwarranted disparity. It may be that, due to the nature of crime prosecuted or other factors differentiating the districts, there are important differences among the populations of defendants being sentenced. For instance, the defendants in one district may actually be more remorseful than those in the others. Unfortunately, strictly numerical comparisons, such as those offered by Heaney, do not make this sort of analysis possible.
147. See Ilene H. Nagel & Stephen J. Schulhofer, A Tale of Three Cities: An Empirical Study of Charging and Bargaining Practices Under the Federal Sentencing Guidelines, 66 S. CAL. L. REV. 501 (1992).
148. Id. at 531, 550.
149. Id. at 540. The Nagel and Schulhofer study does not provide exact numbers because the authors wished to preserve the anonymity of the districts they studied. Id. at 553 n.73.
150. ANNUAL REPORT, supra note 116, at 103.
151. Letter from Susan Katzenelson, Director, U.S. Sentencing Commission Office of Policy Analysis, to Maria Rodrigues McBride, Chief Probation Officer of the District of Connecticut (Feb. 20, 1997) (copy on file with author).
152. 13 F.3d 555 (2d Cir. 1994).
153. Id. at 557. The appellate court upheld the denial of a reduction on these grounds.
154. 880 F. Supp. 605 (N.D. Ind. 1995).
155. Id. at 608-09. The court remarked dryly, "Mr. Schultz performed poorly with respect to the conditions of his pretrial supervision." Id. at 608. In deciding to grant the section 3E1.1 adjustment despite Schultz's poor performance, the court gave weight to the following factors: Schultz's plea was timely (two weeks before trial), he admitted the conduct comprising the offense of conviction, and he acknowledged his alcoholism at the sentencing hearing. None of these factors seems particularly extraordinary, suggesting a strong inclination on the part of the sentencing judge to reward guilty pleas, regardless of other aspects of the post-offense conduct.
156. Barry Letter, supra note 14, at 1.
157. Id. ("This confusion is generated by the interaction and definition of the numerous factors listed in the current acceptance guideline, all of which comprise the vague concept of 'acceptance.'").
158. Both prosecutors and defendants appeal section 3E1.1 decisions at a relatively high rate: it is the third most litigated issue among both groups. Id. at 1 n.1.
159. U.S.S.G. § 1A3.
160. Freed & Miller, supra note 89, at 176. Some interviewees feel that a review of the acceptance guidelines should start by going back to the original philosophy that serves as a foundation for the guideline. Training the operation and rationale of the acceptance guidelines frequently generates the comment, 'Well, which is it, a reduction for the defendant who accepts responsibility for his conduct, or a guilty plea discount?' STAFF DISCUSSION PAPER, supra note 4, at 12.
161. By the term "unusual cases," I mean to encompass cases that go to trial, in addition to even more exceptional cases, such as those involving egregious presentencing behavior like the Schultz case discussed in the previous Part. Though trials may at first blush appear unexceptional, Commission data indicates that nearly 92% of federal criminal convictions result from guilty pleas. ANNUAL REPORT, supra note 116, at 53. Moreover, notwithstanding the strong language of application note 2 against giving a reduction to a defendant who goes to trial, courts seem to vary significantly in their willingness to find acceptance of responsibility in trial-conviction cases. See supra text accompanying notes 146-47. Thus, going to trial seems just such an exceptional circumstance as is capable of triggering disparate applications of section 3E1.1.
162. The district judge in Knight, for instance, apparently insisted that defendants must do one of the acts listed in application note 1 in order to qualify for the adjustment. United States v. Knight, 905 F.2d 189 (8th Cir. 1990).
163. See U.S.S.G § 1A3 ("[Granting broad discretion in the Guidelines] risked a return to the wide disparity that Congress established the Commission to reduce and would have been contrary to the Commission's mandate set forth in the Sentencing Reform Act of 1984.").
164. Brady v. United States, 397 U.S. 742, 753 (1969).
165. Notwithstanding the assertion of the Brady court that guilty pleas are a good indicator of rehabilitative potential, a plea is only a proxy for such potential, and an inquiry that focuses solely on conduct, such as pleading guilty, will necessarily produce somewhat different results than in an inquiry into rehabilitative potential per se. Indeed, focusing on some of the conduct that is emphasized in the application notes may actually undermine the ability of section 3E1.1 to reward rehabilitative potential. For instance, one of the prosecutors I interviewed argued that defendants who plead guilty actually have a higher recidivism rate than those who go to trial.
166. For instance, a plausible reading of the application notes might be as follows: the Commission so wishes to discourage going to trial and obstructing justice that when a defendant engages in either conduct the defendant is appropriately precluded from a reduction based on state of mind. In other cases, however, section 3E1.1 is available and functions to provide a benefit for defendants with rehabilitative potential. Put differently, in cases of trial or obstruction, the cooperation paradigm predominates; otherwise, the remorse paradigm predominates. Such a reading might or might not constitute wise policy, but it would at least be a policy and represent a coherent harmonization of the competing goals of section 3E1.1.
