Summary
noting Petitioner's claim that he was actually innocent of the conduct that formed the basis of a sentencing enhancement "raises none of the concerns that prompted the Second Circuit to carve out the [actual innocence] exception to Congressionally imposed limits on habeas jurisdiction"
Summary of this case from Norwood v. WilliamsOpinion
No. 3:17-cv-2059 (MPS)
01-04-2019
ALEX SALVAGNO Petitioner, v. D.K. WILLIAMS, WARDEN Respondent.
RULING ON MOTION TO DISMISS
Alex Salvagno ("Salvagno") is an inmate confined at Federal Correctional Institution ("FCI") Danbury. He filed this petition for a writ of habeas corpus under 28 U.S.C. § 2241 against D.K. Williams ("Williams"), the warden at FCI Danbury, to challenge the twenty-five year sentence imposed after he was found guilty of racketeering conspiracy, violating and conspiring to violate the Clean Air Act ("CAA") and Toxic Substances Control Act ("TSCA"), and filing false personal income tax returns. Salvagno argues that his sentence violates due process because the sentencing court relied on information that years later turned out to be materially untrue, and that this Court should hear his claim - brought after several earlier challenges to his conviction and sentence failed - because he is actually innocent of the conduct that formed the basis of a substantial sentencing enhancement imposed on him under the United States Sentencing Guidelines. (ECF No. 1 at 6; ECF No. 1-1 at 2-31.)
On January 16, 2018, the Court ordered Williams to show why it should not grant the relief sought by Salvagno. (ECF No. 3.) Williams now moves to dismiss Salvagno's petition under Rule 12(b)(1) of the Federal Rules of Civil Procedure for lack of subject matter jurisdiction. (ECF No. 7-1 at 5-13.) For the reasons discussed below, the Court GRANTS Williams' motion to dismiss and DISMISSES Salvagno's petition.
BACKGROUND
The following facts are drawn from Salvagno's petition, Williams' motion to dismiss, and where applicable, court documents in the underlying criminal case against Salvagno.
The Court considered for purposes of this petition the most relevant (in its view) sentencing-related materials available in the underlying case. In particular, the Court reviewed the transcripts of the testimony of one of the government's expert witnesses—which were only recently filed on the public docket (ECF Nos. 1117, 1118, USA v. Salvagno et al., 02-cr-00051-LEK (N.D.N.Y.))—and the transcript of the trial judge's findings and imposition of sentence (ECF No. 402, 02-cr-00051-LEK (N.D.N.Y.)), in addition to relevant parts of the parties' sentencing briefs (ECF Nos. 394, 395, 02-cr-00051-LEK (N.D.N.Y.)).
I. The Trial
Salvagno was the President and principal owner of AAR Contractor, Inc. ("AAR"), an asbestos abatement company operating in New York. (ECF No. 112 at ¶¶ 1, 9, 02-cr-00051-LEK (N.D.N.Y.).) On January 15, 2003, Salvagno, AAR, and other defendants were indicted in the Northern District of New York on charges of racketeering conspiracy, conspiracy to violate the CAA and TSCA, multiple violations of the CAA, and three counts of filing false personal income tax returns. (ECF No. 7-1 at 2; see ECF No. 112, 02-cr-00051-LEK (N.D.N.Y.).) In short, the Indictment alleged that AAR, with Salvagno at its helm, had performed dry removals of asbestos without appropriate precautions or safety gear at dozens of sites and then falsified air monitoring tests to show that the work was lawfully done. (See generally ECF No. 112, 02-cr-00051-LEK (N.D.N.Y.).) Following a four-month trial, the jury found Salvagno guilty on all charges on March 30, 2004. (ECF No. 7-1 at 2.)
Unless a citation to another case is provided, as here, citations to "ECF No. ___" refer to the docket entries in this case, 17-cv-2059 (MPS).
II. The Sentencing Hearing
Following the guilty verdict, the judge held an eleven-day sentencing hearing, taking testimony from fact and expert witnesses called by both parties. Among the many contested issues, Salvagno and the government hotly disputed the applicability of U.S.S.G. § 2Q1.2(b)(2), which provides for a sentencing enhancement of nine levels if the offense "resulted in a substantial likelihood of death or serious bodily injury." Federal Sentencing Guidelines Manual § 2Q1.2(b)(2) (2000). (See ECF Nos. 394, 395, 02-cr-00051-LEK (N.D.N.Y.).)
The sentencing court used the 2000 edition of the Sentencing Guidelines, because, in its view, the edition in effect at the time of sentencing would have resulted in a significantly higher offense level calculation in violation of the prohibition against ex post facto laws. (ECF No. 402, Dec. 23, 2004 Hearing Tr. 155:9-14 (the "Dec. 23, 2004 Hrg."), 02-cr-00051-LEK (N.D.N.Y).) The language of Section 2Q1.2(b)(2) remains today the same as it was in the 2000 Manual.
The Government's primary expert witness concerning the applicability of § 2Q1.2(b)(2) was Dr. Stephen M. Levin, an expert in asbestos exposure. Dr. Levin testified that he reviewed the indictment, trial testimony, and numerous affidavits, and spoke with three former AAR workers to gain familiarity with the facts on which to base his opinions. (See also ECF No. 1117, Oct. 27, 2004 Hrg. at Tr. 15-17 (the "Oct. 27, 2004 Hrg."), 02-cr-00051-LEK No. 394 (N.D.N.Y.).) Based on his review, Dr. Levin opined that he believed that there was a substantial likelihood of death or serious bodily injury as defined in § 2Q1.2(b)(2), because "there is very significant likelihood that individuals will contract cancer, which I consider a grave threat to human health, as well as severe and impairing asbestosis as a consequence of the exposures that have been described to me . . . ." (Id. Tr. 62:19-24.) In particular, Dr. Levin estimated that a core group of 100 employees who had worked for AAR for 4 years or more without respiratory protection were exposed to asbestos in the range of 160 "fiber-years." (Id. Tr. 63:19-23.) As Dr. Levin explained, a fiber-year is the average number of asbestos fibers per cubic centimeter multiplied by the number of years a person is exposed to that average concentration. (See id. Tr. 30:9-13 ("[F]iber years really means being exposed to a certain concentration of asbestos in the air for a certain period of time because it's accumulated exposure that determines the risk of disease, not just the exposure at any one time.").)
The Government also called one other expert, Dr. Michael Lax, as a rebuttal witness to Salvagno's two experts. (ECF No. 394-2 at 2, 02-cr-00051-LEK (N.D.N.Y.).)
Dr. Levin testified that his estimate of a 160 fiber-year concentration was based on his review of the trial transcript, affidavits, and his discussions with AAR workers. (See, e.g., id. Tr. 56:22 -57:7.) In particular, Dr. Levin testified that at AAR, "for the most part the removal of asbestos was done under dry conditions, as opposed to wet, which would suppress dust and that the removal was carried out at a very rapid pace which resulted in almost constant presence of visible dust in the air, at times described as so heavy as it looked like a snowstorm or a blizzard, and this was a recurrent thing depending on the pace of the work." (Id. Tr. 55:3-10.) He similarly testified that, based on one former employee's description, AAR's removal of asbestos from pipe fittings and elbows led to the "constant presence of visible dust" in the air, due in part to the lack of ventilation:
[The individual] would use a knife, a hammer, anything to pound the material in dry fashion to remove it from the surface that it was adhering to, creating dust in his breathing zone that was very visible and the same individual described frequently performing these operations in enclosed areas where the door would be locked, no ventilation would be present and where there was the constant presence of visible dust that could be seen in the air. Even under relatively low lighting one would see the mist of drifting asbestos fiber flinting in the air across the light source.(Id. Tr. 57:16-25.) Comparing this information to published measurements regarding the air concentrations attending particular abatement activities of the type that AAR's employees engaged in (e.g. "dry rip[ping]" and sweeping and bagging of asbestos debris), Dr. Levin constructed a "range of likely fiber concentrations that would have been present had real measurements been taken during the time that these operations were carried out in this case" and adopted the lowest estimate in the range of 160 asbestos fibers per cubic centimeter. (Id. Tr. 58:1-59:8; Tr. 63:3-8.). Dr. Levin then estimated that the 100 workers in the "core group" were exposed to that air concentration 25% of the time (despite evidence suggesting a greater percentage), resulting in a total average exposure of 40 fibers per cubic centimeter, which he characterized as a "really conservative estimate of their real exposure." (Oct. 27, 2004 Hrg. Tr. 63:8-23.) Multiplying this figure by the four years worked by these 100 employees, Dr. Levin concluded that the average asbestos exposure for each was 160 fiber-years. (Id. Tr. 63:19-23.)
