BOWNES, Senior Circuit Judge.
In this appeal, Clayton Libby, a Massachusetts state prisoner serving a life sentence on a 1971 conviction for murder in the first degree, challenges the district court’s denial of his petition for a writ of habeas corpus. In so doing, petitioner primarily contends that the court erred in deeming harmless a jury instruction on the issue of malice which set up an unconstitutional mandatory presumption. See Sandstrom v. Montana, 442 U.S. 510, 520-24, 99 S.Ct. 2450, 2451-59, 61 L.Ed.2d 39 (1979) (instruction containing presumption which has the effect of relieving the prosecution of the burden of proof on an element of a charged crime violates the Due Process Clause) (hereinafter “Sandstrom error”).1 We affirm.
I.
BACKGROUND
Early in the morning of August 9, 1970, Bruce Cullen, a New Hampshire resident, was stabbed to death in a brawl that erupted outside of a South Boston housing project. Petitioner and George Cooper were indicted and tried for the killing. Cooper was acquitted; Libby, however, was convicted of murder in the first degree.
Although the circumstances in which the stabbing took place are sketchy, the trial record reveals that, on the night of August 8, 1970, petitioner was drinking beer, smoking marijuana, and possibly taking diet pills. Sometime early in the morning of August 9, 1970, petitioner, along with Francis Barton and Kevin Martin, went to George Cooper’s South Boston apartment building and began to converse with Cooper through a rear apartment window. After a while, petitioner and Martin walked to the front of the building where they met several other men. Included among these men were the victim, Cullen, and another New Hampshire resident, Dennis Bates.
At some point, a fight broke out. The reason for the fight is not entirely clear, although there was testimony indicating that it started simply because Cullen and Bates were not from the area. There also was testimony indicating that petitioner and Cullen were arguing about whether Cullen had been in a certain federal prison. In any event, during the course of the fight, Cullen was stabbed nine times. Six of the stab wounds were to his chest; the other three were to-his back or side. One of the chest wounds was to the victim’s heart, and apparently was delivered by a “downward” blow.
No witness testified to actually observing the stabbing.2 Instead, petitioner was incul[735] pated through the testimony of eyewitnesses who observed him both before and after the fight. Specifically, there was testimony that, inter alia, petitioner (1) had been carrying a knife prior to the fight; (2) was seen running away from the site of the fight with blood on his clothes; (3) was seen holding a knife shortly after the stabbing; (4) admitted, on several occasions after the fight, that he had done the stabbing; and (5) made 'threats against anyone who might “snitch[ ] on him.” There also was testimony that petitioner had stabbed Cullen because he thought Cullen was “going to jump him from behind” and/or because he thought Cullen was “beating up Kevin Martin.”
At the conclusion of a seven-day jury trial, the trial judge instructed the jury on theories of first degree murder,3 second degree murder, and manslaughter. In the course of defining malice, which is “the requisite mental element” of murder under Massachusetts law, see Commonwealth v. Huot, 380 Mass. 403, 403 N.E.2d 411, 414 (1980), overruled on other grounds, Commonwealth v. Bray, 407 Mass. 296, 553 N.E.2d 538 (1990),4 the judge told the jury that “[mjaliee is implied in every deliberate cruel act by one against another.” The jury convicted petitioner of first degree murder and recommended a sentence of life imprisonment.
On appeal, petitioner argued, inter alia, that the aforementioned instruction constituted Sandstrom error and required reversal of his conviction.5 More particularly, petitioner asserted that the instruction had the effect of directing the jury to find malice if it found that petitioner had committed a “deliberate cruel” act against the victim despite the fact that a “deliberate cruel” act is not necessarily malicious. The SJC disagreed, holding: “In the context of the facts of this case and in light of the judge’s entire instruction on malice (which is not otherwise challenged), we see neither a substantial likelihood of a miscarriage of justice calling for relief ... nor an unconstitutional presumption dictated to the jury.” Commonwealth v. Libby, 405 Mass. 231, 540 N.E.2d 154, 158 (1989) (hereinafter “Libby I ”). The SJC then went on to affirm the conviction, although it remanded to the superior court for consideration of a previously-filed motion to dismiss the indictment on account of delay.
