U.S. v. PREVATTE

RIPPLE, Circuit Judge.

After the district court reconsidered their sentences, see United States v. Prevatte, 16 F.3d 767 (7th Cir. 1994), Russell Prevatte and Robert Soy were each resentenced to 636 months of imprisonment for maliciously damaging or destroying property by means of an explosive, in violation of 18 U.S.C. § 844(i). The defendants appeal that sentence. For the following reasons, we affirm in part, reverse in part, and dismiss the appeal in part for lack of appellate jurisdiction.

I [2] BACKGROUND A. Facts

We set forth the facts relevant to this litigation in United States v. Prevatte, 16 F.3d 767 (7th Cir. 1994). We assume familiarity with that opinion. In summary, Russell Prevatte, with high school friends Douglas Bergner and Jerry Williams, commenced a series of burglaries in 1990. In 1991, a fourth individual, Robert Soy, joined them.

Williams, a probationary police officer with the Hammond Police Department during the fall of 1991, attended the Indiana State Police Academy from September 16 through December 20, 1991. While at the Academy, he displayed an interest in explosives and researched bombs in the academy library, as well as through the academy instructors. Prevatte later read a book Williams had suggested, the Anarchists Cookbook, and the two discussed, when Williams was home from the academy on weekends, how to manufacture pipe bombs and how to use them near gas meters as a diversionary tactic for burglaries.

On December 23, 1991, Prevatte and Soy detonated the first pipe bomb in a residential alley in Whiting, Indiana. This "test" bombing apparently was designed to measure the damage caused by the bomb and to gauge the response time of emergency services so that later burglaries could be synchronized accordingly. The bomb punctured a gas meter some fifty feet away and resulted in the death of Emily Antkowicz, an elderly woman whose backyard abutted the alley and who was about thirteen feet from the bomb at the time of explosion. It is this incident that led to the charge and sentence that is before us in this second appeal. The later criminal activity of these two defendants and their confederates is set forth in detail in our earlier opinion and need not be repeated here. Suffice it to say that a grand jury returned a twenty-one count indictment against Prevatte and Soy. Among other things, the grand jury charged the defendants with: (1) maliciously conspiring to damage or destroy property by means of an explosive in violation of 18 U.S.C. § 844(i); (2) maliciously damaging or destroying property by means of an explosive in violation of 18 U.S.C. § 844(i); and (3) making a firearm, statutorily termed a "destructive device," in violation of 26 U.S.C. § 5845(f), 5861(f) and 18 U.S.C. § 2.

A jury convicted Prevatte and Soy of fourteen of the twenty-one counts of the indictment, including the violation of 18 U.S.C. § 844(i) that involved the death of Emily Antkowicz. At sentencing, the defendants submitted that the district court should apply the guideline for second degree murder as the one most closely analogous to the conviction. The district court disagreed. The court found that the December 23, 1991 test bombing was, within the meaning of the statute, an act of arson because it involved destruction of property. At the imposition of sentence, the court first noted that 18 U.S.C. § 844(i) refers the court to U.S.S.G. § 2K1.4(c). That section in turn directs the court to the "most analogous" guideline offense from Chapter Two, Part A. The court then noted that the murder statute, 18 U.S.C. § 1111, provides that every murder committed in perpetration of or an attempt to perpetrate any arson, burglary or robbery, is murder in the first degree. Because the offense charged was arson, the court held that first degree murder was the most analogous offense. The court therefore determined that the applicable U.S.S.G. assessment was: total offense level 43, criminal history category I. Accordingly, the court sentenced the defendants to life imprisonment.

On their first appeal to this court, Prevatte and Soy submitted that the district court erred in allowing the jury to hear evidence of the uncharged crimes surrounding the bombing dates. Both defendants disputed the district court's sentencing determination on two grounds. First, the defendants challenged the district court's decision that the guideline for first degree murder, U.S.S.G. § 2A1.1, provided the most closely analogous guideline. Second, the defendants contended that, under § 844(i), life sentences may be imposed only after a jury recommendation, pursuant to 18 U.S.C. § 34. We held that the first degree murder guideline was indeed applicable to the charged crimes. We stated:

[W]e believe that the bombing at issue is sufficiently similar to arson to apply the first degree murder guideline on this basis. This conclusion rests on our understanding of the language and history of 18 U.S.C. § 844(i).