Dr. Levin also testified that based on his review of affidavits and conversations with the AAR employees, he was aware that AAR did not systematically employ negative pressure at its work sites and that any containment efforts that were employed were ineffective. (Oct. 27, 2004 Hrg. Tr. 60:14-61:19.)
Dr. Levin erroneously stated on direct examination that the concentration was 165 fibers per cubic centimeter, but on cross-examination he clarified that the concentration was 160 fibers. (Nov. 3, 2004 Hrg. Tr. 61:14-62:17.)
For example, Dr. Levin added that "at any one time no more than 50 percent of the workers were wearing respiratory protection and at times many fewer than that." (Id. Tr. 55:10-12.)
To calculate how many of the workers would likely suffer injury at this exposure level, Dr. Levin applied the "OSHA risk assessment model," which "was based on the observation of many working populations, and their disease and death experience." (Id. Tr. 63:23-64:3.) Based on that risk assessment model, Dr. Levin concluded that among the group of 100 core AAR workers, he anticipated 20 cases of asbestos-related cancers and 9 of severe and impairing asbestosis. (Id. Tr. 64:3-7.) He also noted that AAR employees who worked for less than four years (and thus were excluded from the group of 100) were also at a substantial risk of contracting these same diseases, though in smaller numbers due to their shorter exposure period. (Id. Tr. 64:8-15.)
Dr. Levin further testified that he would be "surprised" if none of the 100 workers developed asbestos-related illness after 15 years or more. (Id. Tr. 70:19-71:7 ("I would expect a great majority of workers, if we looked at their X-rays 20 years down the road, we are going to see a clear area of asbestos-related scarring on their films. . . . [I]f we look only at that group of hundred workers who worked without respiratory protection for over four years, I would be very surprised to find a single X-ray that looked normal if we followed them 20 years."); see also id. Tr. 66:24-67:4 (same for workers who worked longer than 4 years with AAR).) In sum, Dr. Levin supported his conclusion that Salvagno's conduct had caused substantial likelihood of death or serious bodily injury as defined in § 2Q1.2(b)(2) with a risk assessment model that estimated that 29 of the 100 longest-exposed AAR workers would contract cancer or asbestosis over a 15- to 20-year time frame, based on an estimated average fiber count for these workers of 160 fiber-years. (See ECF No. 1-1 at 15.)
On cross-examination, Dr. Levin acknowledged that if his estimate of 160 fiber-years was incorrect, it would affect the number of individuals he anticipated would suffer serious illness. (ECF No. 1118, November 3, 2004 Hearing Tr. 62:7-13 (the "Nov. 3, 2004 Hrg."), 02-cr-00051-LEK No. 394 (N.D.N.Y.).) Nonetheless, Dr. Levin stated that while there was some uncertainty in the precise number of individuals that would be affected, he was "entirely comfortable" in saying that serious disease would result from the exposure levels described to him:
Q Okay. Now, you would agree, wouldn't you, that if the 160 number that you started with is wrong, your analysis would have a problem, right?
A If the 160 fiber years turned out not to be the dose, whatever the dose ultimately turned out to be, either greater or lesser, would influence the number of people that anticipate would suffer serious illness.
Q And what you did to come out with 160 was you took -- you made an estimate based on some studies, correct?
A Measurements taken during this operation in a number of settings, yes.
[. . . ]
Q And would you agree that making that kind of exposure assessment based on the information you had is bad science?(Nov. 3, 2004 Hrg. Tr. 62:7-64:1.) In other words, Dr. Levin believed that even if his 160-fiber year count was wrong, "given the circumstances I have had described to me in this case with the levels of exposure that have been described," some amount of asbestos-related disease would occur among the workers over a 20- to 25-year timeframe.
A No, I wouldn't call it bad science. It's the best we can do in the absence of actual measurements. And as I indicated, I would much rather have data that were obtained during the actual processes of removal for this group of 4 workers, but those data were not available.
Q Well, is there a point that as a doctor you would just say that you simply don't have enough data to predict illness?
A Absolutely not. You know, given the circumstances I have had described to me in this case with the levels of exposure that have been described, I feel entirely comfortable in saying the exposure levels were sufficient to produce human disease 20 and 25 years down the road, and while I can't precisely quantify it, I have no doubt that many workers will suffer scarring lung disease and unfortunately a number will also develop cancers. My estimate was based on OSHA's risk assessment with my extrapolation from work process measurements done elsewhere and taking a conservative look at that, my understanding is that the level that I described here may well be low given the amount of dry brooming and sweeping that was a continual activity during the course of the larger project, not just at the end of the day, so resuspension of this material ongoing during the day apparently characterized lots of these jobs, and I would give somewhat higher estimate than I offered had I been aware of how often the dry sweeping and cleaning was going on during the day of this project.
As Salvagno recites in his petition, he called two experts at sentencing: Robert Sheriff, a certified industrial hygienist, and Dr. Peter Barrett, a diagnostic radiologist. (ECF No. 1-1 at 18, 20.) Mr. Sheriff testified that the actual fiber-year count for the "core group" would be less than 1.6 fiber-years, which (even assuming that they did not wear respirators at all over four years) would result in "less than one" of those individuals developing asbestos-related disease. (ECF No. 1-1 at 19; ECF No. 1-4 at 11, 13.) Mr. Sherriff noted that this level of asbestos exposure was permitted by U.S. Department of Labor regulations. (ECF No. 1-1 at 20; ECF No. 1-4 at 15-17.) In Dr. Barrett's testimony, he criticized Dr. Levin's reliance on what he described as inapplicable literature and concluded, based on comparable studies, that the fiber-year count for Salvagno's workers was actually closer to 1 or 1.5 fibers in a given year. (ECF No. 1-1 at 21-22; ECF No. 1-5 at 12-13, 14-15.) In addition, Salvagno presented the written testimony of Dr. William Rogers, a physician in occupational medicine, who examined Salvagno's workers annually between 1990 and 1999, including by taking chest X-rays. (ECF No. 1-1 at 23.) According to Salvagno, Dr. Rogers concluded that there was no difference between Salvagno's workers and other asbestos workers he had examined and that his examination findings were "not consistent with individuals working in heavy concentrations of dust without the use of any respirators." (Id.)
The Court here relies primarily on Salvagno's description of his experts' testimony and written affidavits in his petition. (ECF No. 1-1 at 18-23.) Salvagno's petition also attached short excerpts of the testimony of Mr. Sherriff and Dr. Barrett, which the Court has reviewed as well. (ECF Nos. 1-4, 1-5.)
The record materials available to me do not disclose whether the sentencing judge made a specific determination on the record as to whether U.S.S.G. § 2Q1.2(b)(2) applied. Nonetheless, the sentencing judge adopted the enhancements set forth in the PSR, which according to Salvagno included the § 2Q1.2(b)(2) enhancement and which the sentencing judge concluded "were proven at trial and during the comprehensive sentencing hearing," resulting in a total offense level of 40. (Dec. 23, 2004 Hrg. Tr. 156:4-13.) On December 23, 2004, the judge sentenced Salvagno to 20 years on the racketeering count and 5 years on the remaining counts, to run concurrent to each other but consecutive to the 20 years for the racketeering count, for a total 300 of months imprisonment. (Id. Tr. 156:13-22.)
