Watson v. United States

FERREN, Associate Judge,

dissenting:

I see this case from a different perspective — a perspective that leads to reversal.

I.

It is not clear why Judge Gallagher, writing for a plurality, has bothered to discuss the second, prejudice prong under Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). Ante at 1069-1072. That discussion is unnecessary in light of the plurality’s conclusion that appellate counsel was not constitutionally deficient. But, my colleagues have done so and, in the process, have concluded there was no prejudice from consolidation of the indictment charging first degree murder of Timothy Reeves with the indictment charging an armed assault against Richard Knight and an assatilt with intent to kill while armed against Robert Swearinger and Cynthia Durham. The plurality perceives no prejudice because — contrary to the unanimous conclusion of the division that heard Watson’s direct appeal — “it is evident that evidence on the three charges was ‘mutually admissible.’ ” Ante at 1070.

Consequently, suggests the plurality, the original division on direct appeal that rejected mutual admissibility, but affirmed on another ground, made too much of the case as it was.

My colleagues’ conclusion that the “signature crimes” exception justified mutual admissibility is simply wrong. Unfortunately, neither in the division opinion on collateral attack, Watson v. United States, 508 A.2d 75 (D.C.), reh’g granted, judgment vacated, 514 A.2d 800 (D.C.1986), nor in the en banc opinion are the facts sufficiently elaborated to make this issue and its resolution clear. Accordingly, I have attached as an appendix the unpublished memorandum opinion and judgment on the direct appeal, for it contains the recitation of events necessary to understand the consolidation issue. Once it is clear that consolidation could not properly have rested on the “signature crimes” exception, I hope it will become evident why much of the plurality's analysis unravels.

A.

Pursuant to Super.Ct.Crim.R. 13, the trial judge “may order 2 or more indictments ... to be tried together if the offenses ... could have been joined in a single indictment.” Offenses may be joined in a single indictment under Super.Ct.Crim.R. 8(a) if they “are of the same or similar character or are based on the same act or transaction or on two or more transactions connected together or constituting parts of a common scheme or plan.” Because the incidents at issue were not part of the same act or transaction, our inquiry focuses on whether the assaults and the murder were of a sufficiently similar character, or were related enough through a common scheme or plan, to warrant a joint trial — or whether, instead, the trial court abused its discretion in ordering the offenses tried together. Grant v. United States, 402 A.2d 405, 407-08 (D.C.1979). “The appropriateness of joinder in evidence-of-other-crimes cases must be determined by balancing inevitable prejudice to the defendant caused by the joinder against the legitimate probative force of the evidence and expedition in judicial administration.” Tinsley v. United *1074States, 368 A.2d 531, 533 (D.C.1976) (citing Drew v. United States, 118 U.S.App.D.C. 11, 331 F.2d 85 (1964)).

The types of prejudice likely to arise when multiple offenses are tried together were outlined in Drew:

[T]he defendant may be prejudiced for one or more of the following reasons: (1) he may become embarrassed or confounded in presenting separate defenses; (2) the jury may use the evidence of one of the crimes charged to infer a criminal disposition on the part of the defendant from which is found his guilt of the other crime or crimes charged; or (3) the jury may cumulate the evidence of the various crimes charged and find guilt when, if considered separately, it would not so find.

Id. at 14, 331 F.2d at 88. If, however, evidence of one of the crimes charged would be admissible in the defendant’s separate trial on the other, the last two types of prejudice need not be weighed in the balance.1 Because the jury would have evidence of the other crime before it anyway, the possibility that the jury would cumulate the evidence or infer a criminal disposition already exists. In cases of mutual admissibility, therefore, the court already has determined that the probative force of the other crimes evidence outweighs the prejudicial danger.

Although, as a general rule, evidence of other crimes is not admissible in a criminal trial, Drew explains that other crimes evidence is admissible if relevant to:

(1) motive, (2) intent, (3) the absence of mistake or accident, (4) a common scheme or plan embracing the commission of two or more crimes so related to each other that proof of the one tends to establish the other, and (5) the identity of the person charged with the commission of the crime on trial.

Id. at 16, 331 F.2d at 90 (footnote omitted). In denying Watson’s motion for a new trial, the trial judge stated that the indictments were not improperly joined because “the evidence of the first two offenses could have been used in the murder trial to show motive or intent. The evidence clearly showed a pattern of action and a state of mind of the defendant, which was relevant to the murder charge.” Thus, the court (without clearly justifying mutual admissibility) appeared to invoke four of the five Drew criteria.2 Interestingly, the government on direct appeal defended mutual admissibility only on the basis of “motive.” Because my colleagues perceive “signature” crimes, however, I will discuss that “identity” issue as well.