[736] In August 1990, petitioner’s motion to dismiss was denied by the superior court. In September 1990, petitioner filed a timely notice of appeal from this denial. While that appeal was pending, this court handed down its decision in Hill v. Maloney, 927 F.2d 646 (1st Cir.1990). See supra note 5. In Hill, we held unconstitutional a jury instruction quite similar to the one here at issue. Id. at 649-51.6 Relying on Hill, petitioner argued for a second time to the SJC that his conviction should be set aside because the trial judge’s instruction had the effect of setting up an unconstitutional mandatory presumption. Once again, the SJC rejected petitioner’s argument and affirmed his conviction. See Libby II, 580 N.E.2d at 1028.
Finally, petitioner sought relief in the district court by means of a writ of habeas corpus. In a comprehensive memorandum and order, the court applied the three-part test set forth in Hill for review of alleged Sandstrom errors and denied the writ. First, the court determined that the challenged instruction set up an unconstitutional mandatory presumption and therefore constituted Sandstrom error. See Libby v. Duval, No. 86-2187-WD, slip op. at 8-9 (D.Mass. April 20, 1993) (hereinafter “Libby III ”); see also Hill, 927 F.2d at 648-49. Next, the court found that the instructions as a whole did not sufficiently explain the erroneous instruction on malice, and the jury therefore was not properly instructed on the law. See Libby III, slip op. at 10-13; see also Hill, 927 F.2d at 649. Finally, the court concluded that the instruction, although erroneous, was harmless beyond a reasonable doubt. See Libby III, slip op. at 13-19; see also Hill, 927 F.2d at 649. It is from this last conclusion that petitioner appeals.
II.
DISCUSSION
Free access — add to your briefcase to read the full text and ask questions with AI
BOWNES, Senior Circuit Judge.
In this appeal, Clayton Libby, a Massachusetts state prisoner serving a life sentence on a 1971 conviction for murder in the first degree, challenges the district court’s denial of his petition for a writ of habeas corpus. In so doing, petitioner primarily contends that the court erred in deeming harmless a jury instruction on the issue of malice which set up an unconstitutional mandatory presumption. See Sandstrom v. Montana, 442 U.S. 510, 520-24, 99 S.Ct. 2450, 2451-59, 61 L.Ed.2d 39 (1979) (instruction containing presumption which has the effect of relieving the prosecution of the burden of proof on an element of a charged crime violates the Due Process Clause) (hereinafter “Sandstrom error”).1 We affirm.
I.
BACKGROUND
Early in the morning of August 9, 1970, Bruce Cullen, a New Hampshire resident, was stabbed to death in a brawl that erupted outside of a South Boston housing project. Petitioner and George Cooper were indicted and tried for the killing. Cooper was acquitted; Libby, however, was convicted of murder in the first degree.
Although the circumstances in which the stabbing took place are sketchy, the trial record reveals that, on the night of August 8, 1970, petitioner was drinking beer, smoking marijuana, and possibly taking diet pills. Sometime early in the morning of August 9, 1970, petitioner, along with Francis Barton and Kevin Martin, went to George Cooper’s South Boston apartment building and began to converse with Cooper through a rear apartment window. After a while, petitioner and Martin walked to the front of the building where they met several other men. Included among these men were the victim, Cullen, and another New Hampshire resident, Dennis Bates.
At some point, a fight broke out. The reason for the fight is not entirely clear, although there was testimony indicating that it started simply because Cullen and Bates were not from the area. There also was testimony indicating that petitioner and Cullen were arguing about whether Cullen had been in a certain federal prison. In any event, during the course of the fight, Cullen was stabbed nine times. Six of the stab wounds were to his chest; the other three were to-his back or side. One of the chest wounds was to the victim’s heart, and apparently was delivered by a “downward” blow.
No witness testified to actually observing the stabbing.2 Instead, petitioner was incul[735] pated through the testimony of eyewitnesses who observed him both before and after the fight. Specifically, there was testimony that, inter alia, petitioner (1) had been carrying a knife prior to the fight; (2) was seen running away from the site of the fight with blood on his clothes; (3) was seen holding a knife shortly after the stabbing; (4) admitted, on several occasions after the fight, that he had done the stabbing; and (5) made 'threats against anyone who might “snitch[ ] on him.” There also was testimony that petitioner had stabbed Cullen because he thought Cullen was “going to jump him from behind” and/or because he thought Cullen was “beating up Kevin Martin.”