According to Salvagno's sentencing brief, the offense level in the PSR incorporated two charged groups of 34 and 39, the latter of which included the nine-level § 2Q1.2(b)(2) enhancement. (ECF No. 395 at 5-7, 02-cr-00051-LEK (N.D.N.Y.).) The Court was unable to review the PSR itself.
III. Post-Sentencing Proceedings
On August 30, 2006, upon remand in light of United States v. Booker, 543 U.S. 220 (2005) and United States v. Crosby, 397 F.3d 103 (2d Cir. 2005), the district court resentenced Salvagno to the same custodial sentence. (ECF No. 7-1 at 2-3.) Salvagno filed an appeal in which he challenged, among other alleged errors, the district court's conclusion that U.S.S.G. § 2Q1.2(b)(2) applied. See Appellant's Br., United States v. Salvagno, 2008 WL 6153930, at *117-27 (2d. Cir. January 29, 2008) (sufficiency of the evidence challenge to § 2Q1.2(b)(2) enhancement and observing that "[i]f it turns out that Levin was wrong, then there may very well be no AAR workers who have developed asbestos-related disease; Alex Salvagno, however will still be in prison."). The Second Circuit affirmed his conviction. (ECF No. 1 at 2.) Although the Court of Appeals' summary order did not specifically address his challenge to the § 2Q1.2(b)(2) enhancement, it rejected Salvagno's "other challenges to his conviction and to his sentence," which included that challenge. United States v. Salvagno, 343 F. App'x 702, 705 (2d Cir. 2009).
Salvagno filed a motion to vacate the conviction under 28 U.S.C. § 2255 on July 7, 2010, and a substituted motion on December 10, 2010, which was denied. (ECF No. 7-1 at 3.) Both the district court and Second Circuit denied Salvagno's requests to certify an appeal. (Id.) On December 12, 2016, the Second Circuit refused to certify Salvagno's successive § 2255 motion. (Id.) On February 22, 2017, Salvagno filed with this Court a petition for writ of habeas corpus under 28 U.S.C. § 2441 seeking reconsideration of a denial by the Bureau of Prisons of a request for compassionate release pursuant to 18 U.S.C. § 3582(c)(1)(A)(i). (Id.) This Court dismissed that petition on November 7, 2017. See Salvagno v. Dir., Bureau of Prisons, No. 3:17CV318 (MPS), 2017 WL 5159214, at *1 (D. Conn. Nov. 7, 2017). Appeal of that decision is currently pending. Salvagno v. Director, No. 17-3997 (2d. Cir. 2017).
After Salvagno's appeal was filed, Congress passed the First Step Act, which amended 18 U.S.C. § 3582(c)(1)(A) to allow prisoners to file a motion for compassionate release in the sentencing court after exhausting all administrative appeals of a refusal by the Bureau of Prisons to bring such a motion on the prisoner's behalf. See First Step Act of 2018, P.L. 115-___, § 603(b)(1).
IV. This Action
On December 11, 2017, Salvagno filed the § 2241 petition in this case. (ECF No. 1.) Salvagno's petition primarily claims that his twenty-five year sentence violated due process because the sentencing court relied on information that years later turned out to be wrong—specifically, Dr. Levin's estimate that 100 of Salvagno's workers who worked for four years or more were exposed to 160 fiber-years was inaccurate. (See ECF No. 1-1 at 15.) Although he acknowledges that a Section 2241 petition may not ordinarily be used to attack a sentence, Salvagno argues that this Court should reach this claim on the merits because he asserts "actual innocence" of conduct on which his sentence was based, namely the § 2Q1.2(b)(2) enhancement for conduct resulting in a "substantial likelihood of death or serious bodily injury." (ECF No. 1-1 at 9.) His petition attaches an April 18, 2017 declaration by Christopher Knauf, who avers he was the insurance agent for AAR between 1998 and 2001. (ECF No. 1-2, Exhibit 1, Declaration of Christopher Knauf ("Knauf Decl.") at ¶¶ 1-2).) Knauf's declaration asserts that he obtained a loss run report from the New York State Insurance Fund for the period 1990 to 1998, which maintained AAR's workers compensation policy. (Id. at ¶¶ 5-6.) He avers that the report, which is attached to his declaration, "shows that there are no known asbestos related workers compensation claims for that period," i.e. the period from 1990 to 1998 (id. at ¶ 5); he goes on to aver that "I can confirm that for the period 1990 to date [i.e. April 18, 2017, the date of the affidavit], there have been no known asbestos related workers compensation claims." (Id. at ¶ 6.) Knauf's declaration attaches what appears to be a New York State Insurance Fund loss run report for AAR for accidents that occurred between 9/22/1990 and 5/23/1998, which lists 18 claims but does not disclose the nature of those claims. (ECF No. 1-2 at 4.) Salvagno argues that the Knauf declaration and attached documents show that "in the 27 years since the . . . conduct commenced in 1990 (beyond the 15-20 year latency period for asbestos), there have been no reported asbestos injuries actually attributable to [his] offense." (ECF No. 1-1 at 8-9.) Accordingly, Salvagno concludes that he has shown by clear and convincing evidence that Dr. Levin's 160 fiber-count estimate and risk analysis were wrong, and thus that he is innocent of the conduct on which the U.S.S.G. § 2Q1.2(b)(2) enhancement was based. (ECF No. 1-1 at 9.)
On January 16, 2018, the Court ordered Williams to respond to the petition. (ECF No. 3.) On March 12, 2018, Williams moved to dismiss the petition for lack of subject matter jurisdiction. (ECF No. 7.) Salvagno filed a response on April 10, 2018, and Williams filed a reply on April 24, 2018. (ECF Nos. 11, 13.) Salvagno then moved for leave to file a sur-reply, attaching his brief to his motion. (ECF No. 14.) The Court GRANTS that motion and will consider Salvagno's sur-reply.
Salvagno also filed a motion for appointment of counsel on April 12, 2018, following Williams' reply. (ECF No. 12.) In light of my determination that the Court does not have subject-matter jurisdiction, this motion is DENIED as moot. In any event, because no hearing is required to decide his petition, the Court would deny the motion on the merits, as Salvagno has had no difficulty responding to Williams' arguments or presenting his case to the Court. See, e.g., Hodge v. Police Officers, 802 F.2d 58, 60 (2d Cir. 1986) ("In habeas corpus cases, counsel must be appointed for qualified indigents when a hearing is required; the court may appoint counsel at an earlier stage if it deems appointment desirable.").
LEGAL STANDARDS
A "case is properly dismissed for lack of subject matter jurisdiction under Rule 12(b)(1) when the district court lacks the statutory or constitutional power to adjudicate it." Makarova v. United States, 201 F.3d 110, 113 (2d Cir. 2000). The "plaintiff asserting subject matter jurisdiction has the burden of proving by a preponderance of the evidence that it exists." Id. "In resolving a motion to dismiss for lack of subject matter jurisdiction under Rule 12(b)(1), a district court . . . may refer to evidence outside the pleadings." Id. The court construes Salvagno's pro se petition liberally and accepts all factual allegations as true. See Ford v. D.C. 37 Union Local 1549, 579 F.3d 187, 188 (2d Cir. 2009).