Evidence of other so-called “signature” crimes, where remarkably similar elements reveal a common “modus operandi,” will be admitted to establish identity. See Bridges v. United States, 381 A.2d 1073, 1075-78 (D.C.1977), cert. denied, 439 U.S. 842, 99 S.Ct. 135, 58 L.Ed.2d 141 (1978); Arnold v. United States, 358 A.2d 335, 338-39 (D.C.1976) (en banc). Two decisions of this court illustrate the virtual uniqueness implied by a “modus operandi” or “signature”: Calaway v. United States, 408 A.2d 1220, 1226 (D.C.1979) (evidence of prior assault admissible to show modus operandi where government alleges that, in both pri- or assault and offense charged, defendant approached a young woman on pretense of looking for room to rent, forced her to remove her clothes, subdued her with a blow to the jaw, climbed atop her and strangled her); Crisafi v. United States, 383 A.2d 1, 4-5 (D.C.), cert. denied, 439 U.S. 931, 99 S.Ct. 322, 58 L.Ed.2d 326 (1978) (evidence of prior rape admissible to *1075show modus operandi where government alleges that, in both prior rape and offense charged, defendant approached a woman in a public place and introduced himself as “Simone,” later telephoned the woman, saying he was “George,” Simone’s roommate, and urging her to date Simone, and then attempted to rape her on the resulting “date”).

In the present case, the crimes themselves — the assaults and the murder — were not so remarkably similar that they betray the identity of the perpetrator. Cf. Calaway; Crisafi. Indeed, the common elements are far from strikingly similar. Although these were all crimes of violence, the two assaults — one a threat, the other a shooting — were committed exclusively with firearms, whereas the murdered Timothy Reeves suffered both gunshot wounds and multiple stabbings. The crimes also differed in severity: threatening with a gun, as in the Richard Knight assault, is a far cry from murder by gunshot and stab wounds. Finally, the fact that the assaults and the murder took place in the same house does not make otherwise distinguishable crimes remarkably similar to the point of suggesting a common “modus operandi” or “signature," especially because nothing about the house itself figured in the manner in which the crimes were committed.

This case is even a weaker candidate for the “signature” crimes label than we found in Tinsley, supra, where, in both cases, the homicides were by strangulation after the victim had been seen drinking with the defendant. We stressed that the methods of strangulation were different and that the injuries were not so unique as to suggest the “signature” of the killer. Id. at 534-36; See Bridges, 381 A.2d at 1076. In the present case, the connection between the victims is certainly no greater than in Tinsley — a location in common for the violence here, compared with drinking in common in Tinsley — and the methods of violence are more dissimilar than in Tinsley. Frankly, it is inconceivable to me how my colleagues could have carefully focused on these facts and yet concluded that this case reflects “a classic demonstration of crimes with the same modus operandi (signature crimes).” Ante at 1071. The facts simply do not meet the test.

The government urged affirmance on direct appeal for an altogether different reason: the crimes revealed the same motive —a desire to eject someone from the Quincy Street house. Such a superficial connection, however, is not sufficient to establish a motivation sufficient for mutual admissibility. Other crimes evidence can only be admitted under the motive exception when one crime is probative of the reason or incentive for committing the other. For example, in Horton v. United States, 377 A.2d 390 (D.C.1977), Horton robbed an acquaintance named Rodwell. When Horton learned that Rodwell had reported the crime to the police, he and some friends assaulted Rodwell and raped his wife. We held that the counts were properly joined in a single indictment because the evidence of one crime would have been admissible in a trial for the other: Horton’s “revenge” motive in the rape and assault was inextricably linked to Rodwell’s reporting the robbery.3

The same is true of the two assaults in this case, which were linked by a common motive and, therefore, appropriately joined in a single indictment. Watson originally threatened Richard Knight, a friend of Robert Swearinger, in order to get him out of the house. Watson thereby wanted to communicate to Robert Swearinger, who lived there, that Watson and Joan Swear-inger (Robert Swearinger’s estranged wife who lived with Watson) intended to take over the house. The underlying motive was for Watson and Joan Swearinger to reap some economic benefit from the house, which they believed to be her prop*1076erty. Under the same “claim of right,” Watson shot Robert Swearinger and Swearinger’s friend, Cynthia Durham, only six days later. The purpose was the same: to scare Robert Swearinger out of the house.

The same reasoning does not apply to the murder of Timothy Reeves. Reeves occupied the house over a year later under a rental agreement between Reeves’ mother and Joan Swearinger, not by a claim of ownership as Robert Swearinger had asserted. Reeves’ mother had informed Joan Swearinger that she and her family would be moving out. She planned to go by June 28 but her son would stay until July 1. Watson apparently wanted Reeves out before July 1 simply because Watson wanted to clean the house before a new tenant moved in. Watson did not have a continuing desire or need to keep the house empty, which could be the only possible motive linking the three crimes. To the contrary, the evidence indicates that Watson and Joan Swearinger sought out paying tenants. Watson’s purpose — his motive in wanting Timothy Reeves out of the house — was entirely different from his purpose in seeking to evict Robert Swearinger. Therefore, a common motive cannot be said to link the three crimes.

Under the government’s theory, “motive” means little more than “propensity.” Because Watson had used guns to scare people out of the house on two previous occasions, the government asked the jury to believe that he used deadly force to eliminate an intervenor a third time.4 That is “the very inference that the general rule of exclusion of other crimes evidence is designed to prohibit.” C. Wright & K. Graham, 22 Federal Practice and Procedure, § 5239 at 460-61 (1978) (footnote omitted). The first degree murder charge, therefore, should not have been consolidated for trial with the two assaults on the ground of mutual admissibility of the evidence of the other crimes.