United States v. Prevatte, 16 F.3d 767, 780 (7th Cir. 1994). We also noted that, in interpreting the statutory language of § 844(i) in conjunction with 18 U.S.C. § 34, it is unlawful to impose a life sentence absent jury direction. Because the issue of the life sentence had not been submitted to the jury, we held that the imposition of a life sentence was reversible error. We also noted that the district court had not undertaken any analysis of the mental state of the defendants as mandated in application note 1 to U.S.S.G. § 2A1.1. That application note recognizes that a life sentence may not be appropriate for all convictions of first degree murder and notes that a downward departure "may be warranted." The extent of the departure, the note continues, "should be based on the defendant's state of mind (e.g. recklessness or negligence), the degree of risk inherent in the conduct, and the nature of the underlying offense conduct."

On resentencing, the district court considered the defendants' state of mind. The court held that the death of Emily Antkowicz was caused by the defendants' "recklessness and reckless state of mind and behavior." Tr. II at 2223-24; see Tr. IV at 2203-04. It articulated two substantial factors that led it to that conclusion: (1) The defendants knew they were using explosive materials, and obtained information specifically on pipe bombs; (2) The defendants picked a residential area in which to gauge the police and fire department reaction time to explosions because they were aware that such a location would demand quick reaction time. Tr. II at 2223. The district court stated that it was "departing downward from the sentence called for by the murder statute, but not downward in the classification." Tr. II at 2225. It sentenced each defendant to 636 months of imprisonment on the count that involved the death of Emily Antkowicz. Concurrent terms of imprisonment were imposed on the remaining counts.

II [11] DISCUSSION

The defendants submit one challenge to their sentence. Prevatte and Soy contend that the district court erred in resentencing because it did not depart downward in imposing a sentence, as it had stated it would. Restated, the defendants have asked this court to determine whether the district court properly applied guideline § 2A1.1.

1.

We begin our review with a statement of the basic limitations that circumscribe our authority to review sentencing determinations of the district courts. By statutory command, our review of a sentence is limited to cases in which the sentence was (1) "imposed in violation of law," (2) "imposed as a result of an incorrect application of the sentencing guidelines," (3) "outside the applicable guideline range," or (4) "unreasonable." 18 U.S.C. § 3742(e), (f); see United States v. Sablotny, 21 F.3d 747, 749 n. 2 (7th Cir. 1994). In those cases, in which we do have jurisdiction, we must evaluate sentences imposed under the Guidelines with "due deference" to the district court's application of the Guidelines to the facts of the specific case. 18 U.S.C. § 3742(e); see United States v. Randall, 947 F.2d 1314, 1320 (7th Cir. 1991). As a result of this statutory scheme, this court lacks jurisdiction to review, at the request of the defendants, a district court's discretionary refusal to depart downward from the sentencing range determined by the Guidelines or to review the extent of the downward departure. United States v. Blackwell, 49 F.3d 1232, 1241 (7th Cir. 1995); United States v. Shaffer, 993 F.2d 625, 628-29 (7th Cir. 1993) ("We also lack jurisdiction to review the extent of a downward departure. . . . We can entertain neither a claim that the court did not depart enough, nor a claim that no departure was made."). If a district court denies a downward departure on the mistaken belief that it lacked authority to depart, we treat that misapprehension as an error of law over which we have appellate jurisdiction. United States v. Canoy, 38 F.3d 893, 903 (7th Cir. 1994).

Section 3742(e) states, in pertinent part:

The court of appeals shall give due regard to the opportunity, v of the district court to judge the credibility of the witnesses, and shall accept the findings of fact of the district court unless they are clearly erroneous and shall give due deference to the district court's application of the guidelines to the facts.

We turn now to the case before us. The parties suggest different characterizations of the defendants' contentions. The defendants cast those contentions in terms of questions of law in order to invoke our appellate jurisdiction. The government urges that the defendants have presented, at bottom, a contention that the district court did not depart downward sufficiently, an issue over which we do not have jurisdiction. We believe that the most accurate estimate of the situation contains elements of both of these competing characterizations.