Salvagno seeks to bring his claim under 28 U.S.C. § 2241, which the Second Circuit has treated as a jurisdiction-limiting statute. See Cephas v. Nash, 328 F.3d 98, 103 (2d Cir. 2003) (affirming dismissal of claim under 28 U.S.C. § 2241 for lack of subject matter jurisdiction). Under 28 U.S.C. § 2241, "the federal courts have the power to grant writs of habeas corpus . . . [o]n behalf of . . . prisoners who are in custody in violation of the Constitution or laws or treaties of the United States." Poindexter v. Nash, 333 F.3d 372, 377 (2d Cir. 2003) (citation omitted; internal quotation marks omitted). By contrast, under 28 U.S.C. § 2255, "a federal prisoner may seek release upon the ground that the sentence was imposed in violation of the Constitution or laws of the United States . . . ." Id. (internal quotation marks omitted). "Despite their overlapping language and the generality of § 2241, the two sections have been construed as address[ing] different types of claims," with § 2241 applying to the "execution" of a prisoner's sentence and § 2255 applying to the "imposition" of a prisoner's sentence. See id. (alteration in original; internal quotation marks omitted).
Nevertheless, "[i]n some very limited circumstances, claims that fall within the substantive scope of § 2255 may properly be made in a petition filed under § 2241." Id. at 378. Under the so-called "Triestman" exception, "[w]here relief through § 2255 is unavailable as a procedural matter, and the failure to allow for collateral review would raise serious constitutional questions," the Second Circuit has "interpreted § 2255's exception for 'inadequate or ineffective' remedy as making an exception to the general rule that a federal prisoner must use § 2255 [to challenge a conviction or sentence] instead of seeking a writ of habeas corpus under § 2241." Id. (internal citation and quotation marks omitted); see generally Triestman v. United States, 124 F.3d 361 (2d Cir. 1997). To fit within this exception, the application must show that the relief sought is procedurally unavailable under § 2255 and assert a claim of actual innocence "that (a) is prov[able] . . . on the existing record, and (b) could not have effectively [been] raised . . . at an earlier time." Poindexter, 333 F.3d at 378. Salvagno asserts that his § 2241 petition fits within the Triestman exception. (ECF No. 1-1 at 4, 7.)
DISCUSSION
Construed liberally, Salvagno's petition makes essentially a three-step argument that he is actually innocent within the meaning of the Triestman exception. He first argues that the Knauf declaration shows "that in the 27 years since the offense conduct commenced in 1990 (beyond the 15-20 year latency period for asbestos), there have been no reported asbestos injuries actually attributable to Salvagno's offense." (ECF No. 1-1 at 8-9.) Next, he argues that this "clearly and convincingly shows that the 160 fiber count and corresponding risk analysis, including the conclusion that 29 workers would have developed asbestos disease or died by this time . . . were materially inaccurate." (Id. at 9.) Because the conclusion that 29 workers would have developed asbestos-related disease or died "in turn formed the basis of the nine-level [§ 2Q1.2(b)(2)] enhancement," Salvagno claims that the sentencing judge imposed the enhancement based on conduct of which Salvagno is actually innocent. (See id. at 10.)
Accordingly, Salvagno contends that this new evidence shows that he is actually innocent of the U.S.S.G. § 2Q1.2(b)(2) sentencing enhancement, that his claim is provable on the existing record, and that it could not have effectively been raised at an earlier time. (See, e.g., ECF No. 1-1 at 5-6; ECF No. 11 at 10-22.) I disagree because Salvagno's claim is based on a much broader concept of "actual innocence" than any the Supreme Court or the Second Circuit has endorsed, and recognizing such a claim would ignore the restrictions imposed on post-conviction challenges to applications of the Sentencing Guidelines. Further, Salvagno's claim is not provable on the existing record because, even if his new evidence proves that the Government's expert's estimates were wrong, there was other evidence that adequately supported application of the risk enhancement in this case.
Salvagno argues in his brief that the same evidence shows he is actually innocent of conduct which enhanced his sentence under § 3553(a) as well. I assume for purposes of this ruling that the § 2Q1.2(b)(2) enhancement, even after the Booker remand, affected Salvagno's sentence.
I. Salvagno's Claim Does Not Fit Within the Narrow Doctrine of Actual Innocence
A. The Scope of the Triestman Exception
The "actual innocence" doctrine recognized in certain habeas contexts is "narrow" and typically "concerned with actual as compared to legal innocence." Sawyer v. Whitley, 505 U.S. 333, 339-40 (1992) (citation omitted); see also Darby v. United States, 508 F. App'x 69, 71 (2d Cir. 2013) (calling the doctrine "very narrow"). "In the context of a noncapital case, the concept of 'actual innocence' is easy to grasp . . . because it normally means simply that the defendant did not commit the crime." Poindexter, 333 F.3d at 381 (internal citations omitted).
This was the type of actual innocence claim involved in Triestman, where Mr. Triestman had pled guilty to two counts of drug conspiracy and one count of "using" a firearm during and in relation to a drug crime, in violation of 18 U.S.C. § 924(c). 124 F.3d at 363. After his direct appeal and first motion under 28 U.S.C. § 2255 had failed at the Second Circuit, the Supreme Court adopted a new interpretation of § 924(c) in Bailey v. United States, 516 U.S. 137 (1995), such that the statute no longer covered the conduct to which Triestman had pled guilty. See id. at 364-66. Triestman thus filed another § 2255 petition asserting that, under Bailey, he was actually innocent of the Section 924(c) crime. Id. at 365.
In the interim between his petitions, however, Congress adopted the Anti-Terrorism and Effective Death Penalty Act of 1996 ("AEDPA"), which added new restrictions to § 2255. One new requirement was that the Court of Appeals certify that any successive § 2255 motion satisfied one of two narrow prerequisites before a district court could consider it. 124 F.3d at 365. The Second Circuit found that Triestman's actual innocence argument met neither of the two threshold requirements for certification under § 2255. Id. at 371-72. As the AEDPA amendments thus appeared to foreclose judicial review of Triestman's newly available claim of innocence, the Court of Appeals considered whether Triestman could instead seek a traditional habeas remedy under 28 U.S.C. § 2241. Id. at 373. 28 U.S.C. § 2255 contains a savings clause permitting a prisoner to seek habeas corpus relief under § 2241 where it "appears that the remedy by [§ 2255] motion is inadequate or ineffective to test the legality of his detention." 28 U.S.C. § 2255(e). The Second Circuit interpreted this provision to require a district court to allow a petitioner to raise claims under § 2241 that ordinarily must be brought under § 2255 where failing to do so would raise serious questions about the constitutionality of § 2255, and it found that Triestman's claim fit this situation. See Triestman, 124 F.3d at 377, 380 (holding that because "Triestman cannot bring his claim under the newly-amended § 2255" and "preclud[ing] all collateral review in a situation like this would raise serious questions as to the constitutional validity of the AEDPA's amendments to § 2255," "Triestman is entitled to bring a petition for a writ of habeas corpus in the district court pursuant to § 2241(c)(3)"); id. at 378-79 ("[S]erious Eighth Amendment and due process questions would arise with respect to the AEDPA if we were to conclude that, by amending § 2255, Congress had denied Triestman the right to collateral review in this case."). The Second Circuit described its reasoning as a way of threading a needle between the AEDPA's goal of limiting habeas petitions and the principle that courts must, when a statute is susceptible to two or more readings, select the one that avoids constitutional problems. Id. at 377-78. Accordingly, because § 2255 was not available to Triestman, he nonetheless could bring his Bailey claim of actual innocence under § 2241. Id. at 380.