At the conclusion of a seven-day jury trial, the trial judge instructed the jury on theories of first degree murder,3 second degree murder, and manslaughter. In the course of defining malice, which is “the requisite mental element” of murder under Massachusetts law, see Commonwealth v. Huot, 380 Mass. 403, 403 N.E.2d 411, 414 (1980), overruled on other grounds, Commonwealth v. Bray, 407 Mass. 296, 553 N.E.2d 538 (1990),4 the judge told the jury that “[mjaliee is implied in every deliberate cruel act by one against another.” The jury convicted petitioner of first degree murder and recommended a sentence of life imprisonment.
On appeal, petitioner argued, inter alia, that the aforementioned instruction constituted Sandstrom error and required reversal of his conviction.5 More particularly, petitioner asserted that the instruction had the effect of directing the jury to find malice if it found that petitioner had committed a “deliberate cruel” act against the victim despite the fact that a “deliberate cruel” act is not necessarily malicious. The SJC disagreed, holding: “In the context of the facts of this case and in light of the judge’s entire instruction on malice (which is not otherwise challenged), we see neither a substantial likelihood of a miscarriage of justice calling for relief ... nor an unconstitutional presumption dictated to the jury.” Commonwealth v. Libby, 405 Mass. 231, 540 N.E.2d 154, 158 (1989) (hereinafter “Libby I ”). The SJC then went on to affirm the conviction, although it remanded to the superior court for consideration of a previously-filed motion to dismiss the indictment on account of delay.
[736] In August 1990, petitioner’s motion to dismiss was denied by the superior court. In September 1990, petitioner filed a timely notice of appeal from this denial. While that appeal was pending, this court handed down its decision in Hill v. Maloney, 927 F.2d 646 (1st Cir.1990). See supra note 5. In Hill, we held unconstitutional a jury instruction quite similar to the one here at issue. Id. at 649-51.6 Relying on Hill, petitioner argued for a second time to the SJC that his conviction should be set aside because the trial judge’s instruction had the effect of setting up an unconstitutional mandatory presumption. Once again, the SJC rejected petitioner’s argument and affirmed his conviction. See Libby II, 580 N.E.2d at 1028.
Finally, petitioner sought relief in the district court by means of a writ of habeas corpus. In a comprehensive memorandum and order, the court applied the three-part test set forth in Hill for review of alleged Sandstrom errors and denied the writ. First, the court determined that the challenged instruction set up an unconstitutional mandatory presumption and therefore constituted Sandstrom error. See Libby v. Duval, No. 86-2187-WD, slip op. at 8-9 (D.Mass. April 20, 1993) (hereinafter “Libby III ”); see also Hill, 927 F.2d at 648-49. Next, the court found that the instructions as a whole did not sufficiently explain the erroneous instruction on malice, and the jury therefore was not properly instructed on the law. See Libby III, slip op. at 10-13; see also Hill, 927 F.2d at 649. Finally, the court concluded that the instruction, although erroneous, was harmless beyond a reasonable doubt. See Libby III, slip op. at 13-19; see also Hill, 927 F.2d at 649. It is from this last conclusion that petitioner appeals.
II.
DISCUSSION
We agree with the district court that the tripartite Hill test applies to the merits of petitioner’s claim. Accordingly, we organize our discussion within the Hill framework.
A.
Because the Hill opinion rehearses in great detail the legal standards applicable to challenges to jury instructions which set up presumptions, and because respondents7 concede that the instruction challenged here established a mandatory presumption, we do not believe that either a highly detailed discussion of the law of presumptions or an extensive explanation of why the instruction was defective is required.8 Instead, we think it sufficient to note our belief that it was reasonably likely that the jurors construed the trial judge’s instruction as requiring a finding of malice upon a finding that the stabbing was “deliberate” and “cruel.” See Estelle v. McGuire, — U.S. -, -, 112 S.Ct. 475, 482, 116 L.Ed.2d 385 (1991) (habe-as challenges to jury instructions reviewed for “reasonable likelihood” that the jury has applied the challenged instruction in an unconstitutional manner). Thus, because the instruction had the effect of relieving the prosecution of the burden of proof on an element of the crime charged, see Sand-strom, 442 U.S. at 520-24, 99 S.Ct. at 2457-59, the district court’s conclusion that the instruction established a mandatory presumption was clearly correct. Accordingly, we move to step two of the Hill test.