B.

The division on direct appeal accordingly was correct that neither the “signature” crimes exception, nor the “motive” exception, nor any of the others justified mutual admissibility. Furthermore, I believe the division was clear that it would have found plain error requiring reversal but for its conclusion that the evidence of each crime was “simple and distinct,” affording separate presentation of each without confusion. Under Drew, this “simple and distinct” evidence alternative allows consolidation of two indictments for trial “even though [the other crimes] evidence might not have been admissible in separate trials.” Drew, 118 U.S.App.D.C. at 17, 331 F.2d at 91. The division on direct appeal concluded the argument was “sufficiently strong for consolidation on this basis that the trial court did not commit plain error in failing, absent objection, to sever the murder and assault charges.” Appendix at 1083.

Watson’s contention on collateral attack is uncontested: the government did not try the assault and murder cases separately, such that the evidence as to each was “simple and distinct”; nor were the cases kept separate in closing argument. Supra note 4. Significantly, the government in its supplementary brief and at oral argument to the en banc court did not dispute this contention. Nor does any of my colleagues. Indeed, the operating premise of everyone concerned is that the division on direct appeal was absolutely wrong in so concluding.

Watson contends that counsel on direct appeal was at fault — indeed, constitutionally ineffective — in failing to point out to the court a government misrepresentation on page 23 of its brief filed July 17, 1979, *1077which the division obviously accepted, that the “simple and distinct” exception applied on this record.5 I agree with my colleagues that, although counsel should have filed a reply brief pointing out the government’s misleading advocacy, counsel did do a job sufficiently professional to withstand constitutional attack. The division obviously knew the relevant caselaw and was sufficiently directed to the record to have to bear responsibility for its own plain error in resolving the appeal.

But this does not end the inquiry, for the plurality sweeps under the rug the other critical issue: the alleged ineffectiveness of trial counsel in failing to press for a severance of offenses before trial and, especially, as the trial progressed.

II.

A.

In a footnote, ante at 1057 n. 2, the plurality incorporates the discussion of alleged ineffectiveness of trial counsel in the division opinion on collateral attack. Watson, 508 A.2d at 88-91. There, the court found no ineffectiveness of trial counsel because the division on direct appeal had found no plain error:

By concluding on appellant’s direct appeal that there was no plain error, i.e., no error affecting substantial rights, we were saying in effect that in all probability the purported error would not have affected the outcome of his trial. This being so, counsel’s failure to notice such error at trial cannot be deemed “ineffectiveness” within the context of Strickland, since perforce there is no reasonable probability that absent counsel’s (assumed) failure the result of the proceeding would have been different.

Id. at 89 (citations omitted).

This analysis falls short.6 First, it is important to put the ineffectiveness issue in the context of harmless error and plain error standards of review. Theoretically, on an appellate record there may not be plain error requiring reversal for failure to grant a severance, even though there would have been harmful error, and thus reversible error, if counsel had called the grounds for severance to the trial court’s attention. Allen v. United States, 495 A.2d 1145, 1152 (D.C.1985) (en banc). More specifically, under the nonconstitutional standard for harmless error in Kotteakos v. United States, 328 U.S. 750, 66 S.Ct. 1239, 90 L.Ed. 1239 (1946), an appellate court, when addressing an error preserved for review, must reverse when it cannot say with fair assurance that the verdict was not substantially swayed by the error. Under the constitutional standard of Chapman v. California, 386 U.S. 18, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967), the appellate court must reverse unless it concludes the error was harmless beyond a reasonable doubt. In contrast, a court will not reverse for plain error unless “the error complained of is so clearly prejudicial to the complainant's substantial fights as to jeopardize the very fairness and integrity of the trial.” Allen, 495 A.2d at 1154. Roughly, therefore, the difference between *1078harmless error and plain error analysis is that the former mandates reversal if the error might have affected the outcome whereas, according to my colleagues, the latter mandates affirmance unless the error probably would have affected the outcome.

The plurality states that the Strickland test for prejudice from ineffectiveness of counsel appears to be the same as the prejudice inherent in plain error. As stated by my colleagues, however, reversal for plain error virtually presupposes that the error is outcome-determinative, whereas the Supreme Court in Strickland called such an approach “not quite appropriate” for ineffectiveness claims. 466 U.S. at 694, 104 S.Ct. at 2068. The tests for prejudice, therefore, are not the same.

There is, however, another critical distinction. We might find no plain error from the trial court’s failure sua sponte to grant a severance, were we to conclude that separate trials probably would not have affected the likelihood of conviction. We also might find no ineffectiveness of trial counsel if our review were to focus on whether the separate trials that counsel failed to achieve would have made any difference. But there is another way to look at ineffectiveness of trial counsel: an objection noting the error might have caused the trial court to grant the severance; but, even more importantly, if the trial court did err, counsel’s objection would have caused the appellate court to reverse, and grant separate trials, because the error was not harmless on a record that would not permit reversal for plain error. In short, the test for prejudice in Strickland arguably should be directed at the impact of counsel’s failure on the appellate standard of review, with the consequent prospects for reversal and a new trial, not directed solely at the question of probable acquittal as in plain error review.