Triestman's claim of actual innocence, however, bears little resemblance to Salvagno's. Triestman claimed that he was actually innocent of the Section 924(c) offense to which he had pled guilty, because he had engaged in conduct that, in the eyes of the Supreme Court, fell outside the scope of Section 924(c). Id. at 364-66 (citing Bailey, 516 U.S. 137). By contrast, Salvagno "does not claim innocence of the RICO, Clean Air Act (CAA) & Tax crimes for which he was convicted," and nowhere suggests that his conduct was outside the scope of the underlying statutes. (ECF No. 1-1 at 9.) Rather, he seeks to stretch the doctrine of actual innocence to cover a sentencing enhancement for the likely effects of his conduct. But the Second Circuit has never extended the Triestman exception beyond a claim of actual innocence of an offense of conviction. Indeed, the unique historical pedigree of such a claim was central to its holding in Triestman. The court found that denying a remedy to a person who claimed he was "innocent of the crime of which he stands convicted" would raise "an open and significant due process question," because procedural limitations may violate due process when they "offend[] some principle of justice so rooted in the traditions and conscience of our people as to be ranked as fundamental," and "concern about the injustice that results from the conviction of an innocent person has long been at the core of our criminal justice system." 124 F.3d. at 379 (emphasis added; citations omitted). Here, Salvagno's claim that he is "actually innocent" of the § 2Q1.2(b)(2) enhancement—in essence, that new evidence shows that a Sentencing Guideline was misapplied—draws upon no such "traditions" and is far removed from a "concern . . . [that] has long been at the core of our criminal justice system." Id. In short, Salvagno's claim raises none of the concerns that prompted the Second Circuit to carve out the exception to Congressionally imposed limits on habeas jurisdiction in Triestman.
B. The Extension of Actual Innocence in Spence
Salvagno points out that the Supreme Court in other habeas contexts has extended the concept of "actual innocence" to capital sentencing, see Sawyer, 505 U.S. at 350, and that the Second Circuit has endorsed a further extension of the concept to factual innocence of a crime used as a sentencing enhancement in a non-capital case, Spence v. Superintendent, Great Meadow Corr. Facility, 219 F.3d 162 (2d Cir. 2000). In Spence, a petitioner who pled guilty in state court to a robbery was given an increased sentence by the trial judge after being arrested for a second robbery for which he was later acquitted. 219 F.3d at 165. In his subsequent habeas petition, the Second Circuit held that although Spence had procedurally defaulted his due process claim, such default could be excused because Spence "challeng[ed] the determination of his responsibility for the act predicating his enhanced sentence" - whether he in fact committed the second robbery. Id. at 171-72. Salvagno argues that, like Spence, he was eligible for the same sentence irrespective of the enhancement but had his sentence substantially increased on the basis of conduct of which he now claims actual, factual innocence, specifically, the 160 fiber-year count relied on by Dr. Levin. (ECF No. 11 at 10-11; ECF No. 14 at 6-7.)
Nothing in Triestman suggests that I am required, in resolving Salvagno's petition under § 2241, to follow Spence, which arose in a different procedural context: a § 2254 petition in which the defendant had procedurally defaulted his constitutional claim. Id. at 170. As the Second Circuit acknowledged in Spence, "the doctrine of procedural default is based on considerations of comity and finality, and not on a jurisdictional limitation on the power of a federal court under 28 U.S.C. § 2254 to look beyond a state procedural default and consider the merits of a defaulted claim that asserts a constitutional violation." Id. The Second Circuit thus relied on one of the two judge-made exceptions to a procedural default, "whether failure to consider the federal claim will result in a 'fundamental miscarriage of justice' or, in other words, an unjust incarceration," to excuse the procedural default in Spence's case. Id. at 170, 172 (citing, inter alia, Murray v. Carrier, 477 U.S. 478, 495-96 (1986)) (other internal citations omitted). Spence thus does not implicate the Congressionally imposed jurisdictional limits on successive habeas petitions addressed in Triestman.
Even if the narrow concept of actual innocence of an offense of conviction involved in Triestman should be expanded to allow a claim of "sentencing innocence" of the kind raised in Spence to proceed under § 2241, this would still not widen the § 2241 gateway enough to let in Salvagno's "actual innocence" claim. The "sentencing innocence" claim pressed by Salvagno is broader than the one recognized in Spence, which should itself give a district judge pause because of the Supreme Court's and Second Circuit's admonitions that the "actual innocence" exception in any context is "very narrow." See Sawyer, 505 U.S. at 341; Darby, 508 F. App'x at 71. In Spence, the defendant was actually innocent of another offense that, prior to his subsequent acquittal, the trial court used as a basis to enhance his sentence (or, more precisely, as a basis to find that he had forfeited his eligibility for a more lenient sentence, which the trial court had conditioned on his not engaging in other misconduct before sentencing). 219 F.3d at 166. Salvagno, by contrast, admits that he is guilty of all fourteen crimes of which he was convicted, and he does not suggest that the Guidelines enhancement under § 2Q1.2(b)(2) hinges on a prior conviction. (ECF No. 1-1 at 9.)
Salvagno's claim is also missing another ingredient from the Spence recipe. Spence appears to impose a requirement that there be a constitutional violation tied to the claim of "sentencing innocence," one that caused an error in the determination of the sentence. Relying on Smith v. Murray, 477 U.S. 527 (1986), which discusses "actual innocence" in the capital sentencing context, the Spence court stated that, "[w]hen transporting the concept of 'actual innocence' to the sentencing phase of capital trials, the Supreme Court has required that the legal error the defendant raises (his constitutional claim) must have caused an actual error in determining guilt or eligibility for sentence, by precluding the development of true facts or resulting in the admission of false ones. In so ruling, the Court has made clear that the availability of actual innocence exception depends not on the nature of the penalty the state imposes but on whether the constitutional error undermined the accuracy of the guilt or sentencing determination." 219 F.3d at 170-71 (internal quotation marks and citations omitted). The court concluded that Spence's petition did raise such a constitutional error, specifically that the "trial court did not ask the right question [at sentencing]—whether Spence more likely than not committed the second criminal act," of which he was subsequently acquitted at a state court trial. Id. at 171. Unlike Spence, Salvagno fails to articulate an underlying constitutional violation that goes to the accuracy of the sentencing determination in his case. He claims only that later events - namely, that no claims have been reported after (at most) 27 years - show that the sentencing judge's prediction that there was a "substantial likelihood of death or serious bodily injury" in the future turned out to be wrong. Right or wrong, however, that prediction was supported by evidence presented at the sentencing hearing, evidence that Salvagno took full opportunity to challenge and rebut with evidence of his own. Salvagno fails to point to any constitutional defect in the sentencing proceeding, let alone one that "preclud[ed] the development of true facts or result[ed] in the admission of false ones." Id. at 170 (internal modifications omitted).
Even if the judge's prediction, and Dr. Levin's estimate (to the extent the judge relied on it) turned out to be wrong, Salvagno has not shown that any "false [facts]," id. at 170, were considered by the sentencing judge. (See ECF No. 11 at 4-5.) In a variety of contexts, courts in the Second Circuit have defined "false" as referring to statements that are demonstrably untrue when made. See, e.g., United States v. Lighte, 782 F.2d 367, 373 (2d Cir. 1986) (in perjury cases, jury must be instructed that "an answer is knowingly false only if it was untrue when made and known to be untrue by the individual making it") (internal quotation marks omitted); Morse v. Fusto, 804 F.3d 538, 549 (2d Cir. 2015) (in civil case alleging fabrication of evidence, approving jury instruction that "'false' is defined as 'untrue when made and ... known to be untrue when made by the person making it or causing it to be made'"); United States v. Hall, No. 2:10-CR-154, 2011 WL 6291957, at *6 (D. Vt. Dec. 15, 2011) (under false statements statute, "[a] statement is false or fictitious if 'untrue when made, and known at the time to be untrue by the person making it.'" (quoting Leonard B. Sand et al., Modern Federal Jury Instructions (2011)). That definition generally does not include predictions based on available information not known to be untrue, because no one has a crystal ball, as Salvagno's lawyer told the sentencing judge. (ECF No. 1-1 at 26, ECF No. 1-6 at 1 ("[The prosecutor] paints this case as the equivalent of a murder case. In truth, no one has died, no one has become ill, and despite the crystal ball wielded by the prosecutor's 'experts,' there is no hard evidence that anyone ever will become ill or die as a result of defendants' activities.").) Not surprisingly, then, courts have generally refused to apply the label "false" to estimates and predictions at all, absent evidence that the speaker did not believe them, because they are characterized more aptly as opinions than facts. See, e.g., In re Int'l Bus. Machines Corp. Sec. Litig., 163 F.3d 102, 107 (2d Cir. 1998) ("The statements challenged by plaintiffs are expressions of optimism or projections about the future. Each statement plaintiffs challenge in this action concerns an uncertain future event—the payment of dividends. . . . [T]he challenged statements regarding the future payment of dividends were predictions or opinions, and not guarantees" and thus were not actionable under federal securities laws); DeSalle v. A.G. Edwards & Sons, Inc., 804 F. Supp. 436, 438 (D. Conn. 1992) ("Statements of predictions or opinions are only actionable if the speaker disseminated the forecasts knowing that they were false or that the method of preparation was so egregious as to render their dissemination reckless.") (internal quotation marks and modifications omitted); Burton v. Iyogi, Inc., No. 13-CV-6926 DAB, 2015 WL 4385665, at *8 (S.D.N.Y. Mar. 16, 2015) (in fraudulent inducement case, noting that "[a]lthough opinions are generally not actionable in fraud, the expression of an opinion or prediction which the declarant does not himself believe is a false statement of fact.") (internal quotation marks and citation omitted). And Salvagno nowhere suggests that Dr. Levin or the Government did not believe Levin's estimate when it was presented.