B.
Once we have determined that the specific language challenged by a petitioner set up a mandatory presumption, we consider whether other parts of the instruction explained the particular infirm language to the extent that there is no reasonable likelihood that the jurors applied the unconstitutional presumption. See Boyde v. California, 494 U.S. 370, [737]*737380, 110 S.Ct. 1190, 1198, 108 L.Ed.2d 316 (1990). General instructions regarding the presumption of innocence and the government’s burden of proving all elements of a crime beyond a reasonable doubt are insufficient to fulfill this explanatory role. Id. at 651.9 So too are instructions directly contrary to the erroneous one which themselves correctly state the law. Id.10 Instead, there must be other language in the instructions which actually “explains the infirm language sufficiently so that there is no reasonable likelihood that the jury believed it must [in the context of an erroneous malice instruction] find malice if it found petitioner [acted in such a way so as to trigger the unconstitutional presumption].” Id.
Respondents contend that four sections of the instructions, when taken together, sufficiently explain the infirm language. After reviewing these four sections, and after further reviewing the instructions as a whole, we cannot agree.
The first two sections adduced by respondents involve definitions of malice given prior to the unconstitutional instruction.11 In each of these two instances, the trial judge correctly instructed the jurors that malice was not proved where, among other things, there were “extenuating circumstances” sufficient to “reduce the crime to manslaughter.” At most, we think that these two definitions of malice might have allowed the jurors to infer that the presumption set up by the subsequent unconstitutional charge could be rebutted in certain extenuating circumstances. Cf. id. at 653. We do not, however, see how these definitions could have explained to the jurors that the upcoming instruction was not going to mean what it actually and clearly stated. At any rate, we reiterate that- correct instructions, which directly contradict the erroneous instruction, are not sufficient to cure the error. See supra at p. 737 and note 10.
The other two sections cited by respondents are less compelling. The first of these two, which again preceded the infirm instruction, primarily defined the terms “aforethought” and “murder.”12 The second defined the term “premeditated.”13 Both of the passages are jumbled and confusing, especially when compared to the short and relatively straightforward statement which set up the unconstitutional mandatory presumption. Cf. Hill, 927 F.2d at 652 (juxtaposing clumsily-worded correct instruction with clear and concise unconstitutional instruction in deciding that correct portions of charge as a whole did not negate the effects of the presumption-creating language). Moreover, neither passage explicitly touched [738] on the concept of malice, except insofar as the first one briefly discussed the “aforethought” component of the term “malice aforethought.” In light of these deficiencies, we do not see how these, passages could have actually explained the challenged instruction “so as to offset any erroneous impression given by [it].” See id. at 651.
Before concluding our analysis of the entire charge, we pause to note that, because it was framed in irrefutable and unvarying terms (“[m]alice is implied in every deliberate and cruel act by one against another”), we think it at least reasonably likely that the challenged instruction completely removed the element of malice from the case once the Commonwealth established that petitioner had acted deliberately and cruelly.14 Therefore, in conducting our harmless-error analysis, we will regard the instruction as having erected a conclusive mandatory presumption. See Hill, 927 F.2d at 649 n. 3 (distinguishing between conclusive mandatory presumptions and rebuttable mandatory presumptions).
In sum, we agree with the district court that the charge as a whole did not neutralize the effect of the presumption-creating language. Accordingly, the effect of the instruction here was unconstitutional.
C.