167. See Barry Letter, supra note 14, at 1-2. The Judicial Conference argued: The current guidelines try to do too much with one adjustment . . . and consequently it does not serve any of its numerous goals well. Any plea incentive is inextricably intertwined with issues of attitude, other conduct, the government's preparation, etc. which either get lost in the overwhelming thrust to reward a plea, or worse, prevent the plea incentive from working. How does a court reward entry of a plea where the defendant has done something (perhaps submitted a bad urine sample) which arguably prevents the allowance of the first 2-level adjustment? There is no way to provide a reward for the many non-plea incentives listed in § 3E1.1 if the court rewards the entry of the plea; the other incentives either overcome the plea incentive, or vice versa. In a case where the defendant agrees simply to enter a timely plea, and if the court rewards the plea, it must do so with the full three points . . . with no adjustment left to act as an incentive for the other commendable conduct which the guideline attempts to encourage. Id.
168. This concern has been echoed in studies of prohibitions on plea bargaining in state courts: without the clear, predictable benefits associated with negotiated pleas, defendants are more willing to take their chances with a trial. See Teresa White Cams & John A. Kruse, Alaska's Ban on Plea Bargaining Reevaluated, 75 JUDICATURE 310, 311 (1992); Robert A. Weninger, The Abolition of Plea Bargaining: A Case Study of El Paso County, Texas, 35 UCLA L. REV. 265, 311 (1987).
169. This concern has been echoed in studies of prohibitions on plea bargaining in state courts: without the clear, predictable benefits associated with negotiated pleas, defendants are more willing to take their chances with a trial. See Teresa White Cams & John A. Kruse, Alaska's Ban on Plea Bargaining Reevaluated, 75 JUDICATURE 310, 311 (1992); Robert A. Weninger, The Abolition of Plea Bargaining: A Case Study of El Paso County, Texas, 35 UCLA L. REV. 265, 311 (1987).
170. Schulhofer & Nagel, supra note 34, at 267. Schulhofer and Nagel's subsequent survey of plea bargaining in three districts found continuing evidence of this concern. According to [assistant U.S. Attorneys], this practice [of awarding a discount to defendants who go to trial] prompts the belief that the acceptance-of-responsibility discount with a sentence at the bottom of the range does not provide an adequate incentive for defendants to plead guilty because defendants have a good chance of receiving the same benefit even after proceeding to trial. Nagel & Schulhofer, supra note 145, at 550.
171. Stephen J. Schulhofer, Implementing the Plea Agreement Provisions of the Federal Sentencing Guidelines, 3 FED. SENTENCING REP. 179 (1991); see also Freed & Miller, supra note 89, at 176 (commenting favorably on Schulhofer's analysis).
172. See Robert H. Edmunds, Jr., The Need for Predictable, but Not Mandatory, Sentences, 8 FED. SENTENCING REP. 16, 17 (1995) ("Predictability may be a difficult goal to reach, but reliable knowledge of the likely course of a case would be of immense value to attorneys and defendants.").
173. On the other hand, the defenders of section 3E1.1 have observed that the rate of guilty pleas has not appreciably changed since introduction of the Guidelines. WORKING GROUP, supra note 33, at 4. In response, the critics of section 3E1.1 might argue that the rate of guilty pleas has only been maintained because most districts have followed the Connecticut pattern and effectively transformed section 3E1.1 into an automatic plea discount, and because the actors in the criminal justice system have circumvented the Guidelines and found mechanisms other than section 3E1.1 to reward guilty pleas, such as written stipulations that minimize the role in offense or relevant conduct). Cf. Schulhofer & Nagel, supra note 34, at 268 ("[W]e cannot be certain whether the relative stability of the guilty plea rate indicates that the 'acceptance' discount is set at an appropriate level, or whether the system has found alternative ways to grant guilty plea inducements."). While overall national rates of plea bargaining may be appropriate, districts that do not hew to the Connecticut model or circumvent the Guidelines may have an inadequate plea inducement. Critics might also reject the use of pre-Guidelines practices as a benchmark: perhaps pre-Guidelines plea discounts were also insufficiently reliable and effective. For an argument that section 3E1.1 should not be changed to increase the frequency of plea agreements, see Andrew J. Kleinfeld, The Sentencing Guidelines Promote Truth and Justice, FED. PROBATION, Dec. 1991, at 16, 18 ("The value to society . . . of a guilty plea is much less than its value to actors within the justice system.").
174. See, e.g., United States v. Brownstein, 79 F.3d 121, 123 (9th Cir. 1996) (refusing to depart for self-surrender to authorities because self-surrender adequately considered in section 3E1.1); United States v. Dyce, 91 F.3d 1462, 1469 (D.C. Cir. 1996) (no departure for remorse); United States v. Broderson, 67 F.3d 452 (2d Cir. 1995) (restitution not a basis for departure); United States v. Arjoon, 964 F.2d 167, 171 (2d Cir. 1992) (return of stolen property not a basis for departure). But see United States v. Maier, 975 F.2d 944, 948 (2d. Cir. 1992) (notwithstanding section 3E1.1, drug rehabilitation may furnish a ground for departure); United States v. Rogers, 972 F.2d 489, 493 (2d Cir. 1992) (surrender to police and confession one day after robbery may furnish a basis for departure); United States v. Garcia, 926 F.2d 125, 126 (2d Cir. 1991) ("activities facilitating proper administration of justice in the District Courts" may furnish grounds for departure). See generally HUTCHINSON & YELLEN, supra note 43, at 517 (discussing departures for "superacceptance of responsibility" in Second, Third, and Seventh Circuits); WORKING GROUP, supra note 33, at 14 (discussing a sample of 21 case files involving departures for acceptance of responsibility).