As these cases suggest, "falsity" often also includes a requirement that the speaker or writer know a statement is untrue, but I do not rely on the knowledge component here. As Salvagno points out, if a sentencing court relies on "materially untrue" misinformation or "material false assumptions as to any facts relevant to sentencing," King v. Hoke, 825 F.2d 720, 724 (2d Cir. 1987), that may be sufficient to show a due process violation even if the information was not known to be false at the time of sentencing. Id.
Salvagno's attempt to limit his claim of a factual error to the 160 fiber-year estimate - rather than Dr. Levin's ultimate predictions of how many AAR workers would develop asbestos-related disease or die - does not change this analysis. (ECF No. 11 at 9.) He points to no evidence available at the time of the sentencing proceedings suggesting that that the estimate of 160 fiber-years was demonstrably wrong. To the contrary, his petition acknowledges that "the factual predicate of his actual innocence claim was 'inherently unknowable', not capable of being discovered, and in fact non-existent, until after the 15-20 year latency period for asbestos related injury came to completion." (ECF No. 1-1 at 10-11.) In addition, Dr. Levin himself made clear that the 160 fiber-year count was a rough estimate. (Oct. 27, 2004 Hrg. Tr. 109:15-17 ("I would like better data if it were available. My understanding is that no measurements that were truly accurate were taken during this work.").) The only evidence Salvagno cites that was available to the sentencing judge is his own experts' opinions at sentencing. (ECF No. 1-1 at 18-21.) A disagreement between competing experts, however, is not the stuff of "actual innocence." Cf. Gimenez v. Ochoa, 821 F.3d 1136, 1142, 1145 (9th Cir.) (finding that petitioner's new expert testimony showed "simply a difference in opinion—not false testimony" that would rise to a due process violation, and rejecting actual innocence claim based on the other evidence in the case), cert. denied, 137 S. Ct. 503 (2016). Accordingly, Salvagno has not shown that the sentencing judge relied on "false" information when he made his determination under U.S.S.G. § 2Q1.2(b)(2), and thus has not raised any cognizable claim that a constitutional defect infected his sentencing.
C. Post-Spence Treatment of Actual Innocence
I have found no Supreme Court or Second Circuit precedent that has broadened the "actual innocence" gateway in habeas cases to admit a claim that challenges only the application of a Guidelines enhancement not itself involving innocence of a prior offense, as Salvagno's claim does. If anything, Second Circuit cases postdating Triestman and Spence have continued to emphasize the narrow scope of "actual innocence."
In Walden v. United States, the Second Circuit in an unpublished decision assumed—but expressly did not decide—that a Section 2255 challenge based on Apprendi v. New Jersey, 530 U.S. 466 (2000), to the quantity of cocaine required to impose a sentencing enhancement would fall within the "actual innocence" exception, but held that the petitioner had failed to meet his evidentiary burden. 63 F. App'x 568, 569-70 (2d Cir. 2003) ("Where evidence of the drug quantity necessary to trigger a statutory maximum is 'overwhelming' and 'essentially uncontroverted,' there is no 'real threat . . . to the fairness, integrity, and public reputation of judicial proceedings' . . . . The same analysis applies to a district court's failure to submit drug quantity to a jury.").
In Poindexter, the Second Circuit held that a "legal argument" that Poindexter's three prior convictions should have been treated as one for purposes of the career offender enhancement fell outside the "actual innocence" exception, noting that Poindexter did not suggest "that he did not actually commit those three prior crimes" supporting the career offender enhancement, but "simpl[y] challenge[d] . . . the application of the Guidelines." 333 F.3d at 381-82. Poindexter also discussed a Fourth Circuit case, United States v. Maybeck, in which the court addressed exceptions to procedural default (as in Spence), and found that a showing that the defendant did not commit one of the offenses supporting his Guidelines career offender enhancement satisfied the "actual innocence" exception to procedural default. 23 F.3d 888, 891-94 (4th Cir. 1994). The Maybeck court extended the concept of "actual innocence" to non-capital sentencing, but in doing so retained the concept's traditional character of innocence of an offense: "Hence, a defendant in either a capital or non-capital case would, unless excepted from the cause and prejudice requirement, suffer the same general consequence (an enhanced sentence) from being held responsible for an act of which he or she is actually innocent." Id. at 893. Poindexter's discussion of Spence and Maybeck reiterated the same traditional concept of actual innocence: "Maybeck and Spence, on which Poindexter relies, applied the concept of 'actual innocence' to the penalty phase of noncapital offenses, granting collateral relief to applicants who showed that they in fact had not committed the crimes on which the calculation or imposition of their sentences was based." 333 F.3d at 381. Poindexter, which, like Salvagno's petition, involved a claim brought under the Triestman exception, also reiterated the need for a claim of a constitutional violation. See id. at 380 ("Poindexter's reliance on these cases is misplaced, for his argument is a technical one that neither asserts a claim of constitutional dimension nor constitutes a claim of 'actual innocence' as that term is used either in those cases or in habeas jurisprudence generally.") (emphasis added); see also id. at 382 ("This plainly does not assert a constitutional claim."). Salvagno's petition asserts neither traditional notions of actual innocence nor an independent constitutional claim, and so is far afield from the cases applying "actual innocence" to non-capital sentencing.
Later, in Darby v. United States, the Second Circuit confirmed the traditional understanding of actual innocence in § 2255 petitions: "[In Spence,] [w]e found that the defendant had a valid claim that he was actually innocent of the enhanced sentence, but we by no means suggested that the actual innocence exception applies where, as here, the defendant was indisputably guilty of the predicate offenses that led to his enhancement." See 508 F. App'x at 71 Similarly, in Tellado v. United States, this Court rejected the petitioner's argument that his untimely habeas claim under 28 U.S.C. § 2255 should be heard because he was "actually innocent" of the career offender guideline based on a subsequent change in the law. 799 F. Supp. 2d 156, 170 -71 (D. Conn. 2011), aff'd on other grounds, 745 F.3d 48 (2d Cir. 2014). The late Judge Kravitz concluded that Tellado was more like the defendant in Poindexter than the one in Spence, because he did "not assert that he is actually innocent of the act on which his harsher sentence was based." Id. at 171 (internal quotations omitted, emphasis in original). Specifically, Tellado did not claim that he was innocent of "exchanging drugs for money," the "predicate offense[]" on which the career offender enhancement was based, and so Judge Kravitz concluded that Tellado had not met the requirements of the actual innocence gateway. Id. at 171-72. Salvagno's claim that Dr. Levin's fiber-year count was wrong likewise does not suggest that Salvagno is innocent of the offenses of conviction or any predicate offense that forms the basis of an enhancement, but is simply a challenge to the imposition of U.S.S.G. § 2Q1.2(b)(2). See id. at 172 ("A court that is considering whether a petitioner's 'actual innocence' provides a gateway for consideration of an otherwise barred claim on collateral attack of the petitioner's sentence engages in a different inquiry than a sentencing court that is determining whether to apply a[n] . . . enhancement under the Guidelines."); see also Newsome v. Nalley, 128 F. App'x 815, 816 (2d Cir. 2005) ("Newsome also asserted a claim of actual innocence based on the court's finding the quantity of the drugs at issue . . . . Newsome does not assert that he was innocent of the crime as it was actually charged. His claim that the judge impermissibly made factual findings [as to the quantity of drugs] goes to the validity of his sentence, and does not provide a basis for challenging his conviction.").