At the outset, we do not believe that there is a serious question about our authority to determine whether the district court complied with our directions from the first appeal. To the extent that the appellants raise that issue, we can consider it. In our earlier opinion, we directed the district court to impose a sentence other than life and to consider the applicability of application note 1 to U.S.S.G. § 2A1.1. We turn first to the issue of whether the court imposed a life sentence. In United States v. Martin, 63 F.3d 1422 (7th Cir. 1995), this court held that "where a legislatively enacted sentencing scheme has expressly deprived a court of the possibility of imposing a life sentence, a sentence for a term of years exceeding the defendant's approximate life expectancy would ordinarily constitute an abuse of discretion." 63 F.3d at 1434. At the time of sentencing, the district court did not have the benefit of the guidance contained in Martin. Therefore, in sentencing the defendants to 636 months of incarceration, the court quite understandably did not focus on the standard now set out in Martin. Accordingly, we believe that the court ought to revisit the matter with the guidance of Martin. Determining the life expectancy of the individual defendant is a matter that ought to be addressed by the district court.

In our earlier appeal, we also directed the district court to consider the applicability of application note 1 to § 2A1.1 of the Guidelines. As we have mentioned earlier, that note provides that, when the conviction of first degree murder is predicated on a theory other than premeditated killing, life imprisonment is not necessarily the appropriate sentence and that, in such circumstances, a downward departure "may be warranted." Our direction to the district court, therefore, was to consider whether, on the facts of this case, a downward departure was warranted. The district court complied with our mandate when it considered the possibility of a lower sentence. In sum, the district court effectively carried out this court's order by departing to an extent based upon the defendants' "state of mind (recklessness or negligence), the degree of risk inherent in the conduct, and the nature of the underlying offense conduct." U.S.S.G. § 2A1.1, comment. (n. 1).

Having determined that the district court complied with our mandate, we now examine the remaining contentions. As we have noted earlier, we normally have no jurisdiction to review, at the request of the defendant, the decision of the district court as to whether it ought to depart downward and, if so, the extent of that departure. Shaffer, 993 F.2d at 628-29. A charitable reading of the appellants' submission suggests, however, that one issue of law may be present that is subject to our review. The appellants appear to contend that, although the district court had the discretion to depart or not to depart from the guideline range for murder in the first degree, it was obliged, once it made the decision to depart, to impose a sentence that would have been imposed for second degree murder. The appellants point out that the district court found that the placement and detonation of the bomb amounted to "recklessness and reckless state of mind and behavior." Their conduct should have been punished, they therefore argue, as second degree murder.

We do not read application note 1 as cabining the discretion of the district court to that degree. The application note quite explicitly suggests that a departure below that prescribed for second degree murder or for the underlying offense is not likely to be appropriate. This notation is hardly a directive to the district court that any departure must, as a matter of law, reduce the sentence to the level of second degree murder. To hold that a departure must correspond to the base offense level stipulated in § 2A1.2, Second Degree Murder, every time the court finds that a defendant's mental state was less than "intentionally or knowingly," cf. U.S.S.G. § 2A1.1, comment. (n. 1), would negate the congressional determination that death resulting from certain felonies, such as arson, should be punished, not as second degree murder, but as first degree murder. If the Congress had intended that only murders that were premeditated and involved explosions be treated as first degree, there would be no reason to begin the sentencing analysis with § 2A1.1, First Degree Murder, as mandated by 18 U.S.C. § 1111. Indeed, the district court's redetermination of the sentence in this case demonstrates the need for the flexibility that the application note gives to a sentencing court. The district court commented quite extensively on the mental state of the defendants at the time of the crime. We set forth the pertinent parts in the margin. This analysis can be read as a determination by the district court that the defendants engaged in conduct that, although not premeditated, involved a high degree of recklessness and warranted punishment between the level that would be employed for premeditated murder and the level that would be employed for a murder committed recklessly but not in the aggravated manner exhibited here. Such a determination is clearly permissible under the congressional determination concerning the punishment of murder committed in the course of arson. The sentence corresponds to an offense level of 42, which provides that an individual be sentenced to "360 [months] — life."

At Mr. Prevatte's resentencing, the district court stated:

This Court finds that the death of Emily Antkowicz was not caused by the negligence of the defendant, but instead by the defendant's recklessness and reckless state of mind and behavior.

In reaching this conclusion, the Court has considered:

One. That the defendants knew that they were dealing with explosives materials, and in fact obtained information with regard to pipe bombs.