175. U.S.S.G. § 5K2.0 (quoting 18 U.S.C. § 3553(b)).
176. The judge may, of course, take the defendant's remorse into account when selecting a precise sentence from the range available to him under the Guidelines. This, however, may be an unsatisfactory mechanism for rewarding remorse, both because the recommended range may be quite narrow and because other factors may otherwise dictate a sentence at or near the bottom of the range anyway. For instance, judges who automatically sentence at the bottom of the range out of a belief that most Guidelines sentences are too harsh - the author has met at least one judge who admits to this practice - have little practical ability to differentiate between remorseful offenders through a selection of a sentence from a range. Judges may reward remorseful defendants through manipulation of other aspects of the Sentencing Guidelines, such as through generosity on role in offense or relevant conduct, but some judges may be disinclined to use the Guidelines dishonestly, and, in any case, it seems unwise to build policy decisions around assumptions that judges will circumvent or subvert the black-letter law.
177. I do not mean to suggest that these are the only forms of undesirable discrimination that may result from section 3E1.1, but they may have the greatest social salience. Subpart V.F below suggests that section 3E1.1 may also discriminate against naive or inexperienced criminal defendants. Subpart V.G argues that section 3E1.1 may discriminate against defendants who exercise constitutional rights.
178. The defender particularly worried about the ability of private practitioners to represent their clients well during sentencing. He believed that the federal public defenders, who spend a great deal of time on sentencing issues, generally did a good job of protecting the interests of their clients on matters like section 3E1.1, but feared that private practitioners, who rarely work on federal criminal matters full time, were not capable of remaining abreast of developments in sentencing law and practice, especially now in the era of the Guidelines, which have rendered sentencing as complex and technical a matter as the tax code. Notwithstanding these concerns, probation officers in Connecticut maintained that they could and would adequately protect the interests of defendants represented by unskilled lawyers.
179. See United States v. Vance, 62 F.3d 1152, 1158 (9th Cir. 1995) (equating "penetrating examination of the criminal's soul" with "lenience toward those who cry more easily, or who have sufficient criminal experience to display sentiment at sentencing"). As one Connecticut probation officer observed, knowing how to present oneself to the court in an appropriate manner is a learned skill, and not all defendants possess the skill to an equal degree.
180. By this term, I mean to encompass both people of low intelligence and people with psychiatric disorders.
181. Indeed, even if a judge recognized that a defendant's failure to accept responsibility was due to a mental condition, rather than simple unwillingness, there is no apparent basis in section 3E1.1 for the judge to grant the adjustment: in order to qualify for the reduction, a defendant must "clearly demonstrate" acceptance of responsibility, regardless of capacity to do so.
182. United States v. Echevarria, 33 F.3d 175, 178 n.2. (2d Cir. 1994).
183. 901 F.2d 1161 (2d Cir. 1990).
184. Id. at 1164. The appellate court in Altman, unlike the appellate court in Echevarria, remanded the case for reconsideration of the acceptance-of-responsibility issue in light of the defendant's mental illness. Altman, 901 F.2d at 1166. The Vance case may also involve a form of mental incapacity, perhaps related to substance abuse. When initially arrested, Vance was described as "dazed and glassy eyed, and seemed to have difficulty understanding and responding to . . . questions." 62 F.3d at 1155. Portions of Vance's statement at his plea allocution, reprinted in the appellate decision, suggest a continued lack of mental acuity. See id. at 1158. United States v. Schultz, 880 F. Supp. 605 (N.D. Ind. 1995) more clearly implicates a disability related to substance abuse: alcoholism. Schultz's probation officer recommended that the reduction be denied based on Schultz's alcohol-related conduct during his pretrial release. See id. at 608-09. The court, attempting to distinguish Schultz's conduct from his disability, id. at 609, ultimately rejected the officer's recommendation and granted the reduction. However, the case raises the question of how often violations of the conditions of pretrial release - a common basis for denying the adjustment - are due to disability, rather than to simple poor judgment.
185. Michael J. Perlin & Keri K. Gould, Rashomon and the Criminal Law: Mental Disability and the Federal Sentencing Guidelines, 22 AM. J. CRIM. L. 431, 449 (1995) (citing United States v. Spedalieri, 910 F.2d 707, 711-12 (10th Cir. 1990)).
186. Id. at 452; see also James M. Doyle, The Lawyers' Art: "Representation" in Capital Cases, 8 YALE J.L. & HUMAN. 417, 444 (1996) ("Research resoundingly proves that there is a yawning gap between what mental health professionals will diagnose as mental disability and what jurors (or judges) will accept as a mitigating condition.").
187. GEORGE E. DIX & M. MICHAEL SHARLOT, CRIMINAL LAW: CASES AND MATERIALS 790-92 (3d ed. 1987).
188. United States v. Reno, 992 F.2d 739 (7th Cir. 1993).
189. Id. at 744.
190. For some statistical evidence of this effect, see supra text accompanying notes 148-49.
191. 62 F.3d at 1158. As one probation officer observed, "You are just not going to get remorse from some people."
192. Doyle, supra note 183, at 430-31 (citing JAMES C. SCOTT, WEAPONS OF THE WEAK: EVERYDAY FORMS OF PEASANT RESISTANCE (1985)).
