concurring.
I concur in the judgment and in most of Judge Wellford’s opinion. I write sepa*1517rately, however, to record my individual thoughts on use of the “minimal indicia of reliability” test for determining the admissibility of hearsay — a test I would reject in favor of the “probable accuracy” test prescribed in U.S.S.G. § 6A1.3(a) — and on the argument advanced by defendant Catón as to the appropriate standard of proof in guideline sentencing. (Neither of the other defendants has raised the latter issue before the en banc panel.)
To make what follows intelligible, I need to explain that if one assumes Mr. Catón belonged in the criminal history category where the probation officer recommended he be placed under the guidelines (Category I), and if Mr. Caton’s offense level was as calculated by the district judge (Offense Level 14), the sentence range prescribed in the U.S. Sentencing Commission’s Guidelines Manual was a term of imprisonment of not less than 15 months and not more than 21 months.1 Judge Forester ultimately exercised the departure authority conferred by 18 U.S.C. § 3553(b) and sentenced Mr. Catón to imprisonment for 46 months. (This sentence — two months short of four years — was less than one-fifth the 20-year maximum sentence authorized by Congress in 21 U.S.C. § 841(b)(1)(C).)
Now to defendant Caton’s. argument. Relying on Judge Becker’s opinion in United States v. Kikumura, 918 F.2d 1084 (3d Cir.1990), Mr. Catón argues not only that hearsay evidence ought not to have been admitted at his sentencing hearing, but also that the factual findings on which his upward departure was based ought to have been made under “a reasonable doubt standard of proof (or at least a clear and convincing standard).... ” Kikumura, as I read it, does not support Caton’s argument.
The defendant in Kikumura — an international terrorist who was said by a confidential informant to have trained other terrorists at a camp in the Bekaa Valley of Lebanon before coming to the United States to maim or kill Americans by exploding anti-personnel bombs at a Navy recruiting office in New York City — was sentenced to imprisonment for 30 years.2 The defendant’s guideline range was only 27-33 months, so the trial court’s upward departure resulted in a sentence more than 10 times as long as the maximum within-range sentence. With a departure of that magnitude — said to be the largest departure in the history of federal guideline sentencing, see 918 F.2d at 1089 — the Third Circuit concluded that the sentencing proceeding could fairly be characterized as “a tail which wags the dog of the substantive offense.” Kikumura, 918 F.2d at 1100-01, quoting McMillan v. Pennsylvania, 477 U.S. 79, 88, 106 S.Ct. 2411, 2417, 91 L.Ed.2d 67 (1986). And under these truly unusual circumstances, the Third Circuit held, the hearsay statement of a confidential informant could be considered at sentencing only if found to be “reasonably trustworthy” (a standard I should have thought comparable to the one already prescribed by the Sentencing Commission, see U.S.S.G. § 6A1.3(a)), while the departure itself could only be based on facts established by evidence that was at least “clear and convincing.” Kikumura, 918 F.2d at 1100-03. The lower levels of procedural protection historically applicable in sentencing proceedings (i.e., “some minimal indici-um of reliability” for hearsay and a “preponderance of the evidence” standard for factfinding) had to be “ratcheted up,” the *1518Third Circuit said, because of the extraordinary magnitude of the departure from the guideline range. Id. at 1101.
The Kikumura court went out of its way to stress that the-preponderance standard and the minimal indicia of reliability test are “clearly appropriate” and “perfectly adequate” in most sentencing cases. Id. at 1100, 1101.3 “At a garden variety sentencing hearing,” where the sentencing court does not depart upwards “dramatically,” the Third Circuit stated without qualification that “a 'preponderance standard appl[ies]_” Id. at 1103. (Emphasis supplied.)
Mr. Caton’s case is definitely of the garden variety. The sentencing court’s upward departure of 25 months comes nowhere close to the departure of 327 months — more than 10 times the guideline maximum — imposed in Kikumura. There may well be other unusual situations where it would be appropriate to use the “clear and convincing evidence” standard held applicable in Kikumura — other circuits have recognized that possibility4 — but the instant case, in my judgment, does not present such a situation. This case is more like United States v. Mobley, 956 F.2d 450, 458-459 (3d Cir.1992), where the Third Circuit rejected a due process argument in the context of a six-month enhancement of a sentence that would otherwise have been 21 months.
One final point needs to be made: Mr. Catón cannot honestly blame his 46-month sentence on the Sentencing Commission. *1519His ■ is not a case where a discretionless trial judge, mindlessly applying a Rube Goldberg formula dreamed up by a committee of social scientists in Washington, was constrained to impose a sentence that any reasonable person would find aberrational. For better or worse, Mr. Catón was sentenced the old-fashioned way — by a conscientious and objective trial judge giving careful consideration to the entire picture, at least insofar as the elements of that picture were supported by what he considered to be sufficient indicia of reliability, and then exercising his best judgment as to the sentence that would be most appropriate for the particular individual standing before him.5
Sentencing, as our colleague Judge Jones recently said in a very eloquent opinion,6 “is and must remain an intensely human process.” No sensible person wants human beings to be sentenced solely by formula; if the sentencing guideline system is to work in the way that Congress intended, it is essential that judges who send criminals to prison retain a significant measure of discretion to do what seems fair and just in each particular, case. The judge who accepted- defendant Caton’s plea of guilty and set the sentence at 46 months recognized that he had such discretion, and he used it. Acting on the type of information that sentencing judges have been taking into account for hundreds of years, Judge Forester imposed the sentence he thought Mr. Catón deserved — and he did so despite the Sentencing Commission, not because of it.