Two. That the defendants wanted to check reaction time of police to pipe bombs, therefore, the defendants picked a residential area and produced a pipe bomb which they expected and hoped would get enough attention to demand quick reaction of police and firemen.

And again, as I had indicated earlier, they did this as opposed to picking a field or a farm where there would be no residences in close proximity. By the placing of a bomb in a vacant lot far away from active residences would not get the results that the defendants wanted.

By the placing of an active bomb where people live, and especially close to their residences and especially where homes were in close proximity is not negligence, in this Court's opinion, and was by far extremely reckless behavior.

Tr. II at 2223-24 (emphasis supplied).
At Mr. Soy's resentencing, the court stated:

[T]his Court finds that the death of Emily Antkowicz was not caused by the negligence of the defendant, but instead by the defendant's reckless acts and behavior. . . .

Four. The placing of an active bomb where people lived, and especially where people lived at or near the placement of that bomb, and especially where homes were in close proximity is in this Court's opinion not negligent but by far extremely reckless behavior.

And I have also note (sic) that in the letter even the defendants' attorneys indicated that this was basically a reckless act, the letter of 7-28-94.

Tr. IV at 2203-04 (emphasis supplied).

The remainder of the appellants' argument is appropriately characterized as a contention that the district court did not depart downward to a sufficient degree. As we have already made clear, we do not have jurisdiction to review this determination of the district court. See United States v. Dean, 908 F.2d 215, 217 (7th Cir. 1990), cert. denied, 501 U.S. 1206, 111 S.Ct. 2801, 115 L.Ed.2d 974 (1991).

2.

The district court, having observed the defendants and having heard the evidence in the case, clearly is in a better position to evaluate comprehensively the appropriateness of the sentence. That court considered all of the relevant factors, and acted within its discretion. We have no authority to substitute our view of the record. As the case comes to us on a cold record, the sentence imposed of 53 years for a twenty-six (Mr. Prevatte) and twenty-four (Mr. Soy) year-old seems to be a stiff one, but the district court's opportunity to assess the culpability of the defendants and the need for their incarceration was superior to the one afforded us. We assume that, if the statutory conditions are met, the Director of the Bureau of Prisons will fulfill his responsibilities in the future.

18 U.S.C. § 3582(c) in relevant part provides:

(c) Modification of an imposed term of imprisonment. — The court may not modify a term of imprisonment once it has been imposed except that —

(1) in any case —

(A) the court, upon motion of the Director of the Bureau of Prisons, may reduce the term of imprisonment, after considering the factors set forth in section 3553(a) to the extent that they are applicable, if it finds that — (i) extraordinary and compelling reasons to warrant such a reduction;

* * * * * *

(2) in the case of a defendant who has been sentenced to a term of imprisonment based on a sentencing range that has subsequently been lowered by the Sentencing Commission pursuant to 28 u.S.c. [§] 944(o), upon motion of the defendant or the Director of the Bureau of Prisons, or on its own motion, the court may reduce the term of imprisonment, after considering the factors set forth in section 3553(a) to the extent that they are applicable, if such a reduction is consistent with applicable policy statements issued by the Sentencing Commission.

Conclusion

For the aforementioned reasons, we affirm in part, reverse in part, and dismiss the appeal in part for lack of appellate jurisdiction. On remand the district court may adjust the sentence to ensure that the life expectancy of each of the defendants is appropriately considered.

AFFIRMED in part, REVERSED in part and DISMISSED in part

[24] POSNER, Chief Judge, concurring.

I join the opinion of the court, but add some thoughts on the life-sentence issue. The issue is exceptionally difficult, and unfortunately the difficulty is disproportionate to its actual importance other than to the two defendants. But with so much potentially at stake for them, we owe it careful consideration.