193. Id. at 438 (citation omitted).
194. Who is to say that the mistrust, concealment, and hostility displayed by members of subordinated groups to figures of authority is inappropriate? One Connecticut probation officer expressed opposition to treating section 3E1.1 as an inquiry into remorse because many defendants have good reason to be hostile towards the criminal justice system. Yet, such behaviors may make a real difference to actors in the system. For instance, another probation officer maintained that the "tough" attitudes of some defendants may affect declinations and other prosecutorial decisions: "whether the prosecutor likes you, can matter a lot."
195. Doyle, supra note 183, at 436.
196. See id. at 437-38.
197. U.S.S.G. § 1A3 (policy statement).
198. It does seem, however, that the current Sixth Amendment doctrine is amenable to a distinction between a trial penalty and a plea benefit. See infra subpart V.G.
199. See Wilkins, supra note 16, at 191.
200. As a defender put it, "What are you going to do - hook up a lie detector to defendants when they say they are sorry?"
201. United States v. Hendrix, 505 F.2d 1233, 1236 (1974).
202. In a dissenting opinion objecting to the use of a defendant's perjury in sentencing, Justice Stewart pointed out the inadequacy of brief interactions in the courtroom as evidence of a defendant's character: "Indeed, without doubting the sincerity of trial judges, one may doubt whether the single incident of a defendant's trial testimony could ever alter the assessment of rehabilitative prospects so drastically as to justify a perceptibly greater sentence." United States v. Grayson, 438 U.S. 41, 56 n.3 (1977) (Stewart, J., dissenting). Ironically, section 3E1.1 application notes most discourage judges from assessing contrition in precisely the context in which judges are probably in the best position to know something about the defendant's character: sentencing after a trial, in which the judge has had an opportunity to witness the defendant's demeanor - and perhaps testimony - over a great deal of time.
203. In Connecticut, officers typically meet for two to two and a half hours with defendants. Officers are expected to follow up this initial meeting with a visit to the defendant in the defendant's home, but this second meeting often fails to materialize due to scheduling difficulties and resource constraints.
204. Apparently, much of the probation officer's success depends on whether the officer is relaxed in interviews. As one officer stated, "A relaxed probation officer will help the defendant to relax."
205. A Connecticut defender observed, "Most people do not act like themselves in the court-room." As an example, he recalled one particular client who could not stop himself from grinning in the courtroom. The grinning was simply an expression of nervousness, but the defender believes a judge could have easily interpreted the defendant's demeanor as disrespectful and insincere.
206. See United States v. Vance, 62 F.3d 1152, 1158 (9th Cir. 1995) (discussing defendants "who have sufficient criminal experience to display sentiment at sentencing instead of restraining their emotions").
207. 912 F.2d 448, 450 (10th Cir. 1990).
208. See, e.g, United States v. Cordell, 924 F.2d 614, 619-20 (6th Cir. 1991) (no violation of Fifth Amendment privilege against self-incrimination); United States v. Skillman, 922 F.2d 1370, 1378 (9th Cir. 1990) (self-incrimination); United States v. Rogers, 921 F.2d 975, 982 (10th Cir. 1990) (self-incrimination); United States v. Gonzalez, 897 F.2d 1018, 1021 (9th Cir. 1990) (no violation of Sixth Amendment right to trial); United States v. Wivell, 893 F.2d 156, 159 (8th Cir. 1990) (section 3E1.1 not unconstitutionally vague); United States v. Henry, 883 F.2d 1010, 1011 (11th Cir. 1989) (right to trial); United States v. White, 869 F.2d 822, 826 (5th Cir. 1989) (right to trial).
209. See, e.g., United States v. Piper, 918 F.2d 839, 840 (9th Cir. 1990) (self-incrimination); United States v. Oliveras, 905 F.2d 623, 626 (2d Cir. 1990) (self-incrimination); United States v. Perez-Franco, 873 F.2d 455, 461-64 (1st Cir. 1989) (self-incrimination); Dokla, supra note 96, at 1094-97 (self-incrimination); Andrew Neal Siegel, Note, The Sixth Amendment on Ice-United States v. Jones: Whether Sentence Enhancements for Failure to Plead Guilty Chill the Exercise of the Right to Trial, 43 AM. U. L. REV. 645 (1994).
210. Bearing in mind that the Austin court could not effectively reach the Fifth Amendment issue without torturing the language and intent of application note 1, the problem of appellate protection of the Fifth Amendment privilege seems even more difficult. See supra section III.C.1. Heightening the ineffectiveness of appellate policing of the privilege, some appellate decisions hold that a denial of adjustment may be affirmed even if it violates the Fifth Amendment if another basis for the denial exists. United States v. Cousineau, 929 F.2d 64, 69 (2d Cir. 1991); United States v. Ramirez, 910 F.2d 1069, 1071 (2d Cir. 1990).
211. See Dokla, supra note 96, at 662.
212. See, e.g., Bordenkircher v. Hayes, 434 U.S. 357, 363 (1978).
213. See, e.g., United States v. Corbitt, 439 U.S. 212, 222 (1978).
214. In Corbitt, for instance, the Supreme Court upheld a New Jersey statute that permitted judges to reward defendants who did not contest a charge of first-degree murder with a lower sentence than the otherwise mandatory life imprisonment. Id. at 226.