Recognizing Salvagno's "actual innocence" claim would also conflict with two other lines of habeas cases. First, the Second Circuit has held that Guidelines sentencing errors not raised on appeal are not cognizable even on an initial § 2255 petition absent a showing of a "complete miscarriage of justice." See Graziano v. United States, 83 F.3d 587, 590 (2d Cir. 1996) (applying "complete miscarriage of justice" standard to non-constitutional "claims regarding a sentencing court's error in failing to properly apply the Sentencing Guidelines" not raised on direct appeal). If that exacting standard applies to the first § 2255 petition, then surely a standard at least as demanding should apply to challenges to the application of the Sentencing Guidelines under the narrow Triestman exception for § 2241 petitions, which, by definition, involve a habeas petition filed after the prosecution of an earlier § 2255 motion. Second, the Supreme Court has not yet recognized a freestanding innocence claim. See, e.g., McQuiggin v. Perkins, 569 U.S. 383, 392 (2013) ("We have not resolved whether a prisoner may be entitled to habeas relief based on a freestanding claim of actual innocence."). Rather, available precedent suggests such a claim must be paired with a constitutional violation. See House v. Bell, 547 U.S. 518, 534, 555 (2006) (holding that petitioner had satisfied the actual innocence gateway to proceed with his defaulted constitutional claims of ineffective assistance of counsel and prosecutorial misconduct, but had not shown a freestanding actual innocence claim).
Viewing Salvagno's claim through this analogous lens of defaulted non-constitutional Guidelines challenges under § 2255, I would find that Salvagno has not shown a "complete miscarriage of justice" as required to hear his claim. (Salvagno concedes that he "could not have raised his actual innocence claim on direct appeal or in his prior § 2255 motion." (ECF No. 1-1 at 7.).) "To demonstrate a complete miscarriage of justice based upon a sentencing court's miscalculation of the Guidelines, a petitioner must show an error of fact or law of [a] fundamental character that renders the entire proceeding[ ] irregular and invalid." Mercado v. U.S.A., No. 3:17CV488 (MPS), 2018 WL 1541914, at *4 (D. Conn. Mar. 29, 2018) (citations and internal quotation marks omitted). As discussed above, Salvagno does not claim that any fundamental error infected his sentencing, but simply that the prediction required to impose the § 2Q1.2(b)(2) enhancement - a "substantial likelihood of death or serious bodily injury" - was wrong in hindsight. As shown, that prediction was supported by the contemporaneous evidence at the time of the sentencing, which Salvagno had ample opportunity to challenge. Moreover, Salvagno does not claim that the sentencing court imposed a sentence that exceeded any statutory maximum. Accordingly, Salvagno raises no fundamental error, and thus he could not show a "complete miscarriage of justice" in order for his Guidelines-based challenge to be heard. Cf. Mercado, 2018 WL 1541914, at *4 (finding that challenge to application of the career offender Guideline did not constitute a "complete miscarriage of justice" where sentence imposed was less than statutory maximum).
To the extent that Triestman suggested that actual innocence itself constitutes a constitutional violation, the cases making that suggestion and their progeny address actual innocence of an offense of conviction, not innocence of a sentencing enhancement that was not itself based on the commission of an offense. See, e.g., House, 547 U.S. at 555 (holding that petitioner's demonstration of flaws in forensic proof connecting petitioner to murder, in addition to his eyewitness evidence of another suspect, did not rise to a freestanding actual innocence claim).
Salvagno's petition challenges the imposition of the § 2Q1.2(b)(2) sentencing enhancement based on new evidence that supposedly shows that Dr. Levin's estimate of a 160 fiber-year count was materially wrong. He does not claim innocence of any of his offenses of conviction. Allowing Salvagno's § 2241 petition to proceed is neither warranted by the narrow "actual innocence" exception contemplated in Treistman nor compelled by Spence, which arose in a different procedural posture and involved a different claim. Recognizing Salvagno's claim of "sentencing innocence" would drastically expand "actual innocence" claims in the habeas context, because Salvagno neither purports to satisfy traditional notions of actual innocence (i.e. innocence of the offense of conviction or any predicate offense serving as sentencing enhancements), nor points to an independent constitutional violation. Accordingly, I conclude that Salvagno's claim cannot squeeze through the "actual innocence" gateway afforded by the Triestman exception. Thus, the court lacks jurisdiction over his § 2241 petition.
II. Provable on the Existing Record
Salvagno's claim also does not meet the other prerequisite to invoke the Triestman exception—that a claim of innocence be "provable on the existing record." Salvagno does not explicitly address this gateway requirement, but implies that the Knauf declaration alone shows that the 160 fiber-year count was materially inaccurate. (ECF No. 1-1 at 9; ECF No. 11 at 16-21.)
Even assuming that the Knauf declaration and the attached loss run report prove that none of the AAR workers have made asbestos-related claims, I find that Salvagno cannot prove his claim of innocence under USSG § 2Q1.2(b) on the existing record for two reasons: first, my review of the available portions of the sentencing record shows substantially more evidence supporting the imposition of U.S.S.G. § 2Q1.2(b)(2) than in comparable cases upholding the enhancement; and second, Dr. Levin's testimony supported a conclusion that a substantial risk of death or asbestos-related injury existed even if his 160 fiber-year estimate was wrong.
First, even if Dr. Levin's 160 fiber-year exposure estimate is discarded, the existing record contains evidence exceeding that which the Second Circuit and other circuits have found sufficient to impose the U.S.S.G. § 2Q1.2(b)(2) enhancement. For example, in United States v. Thorn, the Second Circuit reversed the trial court's decision not to impose the enhancement. 317 F.3d 107, 117 (2d Cir. 2003). The Court of Appeals held that U.S.S.G. § 2Q1.2(b)(2) applied "if Thorn's Clean Air Act violations made it considerably more likely that a person would develop an injury involving extreme physical pain or the protracted impairment of a function of a bodily member, organ, or mental faculty, or requiring medical intervention such as surgery, hospitalization, or physical rehabilitation . . . [D]eath or serious bodily injury need not result from the offense conduct." Id. at 117. In addition to holding that the lower court erred by applying the enhancement "only when the offense actually caused such illness or death," the Court of Appeals found that the district court's finding that the evidence of harm was too speculative constituted "clear error." Id. at 118. The Court of Appeals cited evidence that "Thorn did not direct his workers to wear respirators," that "Thorn failed to provide respirators, replacement filters, sufficient quantities of protective gear, and decontamination units," and that Dr. Levin, the expert in that case too, had testified both that "there will be asbestos-related disease [among the exposed workers] with virtual certainty" and that one particular worker had a "nil" chance of not developing asbestos-related disease even if he were exposed for substantially less time than the worker had testified. Id. at 115, 119. The Second Circuit concluded that it was clear error for the district court not to find that "Thorn's conduct resulted in a substantial likelihood that at least one, if not several, of his former employees would develop devastating, life-threatening, asbestos-related diseases." Id. at 119.
Salvagno's argument that the Second Circuit reversed in Thorn only because Dr. Levin's testimony was "undisputed" and "unrefuted" is thus unpersuasive, as the court found ample evidence apart from his testimony to support imposition of the enhancement. (ECF No. 11 at 20.)