I am aware that the Court in Strickland focused on whether “there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different” — on “whether there is a reasonable probability that, absent the errors, the factfinder would have had a reasonable doubt respecting guilt”— on whether “the decision reached would reasonably likely have been different absent the errors.” 466 U.S. at 694, 695, 696, 104 S.Ct. at 2068, 2069. I am not convinced, however, that when severance is the issue this emphasis on the reasonable probability of a different result would preclude factoring into the analysis on appeal the certainty of reversal and remand for new trials, without prejudicial other crimes evidence, had counsel preserved harmless error review. Accordingly, I conclude that the plurality’s equating of plain error and ineffectiveness-of-counsel review is defective.

B.

Even if, on this record, the inquiries as to prejudice would be the same for plain error and ineffectiveness-of-counsel review, there is still a critical flaw in the plurality analysis: by rule of court, resolution of a plain error inquiry on direct appeal does not preclude consideration of an ineffectiveness issue, addressing the same subject matter, on collateral attack.7 We have expressly urged counsel on direct appeal to eschew ineffectiveness claims, leaving them for collateral attack when a more complete record can be made. See Proctor v. United States, 381 A.2d 249, 252 & n. 4 (D.C.1977). It follows that, if an ineffectiveness claim on direct appeal does not necessarily preclude a collateral attack on the same ground, see id., we have implicitly announced a rule of law that plain error determinations on direct appeal do not necessarily preclude ineffectiveness claims on collateral attack either, even though they may concern the same trial defect.8 We *1079cannot be understood to have meant something different: that ineffectiveness claims should be reserved for collateral attack, except when the record is complete enough for resolution of that claim on direct appeal —in which case failure to do so there, or to prevail on the issue if raised as plain error, ends the matters as in res judicata or collateral estoppel, respectively. If we are going to draw appellants and their counsel into that intellectual quicksand pit, then we had better do so clearly and intelligently. The plurality opinion altogether ignores these difficult preclusion issues and, I fear, may have inadvertently made some unintended law.9

C.

As I see it, trial counsel’s failure properly to present the severance issue to the trial court was both deficient and prejudicial under Strickland. Minimally effective counsel would have objected to consolidation at the outset because of a lack of mutual admissibility (counsel objected only to joinder of defendants Watson and Joan Swearinger, not to joinder of the offenses). And, counsel should have objected as it became clear that offenses were not being tried separately within the “separate and distinct” rationale. In either case, the trial court’s failure to grant such a motion would have been reversible error on this record. Trial counsel’s default, resulting in continued joinder of the offenses to Watson’s severe prejudice for the reasons elaborated in Drew, see supra Part I, reflects more than a reasonable probability that, but for counsel’s unprofessional error, the result on this record would have been different. Strickland, 466 U.S. at 696, 104 S.Ct. at 2069. Either the trial court would have granted the severance if counsel had insisted, or this court would have found reversible error for failure to do so and ordered separate trials. In any event, I cannot agree that separate trials would not have affected the outcome; given the prejudice from the consolidation and the less-than-strong circumstantial evidence in the murder case, there is a reasonable probability that the results from separate trials *1080would have been different. Thus, I believe Watson is entitled to reversal and remand for further proceedings.

Respectfully, I dissent.

APPENDIX TO DISSENTING OPINION

District of Columbia Court of Appeals

Nos. 13796, 13815

Curtis L. Watson, Appellant, v. United States, Appellee.

Appeals from the Superior Court of the District of Columbia

(Hon. Alfred Burka, Trial Judge)

(Argued October 31, 1979 Decided November 5, 1979)

Before Kern, Ferren, and Pryor, Associate Judges.

MEMORANDUM OPINION AND JUDGMENT

Curtis L. Watson appeals his conviction of first degree murder, carrying a pistol without a license, assault with a dangerous weapon, two counts of burglary while armed, and two counts of assault with intent to kill while armed. As trial court error, he alleges (1) interruption of defense questioning of witnesses and improper comments on the conduct of the case by defense counsel; (2) improper consolidation of the indictment charging first degree murder with the indictment including the other counts, and (3) failure to grant a judgment of acquittal on the first degree murder charge for insufficient evidence of premeditation. We affirm.

I.

On October 18, 1976, a grand jury indicted appellant and Joan Swearinger on charges of armed assault against Richard Knight1 and an assault with intent to kill while armed against Robert Swearinger and Cynthia Durham.2 The charges arose out of two incidents which took place at 1926 Quincy Street, N.E. during the time that Robert Swearinger was living at that address. He was separated from his wife, Joan Swearinger, who was living with appellant.