Having determined that the overall charge did not adequately explain the challenged instruction, we still must ascertain whether the error was harmless. See id. at 654; see also Sullivan v. Louisiana, — U.S. -, -, 113 S.Ct. 2078, 2082, 124 L.Ed.2d 182 (1993) (reiterating that Sandstrom error is subject to harmless-error review). In Hill, because the Supreme Court had sent mixed signals regarding the proper harmless-error analysis to be. employed where there has been a mandatory presumption, we applied two separate approaches. First, we looked at the trial record as a whole to determine whether it was clear “beyond a reasonable doubt” that the error was harmless. See Hill, 927 F.2d at 655 (applying the Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 828, 17 L.Ed.2d 705 -(1967), standard for determining, on direct review, whether a conviction must be set aside because of federal constitutional error); see also Rose v. Clark, 478 U.S. 570, 580-82, 106 S.Ct. 3101, 3107-08, 92 L.Ed.2d 460 (1986) (applying Chapman harmless-error standard to a presumption-creating jury instruction challenged on habeas). Alternatively, we utilized a narrower approach, derived from Chapman, for analyzing the effects of a conclusive mandatory presumption urged by Justice Scalia in a concurring opinion in Carella v. California, 491 U.S. 263, 267-73, 109 S.Ct. 2419, 2421-24, 105 L.Ed.2d 218 (1989) (hereinafter the “Carella test”). See Hill, 927 F.2d at 654-56. Under both approaches we determined that the error was not harmless. Id. at 657.
Since the decision in Hill, however, (and since the district court passed on whether or not the presumption-creating instruction was harmless), the Supreme Court has issued Brecht v. Abrahamson, — U.S. -, 113 S.Ct. 1710, 123 L.Ed.2d 353 (1993), which clarifies that the two approaches employed in Hill are no longer applicable on collateral-review. In Brecht, the Supreme Court announced that the' Chapman “harmless beyond a reasonable doubt” test should not be utilized by courts reviewing claims of constitutional error of the trial type on habeas, id. at-, 113 S.Ct. at 1717; instead, reviewing courts should now look to whether error “ ‘had substantial and injurious effect or influence in determining the jury’s verdict.’ ” Id. at-, 113 S.Ct. at 1722 (quoting Kot-teakos v. United States, 328 U.S. 750, 776, 66 S.Ct. 1239, 1253, 90 L.Ed. 1557 (1946)).
Petitioner contends that the Brecht approach is inappropriate in the conclusive presumption context. He therefore urges us to explicitly adopt the Carella test for determining whether or not an instruction creating such a presumption can be viewed as harmless error. In light of the clear and uncom[739] promising language employed by the Court in Brecht, we decline to do so.
The issue presented in Brecht was whether the prosecutor’s use of petitioner’s post-Miranda silence for impeachment purposes at petitioner’s trial, which violated petitioner’s due process rights under Doyle v. Ohio, 426 U.S. 610, 96 S.Ct. 2240, 49 L.Ed.2d 91 (1976) (hereinafter “Doyle error”), was harmless. In concluding that it was, the majority decided, as we have noted, that the Kotteakos harmless-error standard was the appropriate lens through which to view the claim on habeas. See Brecht, — U.S. at -, 113 S.Ct. at 1722. In so doing, the Court departed from the approach taken in certain other habeas cases where it had assumed the applicability of the Chapman standard. Id. at -, 113 S.Ct. at 1718 (citing Yates v. Evatt, 500 U.S. 391, 111 S.Ct. 1884, 114 L.Ed.2d 432 (1991); Rose v. Clark, 478 U.S. 570, 106 S.Ct. 3101, 92 L.Ed.2d 460 (1986); Milton v. Wainwright, 407 U.S. 371, 92 S.Ct. 2174, 33 L.Ed.2d 1 (1972); Anderson v. Nelson, 390 U.S. 523, 88 S.Ct. 1133, 20 L.Ed.2d 81 (1968) (per curiam)).
In conducting its analysis, the Court began by observing that Doyle error fit into the category of constitutional error known as “trial error.” See Brecht, — U.S. at-, 113 S.Ct. at 1717. These are errors which “ ‘occur[ ] during the presentation of the case to the jury,’ and [are] amenable to harmless error analysis because [they] ‘may be quantitatively assessed in the context of other evidence presented in order to determine the effect [they] had on the trial.’ ” Id. (quoting Arizona v. Fulminante, 499 U.S. 279, 307-08, 111 S.Ct. 1246, 1263-64, 113 L.Ed.2d 302 (1991)). Errors of the trial type have, since Chapman, been'reviewed under the “harmless-beyond-a-reasonable-doubt” standard.