Until last year, the federal criminal code forbade the imposition of a life sentence for crimes involving the destruction of airplanes, motor vehicles, or their places of storage unless a jury recommended a life sentence. 18 U.S.C. § 34. (The requirement of a jury recommendation was repealed by the Violent Crime Control and Law Enforcement Act of 1994, but there is no contention that the repeal affects this case.) There was no such recommendation in this case. So the first time this case was here we vacated the life sentences that the district judge had imposed on the two defendants. 16 F.3d 767, 783-85 (7th Cir. 1994); see also United States v. Williams, 775 F.2d 1295, 1299 (5th Cir. 1985); United States v. Hansen, 755 F.2d 629, 631 (8th Cir. 1985); Ruiz v. United States, 365 F.2d 500 (3d Cir. 1966); cf. Ex parte Goss, 262 S.W.2d 412 ( 159 Tex.Crim. 235, App. 1953), overruled on other grounds in Ex parte Hill, 528 S.W.2d 125, 127 (Tex.Ct.App. 1975). On remand the judge sentenced each of them to 53 years. One defendant was at the time of sentencing 26 years old, the other 24.

The government believes that a jury recommendation is not required for any term of years, no matter how long. It asked for a 100-year sentence for one of the defendants. Yet, with parole having been abolished, it would not make any sense that I can see to interpret 18 U.S.C. § 34 to require a jury recommendation only in cases in which the word "life" appears in the sentence. The qualification ("with parole having been abolished") is important. In a system with parole, a term of years means what the parole board wants it to mean. Before parole for federal prisoners was swept away (effective in 1987) by the Sentencing Reform Act of 1984, a prisoner was eligible for parole after he had served one-third of his sentence or 10 years, whichever came first. 18 U.S.C. § 4205(a). So if he was sentenced to 100 years in prison, or for that matter to life in prison, he would be eligible for parole after 10 years. A sentence to a term of years, no matter how long, was not a sentence of life imprisonment; nor, for that matter, was a sentence of life imprisonment.

It did not follow that there was no difference between a sentence of life imprisonment and one of 30 years, the shortest sentence that would produce the same date of eligibility for parole as a sentence of life imprisonment. The judge's choice between these sentences would signal his view of the gravity of the defendant's conduct, and this might influence the parole board's choice of the actual, as distinct from the earliest possible, parole date. There was even authority, based on 18 U.S.C. § 4205(b)(1), that the judge could, in sentencing to a term of years in excess of 30 years, postpone the date of eligibility for parole beyond 10 years. E.g., United States v. Tidmore, 893 F.2d 1209 (11th Cir. 1990); Rothgeb v. United States, 789 F.2d 647, 652 (8th Cir. 1986). Several circuits, however, our own included, disagreed — precisely because the effect would be to empower the judge to make a term of years a heavier sentence than life imprisonment. United States v. Fountain, 840 F.2d 509, 517-23 (7th Cir. 1988); United States v. Hagen, 869 F.2d 277, 280-81 (6th Cir. 1989); United States v. Castonguay, 843 F.2d 51 (1st Cir. 1988).

With parole abolished and (another innovation of the Sentencing Reform Act) good-time credits reduced to a maximum of 14.7 percent of a sentence of more than four years (see 18 U.S.C. § 3624(b)(1)), a judge could — if we allowed him — use a sentence to a term of years to imprison a defendant for his natural life, thus circumventing the requirement in 18 U.S.C. § 34 of a jury recommendation for a life sentence. This use or rather misuse would be plain if the judge chose a term that was so long, given the defendant's age, that the defendant was certain (barring some medical breakthrough as yet unforeseen) to die before he completed the term. A 150-year term would do the trick, regardless of the age of the defendant. So would a 120-year term for a 30-year-old, or a 70-year term for a person of 80. Good-time credits would not shorten those terms to the point where they could be survived; the terms in my examples would be shortened only to 128, 102, and 60 years.

Would such sentences be illegal unless the jury had recommended life imprisonment? Besides the present case and Martin, of which more later, there are no cases under section 34 itself, although this is not surprising given its limited scope. Cases like Rothgeb v. United States, supra, 789 F.2d at 651, hold, unexceptionably, that if a statute authorizes life imprisonment or, in the alternative, imprisonment for a term of years, the judge's picking a term so long (210 years in that case) as to be tantamount to life is not reversible error. But the absence, in our case, of a jury recommendation for a sentence of life in prison disempowered the sentencing judge to choose between a life term and a term of years, so if he used a term of years to impose a life sentence he was evading a limitation on his authority. Ruiz v. United States, supra, 365 F.2d at 501, declined to rule that "under some circumstances a term of years greater than the prisoner's life expectancy may not be imposed," but without explaining what those circumstances might be. A state case dealing with a statute similar to 18 U.S.C. § 34 invalidated a 75-year sentence, remanding for the imposition of a sentence "reasonably expected to be less than life." Stewart v. State, 372 So.2d 257, 258-59 (Miss. 1979). See also United States ex rel. Curtis v. Blackburn, 748 F.2d 1047 (5th Cir. 1984). And we have expressed skepticism "that `any term of years' means `any term of years less than the age of the universe' rather than `any term of years less than life.'" United States v. Fountain, supra, 840 F.2d 509, 518.