On May 30,1976, Robert Swearinger was entertaining his friends, Cynthia Durham and Richard Knight, at the house. In the early evening, Swearinger and Durham went to a local store to shop for groceries, leaving Knight alone in the house. About 15 minutes later, appellant and Joan Swear-inger entered without knocking.3 Appellant pointed a gun at Knight, and told him that Robert Swearinger did not live in the house anymore. Appellant told Knight that he could leave “the hard way or the easy way.” Knight chose the easy way and left.

Six days later, on June 6, Robert Swear-inger and Cynthia Durham returned to the house between 1:30 and 2 a.m. after partying with friends. They soon went to bed, but Robert Swearinger was awakened by a crash which turned out to be the sound of the front door being kicked in. Moments later, Robert Swearinger saw appellant standing over him. Appellant fired three shots into the bed. The first hit Swearing-er in the head near his left eye and the other struck Durham in her right hand.

Appellant was tried on the charges arising from the two incidents in August 1977. A mistrial was declared, however, when the trial court discovered that several jurors had seen prejudicial publicity about a murder at 1926 Quincy Street.

Joan Swearinger had taken over the Quincy Street house after the shooting of *1081Durham and Robert Swearinger. She leased it to Mamie R. Reeves for one year beginning July 1, 1976. In late June, 1977, Ms. Reeves informed Joan Swearinger that she and her family would be moving out on June 28, but that her son Timothy would be staying in the house until July 1. On June 28, appellant left his family’s delicatessen, where he worked, carrying a pouch which usually contained a .357 magnum. Appellant explained that he had to “go over to the house and take care of some business.”‘ Appellant met Mr. and Mrs. Mabry at the house, who had inquired about renting it. A young man let appellant and the Mabrys into the house, and then went to the basement. The Mabrys later overheard a conversation between appellant and the young man in which the latter asserted his right to stay in the house until July 1 against appellant’s statement that “[w]e’re taking over the house.” Appellant’s co-worker, Robert Mitchell, also had heard him complain about the son of a tenant who would not leave so that the house could be cleaned up for a new occupant.

On June 29, Mamie Reeves returned to the Quincy Street house and found Timothy’s body on the basement floor. He had been shot through the head and stabbed repeatedly. He still had money in his pockets and there were no signs of a struggle.

Appellant was indicted on a charge of first degree murder for the killing of Timothy Reeves on October 13, 1977. Since the assault charges stemming from the previous incidents had not yet been retried, the government filed a motion (which appellant opposed) to consolidate the charges contained in the two indictments for trial. The trial judge granted the motion to the extent of consolidating the charges against appellant, but he severed the charges against Joan Swearinger.

Appellant was tried before a jury and convicted on counts contained in both indictments. He filed a new trial motion which included a claim that the trial court erred in consolidating the charges in the two indictments. The motion was denied. Appellant was sentenced to prison for terms of 20 years to life for first-degree murder, to run consecutively to a 5 to 15 year term for one burglary conviction; for a 5 to 15 year term for the other burglary conviction, to run consecutively to the first; for concurrent 5 to 15 year terms for the two assaults with intent to kill while armed; for concurrent terms of 40 months to 10 years for assault with a dangerous weapon; and for one year for carrying a pistol without a license. This appeal followed.

II.

First, appellant contends that the trial court’s interruptions of defense counsel and its comments on counsel’s performance were so prejudicial as to deny him a fair trial. In evaluating this contention, we must examine the cumulative impact of the trial judge’s participation to determine whether it could have prejudiced the jury’s view of the defendant’s case. See Khaalis v. United States, D.C.App., [408] A.2d [313] [357] (1979); Williams v. United States, D.C.App., 293 A.2d 484, 488 (1972); Jackson v. United States, 117 U.S.App.D.C. 325, 326, 329 F.2d 893, 894 (1964).

The trial court did interrupt defense counsel on a number of occasions to halt questioning that was irrelevant or without a sufficient basis (Tr. 269-70, 338, 375, 377-78), to warn against arguing with witnesses (Tr. 152-53, 155, 555), and to curtail repetitious questioning (Tr. 109-11, 252, 780, 937-40). On each of these occasions, intervention by the trial judge appears to have been justified. Although even valid interruptions may become prejudicial if they are made continuously, Young v. United States, 120 U.S.App.D.C. 312, 314, 346 F.2d 793, 795 (1965), such was not the case here. The interventions by the trial court were limited to keeping defense counsel’s questioning within acceptable bounds.4

*1082In its interruptions of counsel, however, the trial court occasionally did comment on his handling of the case.5 Although it is unrealistic to expect a trial judge to maintain unlimited patience with counsel’s errors, it must be remembered that “[i]n a jury case, a trial judge should exercise restraint and caution because of possible prejudicial consequences of the presider’s intervention.” Jackson, supra at 326, 329 F.2d at 894. The trial judge in this case, however, did not go so far in his comments as to suggest that defense counsel was incompetent or that Watson’s case was insubstantial. Moreover, the comments usually were directed at the prosecutor as well as defense counsel. Finally, the comments cited in note 5 supra arose in the context of an eight-day trial in which 27 witnesses testified. In short, “[t]he trial was long and the incidents relied on ... few. We must guard against the magnification on appeal of instances which were of little importance in their setting.” Glasser v. United States, 315 U.S. 60, 83 [62 S.Ct. 457, 471, 86 L.Ed. 680] (1942). Appellant’s argument that he was prejudiced by the trial court’s comments accordingly fails.