Mr. Caton’s unhappiness with the result in this case reflects a desire that the discretion historically vested in sentencing judges be curtailed more severely than the guidelines curtail it now. I understand this desire, but I do not fully share it.
. As stated in the detailed exposition of defendant Caton’s case at pages 1506-07 of the opinion written by Chief Judge Merritt for the original panel majority, the probation officer recommended “relevant conduct” findings that would have produced a much higher guideline range: imprisonment for 151-188 months. The district judge, in a five-page opinion filed after he had conducted an evidentiary hearing on the matter, explicitly rejected this recommendation. The district judge (Forester, J.) also gave notice that "the Court is considering an upward departure from the [15-21 month] guideline range.”
. The factual details of this remarkable case are set forth at length in the trial court’s sentencing opinion, United States v. Kikumura, 706 F.Supp. 331 (D.N.J.1989). There was no room for doubt that the defendant was guilty of the charges against him; through his lawyer, William Kunt-sler, he waived trial by jury and stipulated to "all of the moving facts that led to the indict-ment_” Id. at 333.
. Kikumura thus cannot be squared with the argument, advanced by Judge Merritt in the case at bar, that Specht v. Patterson, 386 U.S. 605, 87 S.Ct. 1209, 18 L.Ed.2d 326 (1967), and McMillan v. Pennsylvania, 477 U.S. 79, 106 S.Ct. 2411, 91 L.Ed.2d 67 (1986), require application of the Confrontation Clause to the federal guideline sentencing system. Every court of appeals to have addressed such a proposition has rejected it. See Becker and Orenstein, “The Federal Rules of Evidence After Sixteen Years,” 60 Geo. Wash.L.Rev. 857, 888-89 (1992), listing cases at n. 160.
I would add, in this connection, that in Specht Justice Douglas specifically endorsed Williams v. New York, 337 U.S. 241, 69 S.Ct. 1079, 93 L.Ed. 1337 (1949), where the Supreme Court, speaking through Justice Black, concluded that on the facts presented the Constitution did not bar a sentencing judge from considering hearsay information even though the judge used the information in deciding to impose a sentence of death.
As to then-Justice Rehnquist’s opinion in McMillan, it specifically endorsed and applied Patterson v. New York, 432 U.S. 197, 97 S.Ct. 2319, 53 L.Ed.2d 281 (1977), where the Supreme Court rejected a contention that facts affecting the severity of punishment must be proved “beyond a reasonable doubt." McMillan, 477 U.S. at 83-85, 106 S.Ct. at 2414-16. The McMillan court also rejected an argument that due process required use of a "clear and convincing evidence” standard; at least in the case before it (which, as Kikumura notes, was not one in which a sentencing tail was wagging a substantive offense dog), the McMillan court had little difficulty in concluding that "the preponderance standard satisfies due process." Id., 477 U.S. at 91, 106 S.Ct. at 2419. Notwithstanding that the sentencing scheme at issue in McMillan took away the trial court’s discretion to impose a sentence of less than 60 months if the court found by a preponderance of the evidence that the defendant visibly possessed a firearm during the commission of the offense, the McMillan Court could "see nothing in [this] scheme that would warrant constitutionalizing burdens of proof at sentencing.” Id. at 92, 106 S.Ct. at 2419.
Judge Becker and Prof. Orenstein have suggested that the Supreme Court, acting through the appropriate advisory committee, ought to amend the Federal Rules of Evidence — a process subject to congressional approval under 28 U.S.C. § 2074 — to reflect the changed role of sentencing hearings by giving convicted persons “more protections.” Becker and Orenstein, op. cit., at 890. There may be situations where additional protections would prevent injustice, but I doubt that such a situation is presented in Mr. Caton’s case. The guidelines have already given Mr. Catón more protection against the use of hearsay than he would have had before their adoption, because they require more than "some minimal indicia of reliability” — they require sufficient indicia of reliability to support the "probable accuracy" of the hearsay. U.S.S.G. § 6A1.3(a). I can see no justification for conflating “probable accuracy" and "minimal indicia of reliability,” see Note, "Confrontation and Sentencing,” 105 Harv.L.Rev. 1880, 1884 (1992) — and our en banc court endorses the probable accuracy test in the majority opinion— but I am satisfied that the hearsay used against Mr. Catón met the appropriate test. So too, in my view, did the evidence used against defendants Silverman and Woodard.
. See United States v. Lam Kwong-Wah, 966 F.2d 682 (D.C.Cir.1992); United States v. Schuster, 948 F.2d 313, 315 (7th Cir.1991); United States v. Townley, 929 F.2d 365, 370 (8th Cir.1991); United States v. St. Julian, 922 F.2d 563, 569 n. 1 (10th Cir.1990).
. Judge Merritt, in his dissent, suggests that to a degree not likely to have occurred before the Sentencing Commission arrived on the scene, the sentence imposed in this case was somehow controlled by the prosecutor rather than by an unbiased, objective, neutral federal judge. The careful and detailed statement of the facts at pages 1506-07 of the opinion Judge Merritt wrote for the original panel majority does not support this suggestion.
. United States v. Davern, 970 F.2d 1490, 1516 (6th Cir.1992) (Jones, J., dissenting).