215. Dokla, supra note 90, at 662. This seems a rather esoteric distinction in some cases. For an argument that attempting "to discern a meaningful difference" between a penalty and a denial of a benefit "can only induce vertigo," see United States v. Jones, 997 F.2d 1475, 1483 (D.C. Cir. 1993) (Mikva, C.J., dissenting).
216. The structure of section 3E1.1 seems difficult to distinguish from the mechanism upheld in Corbitt, which also provided a benefit for guilty pleas. See Dokla, supra note 96, at 660. Another similarity between the mechanisms that may be of constitutional significance is that both provide a sentence reduction at the sentencing judge's discretion, rather than an automatic adjustment for not going to trial. The Commission may have viewed this discretionary feature as an important characteristic for a plea inducement to pass constitutional muster. See Wilkins, supra note 16, at 191 n.65. For an argument that the plea inducement may be made automatic without risking rejection on constitutional grounds, see Ellen M. Bryant, Comment, Section 3E1.1 of the Federal Sentencing Guidelines: Bargaining with the Guilty, 44 CATH. U. L. REV. 1269, 1303-05 (1995).
217. Indeed, the distinction between benefit and penalty may be less secure in legal doctrine than its supporters appreciate. The Perez-Franco decision, discussed supra, section III.C.1, plainly rejected the distinction in the context of the Fifth Amendment, referring to "the penalty of not receiving a reduction in [ ] offense level." United States v. Perez-Franco, 873 F.2d 455, 463 (1st Cir. 1989). If such a sentiment were incorporated into the Sixth Amendment analysis of section 3E1.1, the provision - at least as it is commonly applied today - might be in some jeopardy.
218. The case of United States v. Jones illustrates an additional manner in which section 3E1.1 may threaten Sixth Amendment rights, though in an oblique and unusual fashion. In Jones, the defendant chose to go to trial, but the sentencing judge awarded the section 3E1.1 reduction, believing the loss of the two-point reduction would represent too great a penalty (30 months) on Jones' failure to plea bargain successfully with the government. 997 F.2d at 1476. However, in order to preserve the incentive structure for plea bargaining, the judge sentenced Jones six months higher within the recommended range expressly to penalize the decision to go to trial. Id. at 1477. On appeal, the case presented substantial difficulties for the D.C. Circuit, producing an initial panel opinion upholding the sentence, an en banc opinion upholding the sentence on different grounds, and three dissents to the en banc opinion. The en banc majority characterized the trial penalty imposed below as, in effect, a reduction of the section 3E1.1 benefit. Id. at 1478. The holding thus opens a loophole in the traditional rule that a penalty may not be imposed for going to trial. Under Jones, a judge may expressly penalize the decision to go to trial when choosing a sentence within a range so long as the defendant receives the adjustment and the penalty does not wholly erase the benefit of accepting responsibility. Judge Mikva's dissent in Jones raises concerns about the constitutionality of the majority's holding. Regardless of the validity of his dissent, however, I have seen no evidence that the Jones scenario has been replayed elsewhere. In other words, even to the extent that Jones impermissibly burdens the right to trial, the case may not constitute a compelling basis for restructuring section 3E1.1. For a lengthy discussion of the Jones case, see Siegal, supra note 206.
219. The Judicial Conference believes that even an automatic plea discount would be upheld by the courts because the "benefit which a plea brings to the system" outweighs the constitutional harms of burdening the right to trial. Barry Letter, supra note 14, at 2-3.
220. For a discussion of the qualities of courtroom proceedings that lend themselves to powerful political statements, and an argument that the abolitionist John Brown's courtroom performance exerted a profound influence over the development of Northern opinion prior to the Civil War, see Robert Ferguson, Story and Transcription, 6 YALE J.L. & HUMAN. 37, 42-45, 55-58 (1994). Notwithstanding famous counterexamples, such as the trials of John Brown and the Chicago Seven, courtroom proceedings and the operation of the criminal justice system generally function to reinforce the legitimacy of prevailing hierarchies and belief structures. See Doyle, supra note 183, at 438-42 (describing constraints imposed on criminal defense attorneys by the need to respect racial and other prejudices of jurors);
221. For a discussion of the qualities of courtroom proceedings that lend themselves to powerful political statements, and an argument that the abolitionist John Brown's courtroom performance exerted a profound influence over the development of Northern opinion prior to the Civil War, see Robert Ferguson, Story and Transcription, 6 YALE J.L. & HUMAN. 37, 42-45, 55-58 (1994). Notwithstanding famous counterexamples, such as the trials of John Brown and the Chicago Seven, courtroom proceedings and the operation of the criminal justice system generally function to reinforce the legitimacy of prevailing hierarchies and belief structures. See Doyle, supra note 183, at 438-42 (describing constraints imposed on criminal defense attorneys by the need to respect racial and other prejudices of jurors); Cara W. Robertson, Representing "Miss Lizzie": Cultural Convictions in the Trial of Lizzie Borden, 8 YALE J.L. & HUMAN. 351, 392 (1996) (discussing trial strategies of Lizzie Borden case as reflecting prevailing gender and class ideologies and reinforcing divisions between elite and working-class elements of Fall River, Massachusetts society); Louis Michael Seidman, Rubashov's Question: Self-Incrimination and the Problem of Coerced Preferences, 2 YALE J.L. & HUMAN. 149, 151 (discussing the focus of criminal justice system on obtaining confessions as an effort to reinforce legitimacy of the system). One might argue, however, that the criminal justice system actually weakens, rather than strengthens, its claims to legitimacy by squelching the authentic voices of defendants.