The Court also noted that, at the other end of the spectrum of constitutional errors are “ ‘structural defects in the constitution of the trial mechanism, which defy analysis by ‘harmless-error’ standards’ ... because they infect the entire trial process.” Id. (quoting Fulminante, 499 U.S. at 309, 111 S.Ct. at 1264). Listed as an example of such a structural defect was deprivation of the right to counsel. Id. (citing Gideon v. Wainwright, 372 U.S. 335, 83 S.Ct. 792, 9 L.Ed.2d 799 (1963)).
After reaching this conclusion, and determining that neither the doctrine of stare decisis nor congressional silence prevented it from considering the merits of respondent’s argument that the Kotteakos standard, and not the Chapman standard, should be employed in determining whether the Doyle error was harmless, see generally id. — U.S. at-, 113 S.Ct. at 1718-19, the Court turned to an analysis of whether the Chapman standard appropriately served certain interests implicated in habeas eases but not . in the direct review context whence it sprang. Pointing to (1) the state’s interest in finality of convictions that have survived direct review within the state court system; (2) the interests of comity; (3) the interests' of federalism; and (4) the interest of maintaining the prominence of the trial itself, the Court decided that application of the Chapman standard to trial errors challenged on habeas resulted in an “imbalance of ... costs and benefits.” Id. at-, 113 S.Ct. at 1721. Accordingly, the Court embraced the less onerous Kotteakos standard, holding that it applies “in determining whether habeas relief must be granted because of constitutional error of the trial type.” Id. at -, 113 S.Ct. at 1722. Like the Chapman test, this “actual prejudice” inquiry presumes that the reviewing court will conduct its harmlessness assessment “in light of the record as a whole.” Id.
After Brecht, we think it apparent that the question of whether to apply the Kotteakos test in conducting our harmless error inquiry turns on whether the conclusive presumption here at issue constitutes “trial error.” Despite the force of some of our dissenting brother’s arguments, we are constrained by the Supreme Court’s teaching that it should be so considered. See Arizona v. Fulminante, 499 U.S. 279, 306-07, 111 S.Ct. 1246, 1261-62, 113 L.Ed.2d 302 (1990) (citing Car-ella). Thus, we train our sights on whether, in light of the whole record, respondents have met their burden15 of demonstrating [740] that the conclusive presumption did not actually prejudice petitioner because it did not have a substantial and injurious effect or influence in determining the jury’s verdict. In our view, respondents have met their burden.
The thrust of petitioner’s harmfulness claim is not that, in the absence of the conclusive presumption, he would have been acquitted. Rather, petitioner’s argument is that the conclusive presumption precluded the jury from convicting him of manslaughter on a theory of “sudden combat.” See Commonwealth v. Richard, 377 Mass. 64, 384 N.E.2d 636, 638 (1979) (the presence of sudden combat constitutes a mitigating circumstance sufficient to reduce a verdict of murder to manslaughter under Massachusetts law). While we concede (1) that there was evidence (ie., the testimony that petitioner stabbed Cullen because he thought Cullen was “going to jump him from behind” and/or because he thought Cullen was “beating up Kevin Martin”) which might conceivably have provided a basis for the jury to have concluded that the government had not proved an absence of sudden combat; and (2) that the conclusive presumption tended to deter the jury from considering this evidence, see Yates, 500 U.S. at 406 n. 10, 111 S.Ct. at 1894 n. 10 (conclusive presumptions tend to deter a jury from considering any evidence for the presumed fact beyond the predicate evidence), we think it extremely unlikely that the jury would have relied on this evidence and returned a verdict of manslaughter. Mary VanGordon testified that the stabbing was administered while the victim was being held by another man. Moreover, the evidence reveals that the victim was. stabbed nine times, with six of the stab wounds being delivered to the .chest area. In our view, such evidence, when combined with the fact that the defense never specifically argued a sudden combat theory to the jury,16 strongly undermines any claim that petitioner was, throughout the entirety of the stabbing, acting in response to sudden combat.
We do not believe that the erroneous instruction had a substantial and injurious effect or influence on the jury’s verdict. Accordingly, we affirm the district court’s conclusion that the instructional error was harmless.17
Affirmed.