It can never be certain that a federal prisoner, or any other prisoner, will not be released until some fixed number of years (the term of the sentence minus the maximum good-time credits) have elapsed or, in the case of a sentence of imprisonment for natural life, until the prisoner has died. There are always escape hatches, such as executive clemency, or section 70002 of the Violent Crime Control and Law Enforcement Act of 1994, 18 U.S.C. § 3582(c)(1)(A), which authorizes (in "extraordinary" circumstances) the release of a federal prisoner who is at least 70 years old and has served at least 30 years of his sentence. But these escape hatches do not warrant our treating a life sentence, or a sentence to a term of years so long as to amount to the same thing, as not really so severe.

The 53-year sentences in this case are not so long as to be the certain equivalent of life sentences. Life expectancy "at any given age is the average number of years remaining to be lived by those surviving to that age on the basis of a given set of age-specific rates of dying." National Center for Health Statistics, Vital Statistics of the United States, 1990, vol. II, mortality, part A, § 6 (life tables), p. 6 (Public Health Service 1994). Currently, the life expectancy of a 26-year-old white American male (the defendants are white) is 48.4 years, and of a 24-year-old 50.3 years. The actual prison terms that these defendants would have to serve if their sentences stand, after subtraction of their maximum good-time credits, would be 45 years, which is slightly short of their life expectancy. And as the definition given above makes clear, a "life expectancy" is an estimate merely of how long the average person of a given age, nationality, and sex (and of course more precise classifications are possible — one might speak of the life expectancy of a person of a given age, etc. who had had a particular form of cancer or lived in a particular area or was of a particular race) is expected to live. Roughly half (not exactly, because we are speaking of the average rather than median age of death) will die earlier. By the same token, roughly half will survive. This shows that even a sentence equal to (rather than, as here, slightly shorter than, at least when the maximum good-time credits are subtracted) a person's life expectancy is less severe than a life sentence. A sentence equal to a person's life expectancy, of which the sentences in this case are an approximation, is a life sentence with a probability of approximately .5, which in the serious affairs of life is a very high probability; imagine being told that you have a 50 percent probability of living out the week. And as I have said, even a sentence of "natural life" is not a life sentence with a probability of 1.

Should a sentence equal to the defendant's life expectancy be classified as a life sentence for purposes of section 34? If the answer is no, where should the line be drawn? If yes, what about the slightly shorter sentences ( if maximum good-time credits are subtracted) imposed in this case? And should good-time credits be considered at all? These are difficult questions; perhaps questions the answers to which are too arbitrary to be given by courts, which shy away from creating numerically precise rules and thus refuse, for example, to create fixed limitations periods for statutes that do not contain any. Hemmings v. Barian, 822 F.2d 688, 689-90 (7th Cir. 1987); Short v. Belleville Shoe Mfg. Co., 908 F.2d 1385, 1394 (7th Cir. 1990) (concurring opinion). It is a matter for Congress, which passed section 34 in a different era, to address — as it has now done by deleting the requirement of a jury recommendation, but only for the future. We must do our best to give meaning to the original statute in a changed era. The best way, I think, is to direct the sentencing judge when choosing a period of years for a defendant not eligible for a life sentence to select a period that in light of the defendant's fundamental demographic characteristics of age and sex (and in light also of his entitlement to good-time credits if he behaves himself as a prisoner), is significantly, though not necessarily greatly, less severe than a sentence of life imprisonment. The sentencing judge must reflect upon the difference between a life sentence and a sentence to a term of years and state for the record his reason for believing that the term of years that he has imposed is indeed significantly less severe than a term of life. The sentencing guidelines do not specify such a procedure, but neither do they purport to override the statute.