222. For a discussion of the qualities of courtroom proceedings that lend themselves to powerful political statements, and an argument that the abolitionist John Brown's courtroom performance exerted a profound influence over the development of Northern opinion prior to the Civil War, see Robert Ferguson, Story and Transcription, 6 YALE J.L. & HUMAN. 37, 42-45, 55-58 (1994). Notwithstanding famous counterexamples, such as the trials of John Brown and the Chicago Seven, courtroom proceedings and the operation of the criminal justice system generally function to reinforce the legitimacy of prevailing hierarchies and belief structures. See Doyle, supra note 183, at 438-42 (describing constraints imposed on criminal defense attorneys by the need to respect racial and other prejudices of jurors); Cara W. Robertson, Representing "Miss Lizzie": Cultural Convictions in the Trial of Lizzie Borden, 8 YALE J.L. & HUMAN. 351, 392 (1996) (discussing trial strategies of Lizzie Borden case as reflecting prevailing gender and class ideologies and reinforcing divisions between elite and working-class elements of Fall River, Massachusetts society); Louis Michael Seidman, Rubashov's Question: Self-Incrimination and the Problem of Coerced Preferences, 2 YALE J.L. & HUMAN. 149, 151 (discussing the focus of criminal justice system on obtaining confessions as an effort to reinforce legitimacy of the system). One might argue, however, that the criminal justice system actually weakens, rather than strengthens, its claims to legitimacy by squelching the authentic voices of defendants.
223. A detailed examination of the technical constitutional questions involved is beyond the scope of this Article, but the consistent upholding of section 3E1.1, see cases cited supra in note 205, suggests that courts would generally be unreceptive to fresh constitutional challenges.
224. Examples might include various types of conduct currently enumerated in the application notes, such as voluntary restitution, turning oneself in to the authorities, voluntary assistance to authorities in recovering fruits and instrumentalities of the offense, and obstructing justice. The new application notes might further suggest presumptive point values for each of these acts. For instance, a defendant who pleads guilty might have a presumptive two-level benefit reduced by one level for obstructing justice, but increased by two levels for a timely plea and voluntary restitution, resulting in an overall reduction of three points. As my proposal is structured, desired and discouraged acts would be treated cumulatively, but the net benefit would be limited to three levels.
225. See supra subpart II.C. In a similar vein, the Judicial Conference of the United States has also observed that the Commission has progressively moved closer to a direct reward for a guilty plea. Barry Letter, supra note 14, at 2.
226. See supra subpart II.D.
227. One of the U.S. Judicial Conference's objections to the current structure of section 3E.1 is that judges may not award a one-point reduction at all, and generally have little choice about awarding the third point when they award the section (a) adjustment to a defendant whose offense level is greater than sixteen; in short, judges have little ability to calibrate the § 3E1.1 adjustment to the particular circumstances of the defendant. See Barry Letter, supra note 14, at 3-4 (discussing the "all-or-nothing" problem). Under my proposal, judges would have greater flexibility as to the amount of reduction, but would have clearer guidance from the Commission and would be more accountable to appellate courts. In short, my proposal is not intended to increase judicial discretion per se. Note that permitting judges to award a one-point reduction under section 3E1.1 may resolve the dilemma of the district court in the Jones case. The sentencing judge in Jones felt compelled to penalize the defendant for going to trial because he could not award less than a two-point reduction for accepting responsibility. See supra note 215.
228. My proposal would not wholly preclude judges from taking account of a defendant's attitude in sentencing: a judge could use remorse as a basis for sentencing at the bottom of the applicable range, or defiance for sentencing at the top. Remorse might also be a basis for departure - indeed, remorse-based departures would grow more easy under my proposal because it would be clear that the Commission did not already take remorse into account in section 3E1.1. However, under my proposal there would be less room overall for remorse-based considerations in sentencing. Attitude would be only one factor most judges would want to take into account in sentencing within a range, and the size of the range is itself rather limited in many cases. Departures would provide greater degrees of recognition for remorse, but would be subject to a much more searching form of appellate scrutiny than the current acceptance-of-responsibility determinations.
229. This observation has also been made by Judge Edward Becker of the Third Circuit. Comments of Judge Becker to Sentencing Workshop (New Haven, March, 1996).
230. This might happen, for instance, if a judge emphasized the remorse aspect of section 3E1.1, thereby preventing the provision from serving as the sort of intermediate assistance benefit that Judge Becker and I envision.
231. U.S.S.G. § 5H1.2.
232. Id. § 5H1.5.
233. Id. § 5H1.11.
234. Id.
235. Id.
236. The sentencing differential already seems to function predictably enough in Connecticut and perhaps in most other districts. However, Schulhofer's and Nagel's research suggests that there are at least a few districts in which the differential is insufficiently clear and reliable to encourage guilty pleas. See supra note 167. My proposal is aimed primarily at, and will have the greatest effect in, such districts, rather than districts like Connecticut.