III.

Appellant next argues that the trial court erred in consolidating for a single trial the indictment charging the two assaults and the indictment charging first degree murder. Under Super.CtCr.R. 13, the trial judge “may order two or more indictments ... to be tried together if the offenses ... could have been joined in a single indictment.” Offenses may be joined in a single indictment under Super.Ct.Cr.R. 8(a) if they “are of the same or similar character or are based on the same act or transaction or on two or more transactions connected together or constituting parts of a common scheme or plan.” Our inquiry, then, focuses on whether the assaults and the murder were sufficiently related to warrant a joint trial or whether, instead, the trial court abused its discretion in consolidating the indictments. See Grant v. United States, D.C.App., [402] A.2d [405], [407] (1979).

At the outset, we note that the indictment charging appellant with the first two (assault) offenses also charged a co-defendant, Joan Swearinger. Thus, in moving to consolidate these cases for trial with the third (murder) charge against appellant, the government presented the court with questions of joining co-defendants as well as joining separate offenses. Alert to these questions, the trial court stated at a pretrial hearing:

[A]fter reading everything and listening to argument the last time, there is no question in my mind that the cases should be severed — they must be severed. It can be resolved one of two ways. Let Mr. Watson be tried for both offenses individually, and Ms. Swearing-er separately. Or try Ms. Swearinger and Mr. Watson for the murder case — I mean for the shooting. And then try Mr. Watson for the other [i.e., murder] matter individually. What does the Government have to say?
MR. BENNER: Well, the Government’s position is in view of the Court’s ruling, we would move to consolidate all the charges against Mr. Watson, and sever the charge against Ms. Swearinger.
THE COURT: That seems to be the logical way to do it.

Appellant’s counsel did not object; in fact, his memorandum in opposition to the government’s motion to consolidate *1083stressed the improper joinder of appellant and Ms. Swearinger, not the joinder of the offenses charged against appellant. At no time during trial, moreover, did appellant’s counsel seek a severance.6 Not until after appellant’s conviction did counsel, in a new trial motion, argue that the assault and murder charges should not have been joined in a single trial.

Under these circumstances, we conclude that we must apply the plain error standard of review. See Watts v. United States, D.C.App., 362 A.2d 706 (1976) (en banc). We agree with appellant that, on the facts here, consolidation of the murder charge with the two assault charges (which are properly joinable) cannot be justified by any of the five exceptions to the general bar against “other crimes” evidence set forth in Drew v. United States, 118 U.S.App.D.C. 11, 331 F.2d 85 (1964).7 However, if “the evidence of each crime is simple and distinct,” Drew also permits consolidation of two indictments for trial “even though [other crimes] evidence might not have been admissible in separate trials.” Id. at 17, 331 F.2d at 91. We conclude that the argument is sufficiently strong for consolidation on this basis that the trial court did not commit plain error in failing, absent objection, to sever the murder and assault charges. See Bell v. United States, D.C.App., 332 A.2d 351 (1975) (defendant initiated joinder of offenses in single indictment and never objected); Franklin v. United States, 117 U.S.App.D.C. 331, 330 F.2d 205 (1964) (where defendants made no objection to joint trial, misjoinder was cured by the verdict).

IV.

Appellant argues, finally, that the government’s evidence was not sufficient to support his conviction for first-degree murder and, therefore, that the trial judge erred in denying his motion for a judgment of acquittal at the close of the government’s case. He contends, more specifically, that the government failed to produce evidence of premeditation which would justify conviction of murder in the first degree.

Viewing the evidence, as we must, in a light most favorable to the government, see Byrd v. United States, D.C.App., 388 A.2d 1225, 1229 (1978), we find facts ample to permit a reasonable jury to find guilt beyond a reasonable doubt. See Jackson v. United States, D.C.App., 395 A.2d 99, 102 (1978). First, appellant had complained about Timothy Reeves’ presence in the Quincy Street house. Prior to the killing, appellant told his co-worker, Robert Mitchell, that he “would kick the son’s ass” if he did not leave the house. (Tr. 315.) Alvin Watson, appellant’s brother, heard him tell Joan Swearinger that “we’re going to get him out of that house.” (Tr. 362.) Such threatening remarks, coupled with some passage of time, are probative of the speaker’s deliberate and calculated intent to kill. See United States v. Peterson, 166 U.S.App.D.C. 75, 79, 509 F.2d 408, 412 (1974); United States v. Mack, 151 U.S. App.D.C. 162, 166, 466 F.2d 333, 337, cert. denied, 409 U.S. 952 [93 S.Ct. 297, 34 L.Ed. 2d 223] (1975 [1972]).

