dissenting.
Although the majority concludes that an Alford plea falls within the definition of “prior sentence” under U.S.S.G. § 4A1.2(a)(l), I cannot agree. Such a reading need not and should not be read into the sentencing guidelines, for to do so may well lead to an unfair result.
An Alford plea results when a defendant pleads guilty and consents to the imposition of a sentence while still proclaiming his innocence of the charged offense. See North Carolina v. Alford, 400 U.S. 25, 91 S.Ct. 160, 27 L.Ed.2d 162 (1970); see also Wilson v. Lawrence County, Mo., 154 F.3d 757, 758 n. 1 (8th Cir.1998). The Alford plea is unique: it is a form of guilty plea entered into despite protestations of innocence. It is a plea that a defendant renders solely out of practical considerations and not as a mechanism for admitting guilt. Before a court may accept an Alford plea, the court must find “strong evidence of actual guilt” independent of the defendant’s statements. Alford, 400 U.S. at 37, 91 S.Ct. 160.
Section 4A1.1 of the sentencing guidelines governs the calculation of a defendant’s criminal history category. Section 4A1.2 provides definitions and instructions for implementing § 4A1.1. When defining “prior sentence,” § 4A1.2(a)(1) of the guidelines states: “The term ‘prior sentence’ means any sentence previously imposed upon adjudication of guilt, whether by guilty plea, trial, or plea of nolo conten-dere, for conduct not part of the instant offense.” Section 4A1.2(a)(l) does not mention Alford pleas. It does, however, require an “adjudication of guilt.” Thus, when considering the proper application of the language of § 4A1.2(a)(l) to Alford pleas, I am forced to conclude that such a plea is not a prior sentence for purposes of criminal history calculations — at least not without indicia that it represented an actual “adjudication of guilt.”
As a general matter, the structure of the sentencing guidelines punishes an individual for both the instant offense and all prior offenses. It does so by totaling points on a matrix comprised on one axis by the characteristics of the instant offense, and on the other axis by points accrued through prior offenses. Normally, the law requires that the government prove conduct of the instant offense by at least a preponderance of the evidence. However, when sentencing, prior convictions need not be proven: they are automatically entered on the criminal history axis. The prior conviction itself is proof of the defendant’s guilt. In a situation such as this one, where the defendant argues that one of his prior convictions was based on an Alford plea, the usual assumptions about prior convictions may not necessarily hold. While an Alford plea should require independent proof of guilt to sustain the conviction, there may be instances where that is not the case.
Here, Mackins asserts that no independent evidence of guilt exists. On the record before this court, there is no evidence that Mackins did commit the crime to which he pleaded guilty under protestations of innocence. It is clear that Mac-kins was concerned primarily with avoid*271ing imprisonment in Rikers Island prison at the time he agreed to an Alford disposition on charges of grand larceny. He had spent more than one year at Rikers Island awaiting various pre-trial stages of his case. Had he persisted in seeking a trial, he would have been returned to Rikers Island for an indefinite period of time while the case proceeded. In addition, the state had charged him with attempted murder, grand larceny in the fourth degree, robbery in the first degree, criminal possession of a weapon, attempted murder in the second degree, six counts of robbery in the first degree, attempted grand larceny in the fourth degree, and four counts of kidnapping. After Maekins’s co-defendant was acquitted at a separate trial, the state offered to dispose of this case if he would plea (under protestations of innocence) to a Class A misdemeanor for which Mackins would serve no time beyond the year he had already served at Rikers Island and for which he would not be subjected to a parole violation. Mackins had a tremendous incentive to take the plea rather than return to Rikers Island and face the risk of trial on the multitude of charges against him. The Alford plea represented a virtual “no lose” situation in which the circumstances now indicate Mackins may not have been guilty of one of the crimes that contributes to his criminal history category of IV.