237. In the spirit of current application note 1, my proposal specifies that a defendant cannot be penalized for failing to provide incriminating evidence or information. For instance, in the Austin case, the defendant would receive a two-point reduction for pleading guilty. The defendant might have qualified for the third point by providing information regarding the missing guns to the authorities. However, the defendant's two-point benefit could not be reduced for failing to provide such information.
238. See, e.g., Brady v. United States, 397 U.S. 742, 752 (1970) ("[W]ith the avoidance of trial, scarce judicial and prosecutorial resources are conserved for those cases in which there is a substantial issue of the defendant's guilt or in which there is substantial doubt that the State can sustain its burden of proof."); United States v. Quejada-Zurique, 708 F.2d 857, 861 (1st Cir. 1983) (arguing that any legitimate system that allows the negotiation of pleas will necessarily and permissibly feature plea inducements); Bryant, supra note 213, at 1292-93 (discussing cases upholding plea bargaining based on judicial economy). On the other hand, the elimination of the section 3E1.1 incentive might not actually result in a substantial decrease in plea bargaining. Even without a reward for doing so, many defendants may prefer to dispose of their cases as quickly and painlessly as possible. See MILTON HEUMANN, PLEA BARGAINING: THE EXPERIENCES OF PROSECUTORS, JUDGES, AND DEFENSE ATTORNEYS 69-71 (1978) (discussing eagerness of the defendants to plead guilty and "get it over with"). As a Connecticut public defender told me, for many defendants "the thought of appearing before a jury and having a judge glaring down at them is a real nightmare." Moreover, the elimination of section 3E1.1 might not, in fact, end the practice of rewarding guilty pleas. As will discussed further below, prosecutors and judges could still provide hidden plea discounts.
239. Barry Letter, supra note 14, at 5.
240. Sentencing differentials for guilty pleas seem an effective mechanism for discouraging defendants from going to trial, see Nagel & Schulhofer, supra note 145, at 524 n.56, 550, but the effect of section 3E1.1 on other forms of post-offense conduct is less clear. With respect to encouraging desired acts, many of the acts enumerated in application note 1, most of which would be carried forward in my proposal, are quite costly to the defendant or can only be carried out by limited classes of defendants. For instance, voluntary restitution may impose substantial costs and, in any case, is only available as an option to perpetrators of property crimes. With respect to the flipside of desirable post-offense conduct, discouraging disfavored acts, the record in Connecticut may provide some clues. In standard plea agreements in Connecticut, the government promises to recommend the reduction if the defendants cooperate with the probation office and disassociate themselves from criminal activities. According to prosecutors, up to 10% of the defendants evidence some form of continuing criminal conduct after entering into such agreements, generally by failing a drug test. A public defender informed me that defendants regard plea agreement forms as "lawyers' gobbledygook" and do not take into account the possibility of losing section 3E1.1 credit when considering presentence misconduct. However, a probation officer took the opposite position, opining that many defendants are, in fact, deterred from misconduct as a result of the possibility of receiving a longer sentence. In sum, both facets of the section 3E1.1 incentive - encouraging the good and discouraging the bad - may have an effect on post-offense conduct other than the plea-or-trial decision, though probably not a dramatic one.
241. Schulhofer & Nagel, supra note 145, at 547-49. Although aggressive probation officers may subvert these forms of plea bargaining through the information and recommendation contained in a presentence report, some judges may disfavor interference with plea agreements. See Letter from Francesca D. Bowman, Chair, Probation Officers Advisory Group to the United States Sentencing Commission, to Judge Richard P. Conaboy, Chairman of the United States Sentencing Commission 2 (Jan. 30, 1996) (concluding, based on national survey of probation officers, that courts "almost universally defer to the plea agreement, especially when it is more favorable to the defendant than the presentence report"). For further discussion of the discretionary authority of prosecutors under the Guidelines, see Freed, supra note 1, at 1723-24.
242. HEUMANN, supra note 234, at 158. Professor Heumann's conclusions were based on a study of state court practices in a nonguidelines context. However, a federal public defender assured me that precisely the same phenomenon of implicit plea bargaining occurs in the federal system today. He observed that defenders often enter guilty pleas without negotiating with the government for leniency first, because the belief that guilty pleas should be rewarded is so deeply entrenched that defendants may get the full benefits of a guilty plea without any assistance from the government. He further asserted that, were section 3E1.1 to be eliminated altogether, he would still advise his clients that they would receive more favorable treatment at sentencing if they pled guilty.
243. Richard S. Frase, Implementing Commission-Based Sentencing Guidelines: The Lessons of the First Ten Years in Minnesota, 2 CORNELL J.L. & PUB. POL'Y 279, 316-17 (1993) (analyzing regression analysis of Minnesota sentencing data). Research on experiments with plea-bargaining bans further reveals the deep roots of plea benefits in the criminal justice system: plea differentials persist even when negotiation is formally prohibited. See, e.g., Weninger, supra note 166, at 312.
244. Schulhofer, supra note 168, at 180 (discussing effect of remorse-oriented decisions on plea bargaining).
245. Wilkins, supra note 16, at 190-91.
246. See, e.g., Freed & Miller, supra note 89, at 176 (proposing relabeling § 3E1.1 a "plea benefit" and arguing that the benefit should "be presumptively granted to all defendants who plead guilty, except where the prosecutor demonstrates by clear and convincing evidence that a proportionately smaller discount is warranted by exceptional circumstances"); Barry Letter, supra note 14, at 6 (suggesting changes to § 3E1.1 that would result in an automatic one- or two-point reduction for a guilty plea).