Robert Mitchell also testified that appellant left the family store on the night of June 28 carrying the pouch for his .357 magnum. As he left, appellant stated that he intended “to go over to the house and take care of some business.” (Tr. 321.) Although Mitchell did not actually see the gun that night, a reasonable juror could infer that appellant had carried a gun to the murder scene. The fact that a defendant carried a deadly weapon to the scene of *1084the crime is strong evidence of premeditation — of a calculated intent to kill.8 See Frendak v. United States, D.C.App., [408] A.2d [364] [371] (1979); O’Connor v. United States, D.C.App., 399 A.2d 21, 26 (1979); Peterson, supra [166 U.S.App.D.C.] at 79, 509 F.2d at 412.

The testimony of Dr. Brownlee, the medical examiner, also supports the jury’s conclusion that the homicide was planned. He testified that Reeves probably was asleep when he was attacked. (Tr. 488.) The stab wounds suggested that Reeves was lying on his back and there were no defensive wounds or other indications of a struggle. (Tr. 487.) The manner of killing can also provide a strong indication of whether the killer has premeditated the act. See Byrd, supra at 1230. The facts cited by Dr. Brownlee tend to indicate that the killer deliberately stabbed Timothy Reeves before Reeves had a chance to fight back.

Although the evidence supporting premeditation is circumstantial, the jury may infer premeditation from the surrounding facts and circumstances. See Jackson, supra; Calhoun v. United States, D.C.App., 369 A.2d 605 (1977); United States v. Harris, 140 U.S.App.D.C. 270, 435 F.2d 74 (1970) cert. denied, 402 U.S. 986 [91 S.Ct. 1675, 29 L.Ed.2d 152] (1971). The facts established here provide adequate support for the jury’s conclusion.

Accordingly, it is

ORDERED AND ADJUDGED, that the judgment on appeal here be, and it is hereby, affirmed.

Per Curiam.

For the Court:

/s/ Alexander L. Stevas

Alexander L. Stevas

Clerk

. Watson does not contend that he was confounded in the presentation of his defenses by consolidation of the indictments, so we need not consider the first type of prejudice.

. The term "pattern of action" used by the trial judge could encompass both the identity and the common scheme or plan criteria. I will discuss "identity” in the text below. The dissimilarities in the character of the crimes, coupled with the passage of more than one year between the assaults and the murder, belie any implication that the three crimes were part of a "common scheme or plan.” See Tinsley v. United States, 368 A.2d 531, 534 n. 4 (D.C.1976) (“The dissimilarities in the two homicides and their separation as to time of occurrence by nearly six months excludes any theory of common scheme or plan.”) Finally, no one suggests that the intent exception applies.

. Accord, United States v. Cyphers, 553 F.2d 1064 (7th Cir.), cert. denied, 434 U.S. 843, 98 S.Ct. 142, 54 L.Ed.2d 107 (1977) (proper for government to introduce evidence of an intended heroin purchase as evidence of need for a large sum of money as a motive in a robbery trial); United States v. Johnson, 542 F.2d 230 (5th Cir.1976) (prior conviction and outstanding arrest warrant were admissible in trial for threatening officers to show why defendant had a motive to try to escape).

. For example, in closing argument the prosecutor stated: "Mr. Reeves made the mistake of doing the same thing that Richard Knight and Robert Swearinger did. He wanted to stay in the house.” The prosecutor stated further: "He [Watson] went into that house on two different occasions and either killed or tried to kill sleeping people.” Finally, the prosecutor argued: "[T]he Government asks you to look at the circumstances of this case, to look at the similarity between all of the cases and the way they were committed and the motive. Why was each one of these people assaulted, maimed, murdered?"

. Absent a transcript, I cannot properly address oral argument on direct appeal.

. Assume counsel moves for a severance and the trial court errs in denying the motion. The defendant will prevail on appeal unless the government carries the burden of proving the error harmless. Next, assume counsel, in the same case, fails to move for a severance. In that instance, the defendant has the burden of proving either plain error or prejudice from ineffectiveness of counsel. As a consequence of this shift in the burden of proof and the standard for reversal, the defendant very likely will fail even though he or she would have received a new trial had counsel preserved the error. This does not necessarily mean, however, that an appellate court equates review of the alleged ineffectiveness of trial counsel with review for plain error. At least, we have never carefully reasoned and clearly held that in every case the Strickland test for prejudice (a "reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different”) is the same as the traditional test for finding plain error (“the error complained of is so clearly prejudicial to the complainant’s substantial rights as to jeopardize the very fairness and integrity of the trial"). In sum, it is not necessarily true that our respective standards of review for plain error and for ineffectiveness of trial counsel are congruent. If they were, by the plurality’s logic a finding of plain error would imply, in every case, ineffectiveness of counsel.

. Cf. Restatement (Second) of Judgments § 20(c) (1982) (personal judgment for defendant, although valid and final, does not bar action by plaintiff on same claim where by statute or court rule such judgment does not operate as a bar), quoted with approval in Rhema Christian Center v. D.C. Bd. of Zoning Adjustment, 515 A.2d 189, 196 (D.C.1986).

