Herrera v. LeMaster

F I L E D
United States Court of Appeals
Tenth Circuit
UNITED STATES COURT OF APPEALS
September 20, 2005
TENTH CIRCUIT
Clerk of Court

RUBEN R. HERRERA,

Petitioner-Appellant, No. 03-2164
v. (D. of N.M.)
TIM LEMASTER, Warden, New (D.C. No. CIV-97-542-JP/LFG)
Mexico State Penitentiary,
ATTORNEY GENERAL FOR THE
STATE OF NEW MEXICO,

Respondent-Appellees.

ORDER AND JUDGMENT *

Before HENRY , LUCERO , and TYMKOVICH , Circuit Judges.

This is the fourth appeal of Ruben Robert Herrera’s 1982 conviction for

first degree murder outside a nightclub in Las Vegas, New Mexico. The question

in this appeal is whether on remand the district court correctly applied Brecht v.

Abrahamson, 507 U.S. 619 (1993), in denying Herrera’s petition for a writ of

habeas corpus. Under Brecht, a conviction based on the admission of evidence in

*
This order and judgment is not binding precedent, except under the
doctrines of law of the case, res judicata, and collateral estoppel. The court
generally disfavors the citation of orders and judgments; nevertheless, an order
and judgment may be cited under the terms and conditions of 10th Cir. R. 36.3.
violation of the Constitution will be reversed if the evidence had a “substantial

and injurious effect or influence on the jury’s verdict.” Herrera argues the

introduction at trial of a bullet, bullet clip, and bullet casing seized under a

warrant deemed to lack probable cause improperly influenced the jury’s verdict.

Because we do not believe that Herrera has met the Brecht standard, we affirm.

Our jurisdiction is pursuant 28 U.S.C. §§ 1291 and 2253.

I. BACKGROUND

A. Procedural History

This appeal follows three previously reported decisions. In 1985, the New

Mexico Supreme Court affirmed Herrera’s murder conviction on direct appeal,

concluding that despite the admission at trial of evidence obtained pursuant to a

search warrant that lacked probable cause, the error was harmless. State v.

Herrera, 694 P.2d 510 (N.M. 1985) (Herrera I). Herrera filed a habeas corpus

petition in 1998, which was denied by the district court, also on the ground that

the error was harmless. On appeal, we reversed and remanded for further

proceedings so that the district court could apply the correct standard of review

for harmless error under Brecht. Herrera v. Lemaster, 225 F.3d 1176 (10th Cir.

2000) (Herrera II). An en banc panel of this court reaffirmed the application of

Brecht as the proper standard for habeas review in Herrera v. Lemaster, 301 F.3d

-2-
1192 (10th Cir. 2002) (Herrera III). Herrera appealed that decision to the United

States Supreme Court, which denied certiorari review.

On remand, the district court referred the case to a magistrate judge for a de

novo review of the record. The magistrate judge concluded that admission of the

illegally seized evidence at trial was harmless under Brecht. Herrera filed

objections. The district court adopted the magistrate judge’s recommended

disposition, and denied the petition. Herrera appealed.

B. Factual History

The factual background of this appeal is described in previous opinions,

and we briefly summarize only the relevant facts here. On November 5, 1982, the

victim, Leroy Lovato, encountered Ruben Herrera at the El Nido nightclub in Las

Vegas, New Mexico. A brief dispute occurred between Lovato and Herrera after

Lovato accused Herrera of harassing Lovato’s cousin. Herrera told Lovato to

mind his own business and then left El Nido by himself.

About one hour later, Lovato and a friend, Phillip Arellanes, left El Nido.

They intended to join two women, Denise Martinez and Debra Farrell, outside the

club. As the four were talking on the sidewalk near Lovato’s car, Arellanes

noticed Herrera crossing the street and heading toward the group. Arellanes

commented to Lovato that Herrera was approaching. The two women, fearing an

altercation, turned to leave.

-3-
 Sensing danger, Arellanes told Herrera that neither he nor Lovato wanted

any trouble. Herrera ignored him and turned toward Lovato. Lovato reached out

his hand, apparently to shake Herrera’s hand. Instead of returning the gesture,

Herrera pulled out a gun, shot Lovato in the neck (he died that night at a

hospital), and then pointed the gun at Arellanes. Arellanes begged for his life.