I do not think that the judge should also consider the defendant's health. Apart from the complexities and uncertainties involved in computing life expectancies on the basis of the health of a specific person, a sentence reflects a judgment about the gravity of the defendant's conduct. A person is not less an arsonist for having only six months to live, and a five-month sentence for arson might be thought to depreciate the gravity of his crime. But no one could suppose that a sentence of, say, 40 years for the defendants in this case could be criticized as making light of the seriousness of their crimes.

I have not been able to make up my mind whether the judge should take account of the defendant's race. The use of race as a criterion for official action is highly disfavored; on the other hand the differences in adult life expectancy between blacks and whites in this country are so dramatic that to ignore them in computing a defendant's life expectancy might make it difficult to pick a sentence consistent with section 34. The problem is not important here, because while black and white life expectancies differ greatly, the difference between white and total life expectancies is not great, because blacks comprise a relatively small proportion of the population. These defendants are white, so the use of race-neutral rather than race-specific life expectancy figures would not greatly alter the assessment of what a term of years significantly less than life would be for them.

I do think the maximum good-time credits should be subtracted. The defendant should not be allowed to say, "My sentence is too long because I am planning to be a bad boy in prison and so won't earn the maximum good-time credits."

There is no indication that the district judge was endeavoring to impose a sentence significantly less severe than life in prison, and since our previous opinion did not tell him to do this, no inference that he was endeavoring to do it can be drawn from his silence. The court is therefore right to remand the case for resentencing. The difficulty comes with the directions to the district judge. United States v. Martin, 63 F.3d 1422, 1434 (7th Cir. 1995), an opinion which I joined, holds that "a sentence for a term of years exceeding the defendant's approximate life expectancy would ordinarily constitute an abuse of discretion" under the pre-amended section 34 unless, of course, the jury recommends life imprisonment. In the present case, involving younger defendants, the sentences do not exceed the defendants' life expectancies if, as I believe should be done, maximum good-time credits are subtracted, unless, contrary to my view, individual life expectancies rather than group life expectancies are used and the defendants can show that because of poor health or poor genes or other factors they are not likely to live as long as the average member of their demographic group. Martin does not resolve these issues concerning good-time credits and individual versus group life expectancies (and the bounds of the group), and I interpret Judge Ripple's opinion for the court in the present case to leave them open as well. This is appropriate in light of the difficulty of the issues and the absence of briefing and argument directed to them and a factual record bearing on them.

The passage from Martin that I quoted and that is also quoted in Judge Ripple's opinion states a good rule of thumb, keying the maximum sentence to the defendant's life expectancy. It is not a rigid rule that in all circumstances a sentence exactly equal to the defendant's life expectancy is per se in compliance with section 34, while a sentence that exceeds that life expectancy by even a day is a per se violation of the statute. (Should a 90-year-old commit arson, it might unduly depreciate the gravity of his act to sentence him to a term of years equal to his life expectancy.) What is required is the judge's reasoned choice of a sentence that will fulfill the purposes of section 34 in the post-parole world. I have tried, in amplification of Martin and of the majority opinion in this case, to indicate some of the considerations that bear on that choice.

[39] NORGLE, District Judge, concurring.

I join the opinion of the court and write to emphasize two points: first, to suggest the dangers of giving excessive weight to and the risk of the application of inferences drawn from general statistics to a specific sentencing decision dealing with life expectancy in federal felonies; second, to review the trial court record of the sentencing judge's remarks.

Adding to the bramble bush of Federal Sentencing Guidelines interpretation, trial courts are now faced with determining life expectancy when sentencing in certain cases. Judges must be ever conscious of the risks of interpreting statistical information, especially in the absence of expert testimony. Judges exercising the awesome power of long-term sentencing must not lose sight of the individual standing before the court. Our common experiences in life lead us to express concern about the application of the precise classifications mentioned so eloquently by Judge Posner. Will we, for instance, confront longevity issues regarding smokers versus nonsmokers? Will we consider the longevity of federal inmates versus the general population? How will we deal with the longevity of a defendant not of a particular race but of mixed racial heritage? "Because of the risk that the factor of race may enter the criminal justice process, we have engaged in `unceasing efforts' to eradicate racial prejudice from our criminal justice system." McCleskey v. Kemp, 481 U.S. 279, 107 S.Ct. 1756, 95 L.Ed.2d 262 (1987) (citing Batson v. Kentucky, 476 U.S. 79, 106 S.Ct. 1712, 90 L.Ed.2d 69 (1986)).