. In Proctor v. United States, 381 A.2d 249, 252 n. 4 (D.C.1977), we said:

*1079We do not intend, by our resolution of the ineffectiveness claim here or by our reference to appellant’s failure to utilize D.C.Code 1973, § 23-110, to preclude the appellant from raising the issue in a collateral proceeding under § 23-110 at a later time. Our decision resolves only the specific contentions before us on the trial record. If the appellant is able to furnish better evidentiary support for the present contentions based on material not provided in the record on appeal, or to frame and document contentions of ineffectiveness which we have not considered here, he is entitled to seek collateral relief in the Superior Court. See United States v. Brown, 155 U.S.App.D.C. 177, 179, 476 F.2d 933, 935 (1973), and DiAngelo v. United States, 406 F.Supp. 880, 884-85 (E.D.Pa.1976), [aff’d, 566 F.2d 1168 (3rd Cir.1977),] both of which dealt with the federal analogue of our § 23-110 proceeding, 28 U.S.C. § 2255 (1970). We express no view on whether appellant has a basis for such collateral relief in this case.

Accordingly, we acknowledged that appellant was not entitled to seek collateral relief for exactly the same alleged ineffectiveness of counsel, on the same record, that we considered on direct appeal. Arguably, therefore, if the ineffectiveness of trial counsel alleged on collateral attack concerned exactly the same formulation of the issue and the same portions of the record that served as the basis for plain error review on direct appeal, there may be a serious question of preclusion. Compare Arnold v. United States, 436 A.2d 1302, 1303 (D.C.1981) (all issues raised or capable of being raised on review of record on direct appeal are deemed precluded from further litigation by familiar rules governing issue preclusion) with Head v. United States, 489 A.2d 450 (D.C.1985) (where defendant fails to raise available challenge to conviction on direct appeal, defendant may not do so on collateral attack without showing cause for failure to do so and prejudice as a result). As indicated earlier, however, I am not satisfied that ineffectiveness of counsel and plain error present exactly the same issue. Furthermore, given Proctor and it progeny, I do not believe counsel should be expected to know, without clearer direction by this court, that advocacy of plain error on direct appeal could preclude collateral attack for ineffective assistance of trial counsel on the same subject matter.

. The government initially agreed that Watson cannot have been prejudiced by the failure of counsel on direct appeal to raise the ineffectiveness issue. Government Brief filed Sept. 30, 1986, at 53 n. 63. Later in the same brief, however, at 55-58, the government urged the approach adopted by the majority, ante at 1057 n. 2, incorporating Watson, 508 A.2d at 88-91: that this court’s plain error review of the joinder issue on direct appeal precludes de novo consideration, on the same record, of trial counsel’s ineffectiveness as to the same issue.

. Specifically, the indictment charged violations of D.C.Code 1973, § 22-1801(a), burglary while armed; § 22-1801(a), burglary; and § 22-502, assault with a dangerous weapon.

, In relation to this incident, the indictment included violations of D.C.Code 1973, § 22-1801(a), burglary while armed; § 22-1801(a), burglary; § 22-501, assault with intent to kill while armed; § 22-501, assault with intent to kill; and § 22-502, assault with a dangerous weapon.

.Joan Swearinger had inherited the Quincy Street house from her parents, and had quar-relled with Robert Swearinger over possession of the property.

. One particular incident, however, is worthy of an additional comment. (Tr. 437.) In his cross-examination of a police officer assigned to the Mobile Crime Laboratory, defense counsel asked whether the officer had determined whether Reeves’ assailant was right- or left-handed. The trial court dismissed the witness because defense counsel was persisting in ask*1082ing questions which witnesses were not qualified to answer. Although termination of the defendant’s cross-examination of a government witness may be an overly-harsh sanction for improper questioning, defense counsel did not object to the dismissal of the witness. The termination of questioning did not constitute plain error, see Watts v. United States, D.C.App., 362 A.2d 706 (1976) (en banc).

. For example: "If you don’t get something relevant, and I mean immediately, this witness will be excused permanently. You lawyers are playing games in here now. I will not have that." (Tr. 372.) "The Government has been carrying this case on too far also. You all are going to conduct this case in a lawyer-like manner.” (Tr. 377-78.) "I'm warning you_ One more interruption, one more multiple question, one more failing to let him complete an answer ... and cross-examination is terminated. I’m giving you this one last warning.” (Tr. 939.)

. At oral argument counsel for appellant argued, apropos of his judicial misconduct allegation, see Part II supra, that trial counsel may have been too intimidated to seek a severance once it became clear that there were obstacles to defending three charges at one trial. We reject this contention.

. The exceptions are:

(1) motive, (2) intent, (3) the absence of mistake or accident, (4) a common scheme or plan embracing the commission of two or more crimes so related to each other that proof of the one tends to establish the other, and (5) the identity of the person charged with the commission of the crime on trial. [Drew, supra at 16, 331 F.2d at 90 (footnote omitted).]

. This is true even though the weapon carried to the scene may not be the same weapon that caused death. The testimony here indicated that Timothy Reeves was shot with a .25 caliber bullet.