247. By guaranteeing a defendant an automatic reduction in his sentence, the prosecution will be less likely to bargain over the criminal conduct charged on the presentencing report. This provides for a more honest application of the Sentencing Guidelines' offense levels for categories of criminal conduct. Thereby, the defendant will be charged with the crime he actually committed and still benefit from pleading guilty. Bryant, supra note 213, at 1303 (citations omitted).
248. See supra note 242.
249. The Freed-Miller proposal envisions a stronger presumption than I do: their presumption can only be overcome when a prosecutor demonstrates by clear and convincing evidence that a smaller discount is warranted by exceptional circumstances. I retain greater discretion in the hands of judges so that judges can more fully consider the whole range of a defendant's post-offense conduct than just the plea-or-trial decision. The Freed-Miller proposal also provides a substantially more generous benefit than does my proposal. Freed and Miller set the plea benefit at a level at which it represents a 40% discount, mirroring pre-Guidelines practices. Freed & Miller, supra note 89, at 176. In the interests of proceeding with caution and due deference to existing practices, I have left the plea benefit at the existing two-level reduction, which represents on average only a 15 to 20% discount. See id. The increase in the benefit proposed by Freed and Miller may be desirable; as the Schulhofer-Nagel research demonstrates, the two-point reduction is insufficient to discourage prosecutors and judges from employing less honest and open methods of rewarding guilty pleas. See Schulhofer & Nagel, supra note 145, at 547-49. The U.S. Judicial Conference has also endorsed an expansion of the scope of the section 3E1.1 benefit; its proposal permits up to a four-level reduction for defendants with an offense level greater than sixteen. Barry Letter, supra note 14, at 5. Nonetheless, I am mindful of Judge Kleinfeld's admonition that "the value to society of a guilty plea is less than its value to actors within the justice system." Kleinfeld, supra note 170, at 18-19. If guilty pleas are overencouraged (as Kleinfeld believes they would be with more than a two-point reduction), innocent defendants may be tempted to plead guilty and guilty defendants may escape with less prison time than is socially desirable.
250. The two defendants might be distinguished based on other aspects of the Guidelines besides section 3E1.1, such as departures or sentencing within the prescribed range. However, the range may not be substantial, and departures may be more or less difficult depending on the appellate circuit. Less honest methods of distinction are also available, such as undeserved section 5K1.1 departures, but the Guidelines should be wary of encouraging resort to such devices. Cf. Susan Winarski et al., U.S. Sentencing Commission Staff Working Group Report on § 3E1.1: The Acceptance of Responsibility Reduction, FED. SENTENCING REP. 336, 341 (1992) (arguing that one benefit of allowing judges to use section 3E1.1 to reward defendants "who make a good faith effort to redress the harms [they have caused]" would be to "remove the pressure on courts to resort to unguided departures").
251. The court is unable to distinguish between a begrudging, reluctant timely plea (for which the full three points must be awarded), and a timely 'plea plus' where the defendant pleads as well as shows genuine remorse, demonstrates assistance to the authorities, has undergone post-offense rehabilitative efforts, and/or some of the other factors which the guideline attempts to reward. Barry Letter, supra note 14, at 1. See also COMMITTEE ON CRIM. LAW AND PROBATION ADMIN., JUDICIAL CONFERENCE OF THE U.S., REPORT AND RECOMMENDATION OF THE JUDICIAL CONFERENCE OF THE UNITED STATES FOR AMENDMENT TO THE SENTENCING GUIDELINES 11 (1991) (recommending that the Commission consider revising section 3E1.1 "to recognize and encourage affirmative actions demonstrating acceptance of responsibility other than entry of a guilty plea"), reprinted in WORKING GROUP, supra note 33, at 48, 51.
252. Wilkins, supra note 16, at 191.
253. See, e.g., Winarski et al., supra note 246, at 341 (stating that one drawback of providing incentive for defendants to do more than just plead guilty is that such an incentive "may treat defendants differently based on whether they (or their friends or relatives) can afford to pay restitution, and it may be seen as unfair to defendants whose crimes cause harms that, by their nature, cannot be redressed easily or that otherwise do not fit the considerations listed in Application Note 1 . . .").
254. If this reform were adopted, section 3E1.1 should be restructured so that it could provide a net enhancement, rather than just a net reduction. Otherwise, defendants who go to trial would generally not be subject to penalties for obstruction or reckless endangerment because such defendants would presumptively not receive a section 3E1.1 adjustment anyway. In short, such defendants would have little to lose by obstruction or endangerment.
255. Interestingly, amendment 24 moves towards this end, but in a rather modest fashion. Proposed application note 2(b) makes clear that obstruction "does not necessarily disqualify the defendant" from receiving an acceptance-of-responsibility adjustment, although obstruction might weigh into the "totality of the circumstances" under which a defendant who pleads guilty may be denied the adjustment.
256. 18 U.S.C. § 3553.
257. Freed, supra note 1, at 1726-27.
258. Id. at 1684 n.5 (citing a survey of federal district judges).