Life expectancy computations in multi-defendant cases, where the defendants are of varying ages, genders, and races will require great reflection, careful consideration, and significant fact-finding. Numerous cases in the federal courts involve an international clientele. For instance, sentencing foreign nationals who commit crimes in the United States may be complicated. Life expectancy varies from country to country. As demonstrated in the Central Intelligence Agency's The World Factbook, the life expectancies differ greatly: 1990-1995 Projected Country Average Life Expectancy Brazil 66.2 years India 60.4 years Nigeria 52.5 years Rwanda 46.2 years United States 75.8 years

The United States Central Intelligence Agency, The World Factbook (1994). Individuals being sentenced for federal felonies will risk exposure to a doubled-edged sword of inferences drawn from general statistics. Should the court apply United States life expectancy tables when sentencing a visiting Rwandan who will serve her time in a federal prison? How about a visiting Brazilian national who commits a federal crime? Though not at issue here, it may indeed be difficult for the trial judge to look at several defendants and justify in the record why a person of one age, gender, national origin, or race will receive more or less prison time in light of inferences drawn from general statistics. The trial court's articulation of the reasons for a particular sentence involving life expectancy will be of great importance.

In the case at bar, the defendants are two white male citizens from the same region of the United States. Both received the same sentence. That the sentences are the same, however, does not illustrate the absence of thoughtful imposition of individualized sentences. Indeed, the trial record and the court's articulation at the sentencing hearing reflect otherwise. The trial court rejected the government's position that "the court can sentence Mr. Soy to any number of years, if it pleases, as long as the words L-I-F-E aren't there." (Re-Sentencing, vol. 15, at 2206.) And "the term of 100 years would be appropriate in this particular case so that as to not allow Mr. Soy ever to return to the streets. . . ." (Re-Sentencing, vol. 15, at 2206.)

The trial court's rejection, albeit brief, indicated an awareness of the life expectancy issue. The trial court stated:

And unless my mathematics is off, the number of years that I sentenced the defendant was 53 years. Obviously, it is this Court's feeling that if I give him 100 years that's beyond life, because I don't expect him to live 120 years, unless there's long genes in his family. There's not very many people that go that far, and the likelihood would be very small.

(Re-Sentencing, vol. 15, at 2211.)

Perhaps the trial judge should have articulated in greater detail what he deemed obvious, because to some the obvious was not. As I see the record, the trial judge may well have done what I and my colleagues would direct, that is, to choose a term of years for a defendant not eligible for a life sentence which is significantly, though not necessarily greatly, less severe than a sentence of life imprisonment. When the trial judge said, "I don't expect him to live 120 years, unless there's long genes in his family," (Re-Sentencing, vol. 15, at 2211) he may have satisfied Judge Posner's legitimate concern, in which I join, about sentencing a defendant to a term so long that the defendant is certain to die while incarcerated (barring some medical miracle as yet unforeseen). The sentencing judge engaged in a quick but independent reasoning process beyond intuition. The record does not show that the judge sentenced the defendant to a term that would "do the trick." The undoing here is not the absence of reflection but the brevity of articulation regarding the difference between a life sentence and a sentence for a term of years.

The trial court contemplated the probability of some period of years after the defendants served their time in custody. The judge belabored the record by reading from the printed page all of the conditions of supervised release including reporting to a supervised release officer, working at a lawful occupation, geographical restrictions, submitting written reports within the first five days of each month, and so on. The period of supervised release was five years.

Some might argue that the judge merely met the formalistic requirement of imposing supervised release. Something more from the trial judge on this point might have resolved the issue.

I recognize that life expectancy statistics may serve as a palladium for the statutorily afforded rights of the individual defendants as balanced with the risks to society of early release. I agree with Judge Ripple that on remand the key point is to ensure that the life expectancy of each defendant is appropriately considered by the district court. Aware that Martin stops short of forcing district court judges to use actuarial tables, 63 F.3d at 1434, the trial court in the present case may find it difficult to determine a time span in prison which would not extend beyond life expectancy without reference to such information. The experienced trial judge will also have the thoughtful analysis of Martin to assist in articulating more fully his reasoning regarding life expectancy, thus, leaving less to draw by inference from the cold record, as he attempts to carry out the post-parole legislatively enacted sentencing scheme specifically at issue.