The guidelines recognize the difference between an Alford plea and other types of pleas. The Supreme Court decided the Alford case more than ten years before the sentencing guidelines incorporated § 4A1 in 1987. We can presume that the Sentencing Commission intentionally excluded Alford pleas from § 4A1.2 because the Commission did not include the term in its list of what constitutes “adjudication of guilt.” Additional support for this presumption exists because the Sentencing Commission expressly refers to the “Alford plea” in § 8A1.2 of the guidelines. The commentary at n. 3(g) of § 8A1.2 states: “ ‘Prior criminal adjudication’ means conviction by trial, plea of guilty (including an Alford plea), or plea of nolo contendere.” Because the Sentencing Commission expressly included Alford pleas within the definition of “guilty plea” in § 8A1.2, yet simultaneously excluded such pleas from § 4A1.2, we must assume that this exclusion was intentional. If the Commission intended to include Alford pleas in both sections it would have done so.
In addition, fundamental fairness requires that we allow district courts the discretion to examine the particular circumstances of cases in which criminal history calculations include Alford pleas. The sentencing guidelines are rigid, and they allow very little discretion to district court judges. This ease presents this court with the opportunity to grant district judges discretion in an area where unique circumstances are relevant to the fundamental fairness of a defendant’s ultimate sentence. Because Alford pleas are submitted under protestation of innocence, a district judge must be particularly vigilant about ensuring that sufficient evidence was adduced at the time the Alford plea was taken before including an Alford plea in a calculation of criminal history category; such vigilance is necessary to ensure that the guilty plea is properly treated as an adjudication of guilt.
David Mackins, who is twenty-four years old, appeals his thirty-year sentence for various drug offenses. At sentencing, the District Court adopted the recommendations of a pre-sentence report that assessed Mackins a total of eight criminal history points, placing him in a criminal history category of IV. Excluding the Alford plea leaves Mackins with a criminal history category of III and could result in a lesser sentence. In an instance such as this one, a district court should have the freedom to hold an evidentiary hearing in which the court can determine whether independent evidence of guilt underlay the Alford plea.
*272In a recent law review article discussing the federal sentencing guidelines, two law professors, Marc L. Miller and Ronald F. Wright, explain the problems of removing substantial discretion from judges who preside over a process as complicated as federal sentencing:
It is hard to know where to begin in describing the disaster that has become federal sentencing reform over the past twenty years. This disaster is all the more disheartening because the reform started with so much promise. Few reform efforts — especially in the area of criminal justice, and especially in the federal system — have had as much hope or thought at their core.
However, the sentencing guidelines that emerged from the new administrative process have been one of the great failures at law reform in U.S. history. The collapse was quick, and it has become difficult to defend the current system as the reasoned and principled system we believe Congress and reformers envisioned. The current guidelines are widely hated and in many ways dysfunctional. The expert agency that creates and monitors the guidelines' — the U.S. Sentencing Commission — has morphed into an ineffectual caricature of an administrative agency. Rather than achieving honest, wise or equal sentencing, the primary effects of the guidelines (albeit in conjunction with other developments) have been to occupy increasing portions of the federal judicial workload, to raise the analysis of probation officers above the arguments of lawyers and the reasoning of judges, to shift the type of offenders in the federal system, to shift offenders from state to federal systems, and to pour offenders into federal prisons, for longer and longer periods.
Marc L. Miller & Ronald F. Wright, Your Cheatin’ Heart(land): The Long Search for Administrative Sentencing Justice, 2 BUFFALO CRIM. L. REV. 723, 723, 726 (1999) (footnotes omitted).
Many federal judges and others agree that the guidelines are a disaster and can result in unfair, heavy prison sentences. Here, Mackins must serve approximately thirty years in prison, most of his adult life, with a substantial part thereof attributable to an alleged prior crime that he steadfastly denies. This case relates to a small area of the guidelines where this appellate court by its decision can improve guideline administration by giving the sentencing judge the power to inquire and determine whether or not independent evidence of guilt supports the Alford plea.
I would remand this case to the District Court for possible resentencing consistent with this opinion.