The Ninth Circuit has twice found that even less evidence of harm than in Thorn was sufficient to impose the § 2Q1.2(b)(2) enhancement. In United States v. Pearson, the Ninth Circuit affirmed the district court's imposition of the enhancement where "Pearson violated, and instructed others to violate, the work practice standards, resulting in conditions where asbestos was not stored, or removed properly." 274 F.3d 1225, 1235 (9th Cir. 2001). The supporting evidence was simply that: an "[in]appropriate amount of water was used to wet the asbestos"; "dry asbestos was 'all over the place'"; "[c]ontainment walls were pulled away from the ceiling, with work continuing"; "[s]ome of the negative air machines were clogged"; and one inspector "saw bags of asbestos outside the containment area." Id. at 1229. Similarly, in United States v. Yi, the Court of Appeals affirmed the application of the enhancement based on "[e]vidence presented at trial . . . that the work crew removing the [] ceilings did not wear proper respirators, were exposed to dry ceiling material, and that on-site dust far exceeded industry-recommended levels of asbestos," as well as evidence presented at sentencing that chrysotile, the type of asbestos involved, was carcinogenic. 704 F.3d 800, 806 (9th Cir. 2013). Although the government provided no expert in that case, the Ninth Circuit recognized that "there is no . . . requirement" to provide expert testimony in order to prove the enhancement. Id.
Notably, Thorn does not suggest that the quantum of evidence it cited was necessary to apply the enhancement, as opposed to the lesser showings in Pearson and Yi. But the evidence supporting the enhancement in this case - even when the 160-fiber estimate is discarded - equals even the showing described in Thorn. For example, as in all three cases, the sentencing record contained evidence, relied on by Dr. Levin, that "at any one time no more than 50 percent of the workers were wearing respiratory protection and at times many fewer than that," that AAR did not systematically employ negative pressure at its work sites, and that the few containment efforts it did undertake were largely ineffective. (Oct. 27, 2004 Hrg. Tr. 55:10-12, Tr. 60:14-61:19.) Dr. Levin also repeatedly testified that, as in Yi and Pearson, "for the most part the removal of asbestos was done under dry conditions," resulting in the "presence of visible dust in the air, at times described as so heavy as it looked like a snowstorm or a blizzard." (Id. Tr. 55:3-10, Tr. 57:16-25; see also id. Tr. 58:1-59:8 (suggesting that "dry rip[ping]" and sweeping and bagging of asbestos debris occurred); see also id. Tr. at 125:24-126:2 ("Mr. Valachovic [AAR worker] told me when he worked in these enclosed spaces where dust concentrations were so high in the air, he would experience eye, nose and throat irritation as a result of the dust in the air."). Finally, much like his testimony in Thorn, Dr. Levin stated that he would be "very surprised" to see a normal X-ray among the 100 workers after 20 years and expected a great majority to show asbestos-related scarring. (Id. Tr. 70:19-71:7.) Such evidence meets, or even exceeds, that found sufficient to impose the enhancement in Thorn, Pearson, and Yi.
The fact that, as of April 2017, there have been no reported claims does not by itself refute Dr. Levin's prediction about what might be seen on X-rays.
Second, even accepting as true the petition's assertion that Dr. Levin's exposure estimate is wrong, Dr. Levin's own conclusion that, based on the above evidence, Salvagno's workers were exposed to a "substantial likelihood of death or serious bodily injury" would be unchanged. On cross-examination, Salvagno's lawyer specifically asked Dr. Levin what would happen if, as Salvagno now claims, his 160 fiber-year estimate was wrong. (Nov. 3, 2004 Hrg. Tr. 62:7-64:1.) While Dr. Levin acknowledged that this would affect the number of people he expected would suffer serious illness (Id. Tr. 62:7-13), he asserted that he would nonetheless feel entirely comfortable predicting that some serious asbestos-related illness would occur:
Q Well, is there a point that as a doctor you would just say that you simply don't have enough data to predict illness?
A Absolutely not. You know, given the circumstances I have had described to me in this case with the levels of exposure that have been described, I feel entirely comfortable in saying the exposure levels were sufficient to produce human disease 20 and 25 years down the road, and while I can't precisely quantify it, I have no doubt that many workers will suffer scarring lung disease and unfortunately a number will also develop cancers. My estimate was based on OSHA's risk assessment with my extrapolation from work process measurements done elsewhere and taking a conservative look at that, my understanding is that the level that I described here may well be low given the amount of dry brooming and sweeping that was a continual activity during the course of the larger project, not just at the end of the day, so resuspension of this material ongoing during the day apparently characterized lots of these jobs, and I would give somewhat higher estimate than I offered had I been aware of how often the dry sweeping and cleaning was going on during the day of this project(Nov. 3, 2004 Hrg. Tr. 62:7-64:1 (emphasis added).) He further suggested that his conclusion rested in part on his medical experience, which was independent of his 160 fiber-years estimate:
Q And in addition to the studies that you have discussed in your direct and cross-examination, have you personally observed asbestos disease in your patients?(Nov. 3, 2004 Hrg. Tr. Tr. 67:20-68:19; see also id. Tr. at 69:17-19 ("So, yes, I have seen people with lesser exposures than I believe occurred among these employees who developed significant illness.").) This is all the caselaw requires to impose the § 2Q1.2(b)(2) enhancement. See Thorn, 317 F.3d at 119 ("Thorn's conduct resulted in a substantial likelihood that at least one, if not several, of his former employees would develop devastating, life-threatening, asbestos-related diseases." (emphasis added)).
A Unfortunately, I have seen it all too often.
Q And can you give the Court some idea of approximately how many patients your organization has medically examined with regard to asbestos related disease?
[Court overruled objections]
A I personally have examined thousands of workers myself in the last 24 years . . . . I have, in fact, been responsible for medically monitoring and taking care of many patients who have developed asbestos related disease.
Q And based upon your interview of these people, did they have similar exposures to AAR employees, what has been described to you?
A Well, many of the patients that I have seen are people who had lesser exposures than I have heard described for the AAR employees.
Accordingly, even if Dr. Levin's 160 fiber-count estimate was flawed, the existing record contains evidence exceeding that in comparable cases in which the § 2Q1.2(b)(2) enhancement was imposed. Moreover, Dr. Levin acknowledged that even if his 160 fiber-year count was wrong, as Salvagno's petition now argues, he still believed that the exposure levels would cause asbestos-related disease in Salvagno's workers based on the level of exposure described. Cf. Cephas v. Nash, 328 F.3d 98, 107-08 (2d Cir. 2003) (concluding that petitioner could not demonstrate actual innocence of a continuing criminal enterprise charge on the existing record because the jury had found him guilty on substantive narcotics charges that could form the predicate violations). In light of the existing record evidence of harm and Dr. Levin's own testimony concerning Salvagno's claimed error in his analysis, Salvagno cannot show that he is actually innocent of § 2Q1.2(b)(2) based on the existing record. Thus, even if Salvagno's claim sounds in actual innocence, he cannot prove his claim on the existing record and thus cannot invoke the Triestman exception for § 2255-type relief.
CONCLUSION
Salvagno's motion for leave to file a sur-reply is GRANTED. (ECF No. 14.) Because Salvagno does not assert a cognizable claim of "actual innocence" and because his claim in any event cannot be proved on the existing record, Salvagno's § 2241 petition does not fall within the Triestman exception for challenges to the imposition of his sentence. Accordingly, the Court GRANTS Williams' (ECF No. 7) motion to dismiss for lack of subject matter jurisdiction and DISMISSES Salvagno's petition (ECF No. 1) for writ of habeas corpus. Salvagno's motion to appoint counsel (ECF No. 12) is DENIED as moot. The Clerk is directed to close the case.
IT IS SO ORDERED.
/s/_________
Michael P. Shea, U.S.D.J. Dated: Hartford, Connecticut
January 4, 2019