Herrera threatened Arellanes and warned him not to call the police. Herrera then

crossed the street and fled the scene. Although neither Martinez nor Farrell

testified to seeing Herrera with a weapon aimed on Arrellanes, after the report of

the gun shot, both turned to see Lovato lying on the ground bleeding. Herrera

was walking across the street with his hand to his side. Farrell testified she saw

Herrera sticking a gun into his pants. Martinez saw Herrera cross the street but

did not see him with a gun. Arellanes and the two women then went to a nearby

apartment to call for help.

Two other witnesses, Robert Arellanes and Julian Bustamante, also saw

Herrera leave the scene. Both Robert Arellanes, a cousin of Phillip Arellanes,

and Bustamante knew Herrera; Bustamante had been a high school acquaintance.

While their view of the actual shooting had been partially blocked by a parked

car, Robert Arellanes saw Herrera cross the street after the shooting and place a

gun into his pants. Although Bustamante saw Herrera crossing the street holding

his side, he did not see a gun. Both witnesses had a clear view of Herrera and

-4-
later were readily able to identify him to police. They observed no one else near

the scene of the shooting.

After leaving to report the shooting, Phillip Arellanes and the two women

returned to the murder scene. The police arrived and conducted an investigation.

Investigators found a spent 9 millimeter casing near Lovato’s body but did not

find a gun. In addition, the police arrested a “hysterical” Arellanes for interfering

with the investigation after he became disruptive and ignored repeated warnings

to relax. Later at the police station, officers identified Herrera as being at the

scene after obtaining his name from Arellanes. The police then presented a photo

array to Arellanes, Farrell and Martinez, each of whom positively identified

Herrera.

Three hours after the shooting, the police arrested Herrera in the vicinity of

his residence. He was taken to the Las Vegas police station and questioned.

Investigators also took samples from his hands for use in a gun powder residue

test.

The police then obtained a warrant to search Herrera’s residence. During

the search they found a bullet, clip, and a bullet casing from a 9 millimeter gun.

Ballistics testing later determined that the spent casing found at the scene of the

crime and the casing found in Herrera’s residence were fired from the same gun.

-5-
 The government charged Herrera with first-degree murder and aggravated

assault.

II. DISCUSSION

Before turning to the question of harmless error, we must address two

issues that have stalked this case for a number of years, and greatly complicate

our review. The first centers on the search warrant police obtained to enter

Herrera’s home the day of the shooting. Although we do not resolve the issue in

this appeal, we note that the record strongly suggests that the Las Vegas police

did not violate Herrera’s Fourth Amendment rights in obtaining a search warrant

and searching his home the night of the murder.

That observation brings us to the second complication: Law of the case. In

Herrera III, the state asked us to revisit the New Mexico Supreme Court’s finding

that the search warrant violated the Fourth Amendment. We declined this

invitation, finding that the argument had not been timely raised. Herrera III, 301

F.3d at 1196–97.

In this appeal, the government asks us to disregard our prior determination.

We again decline the invitation. Consequently, our review of the evidence of

conviction will be through the lens of Brecht v. Abrahamson, 507 U.S. 619

(1993).

-6-
 A. Law of the Case

The state argues that we need not address the effect of the illegally seized

evidence on the outcome of the trial, because the police acted in good faith in

obtaining the warrant. Under the so-called good faith exception to the

exclusionary rule set forth in United States v. Leon, 468 U.S. 897 (1984),

evidence obtained pursuant to a constitutionally defective search warrant is

admissible at trial if the officers executing the search warrant reasonably relied on

the warrant and there is no evidence the officers mislead the magistrate issuing

the warrant. Id. at 920–21. Thus, the government urges us to ignore the prior

determinations that the warrant was illegal and that the evidence seized pursuant

to it should be suppressed.

This issue had an unfortunate history in the New Mexico state courts.

Although the United States Supreme Court decided Leon after Herrera’s trial, it

was precedent during the direct appeal to the New Mexico Supreme Court. When

the New Mexico Supreme Court first considered Herrera I in 1985, the state

argued the good faith exception to the exclusionary rule set forth in Leon need not

be considered in light of the substantial evidence supporting probable cause for

issuance of the warrant. The court agreed and determined that its “disposition of

th[e] case [was] such that [it did] not find it necessary to discuss the good faith

exception at [that] time.” Herrera I, 694 P.2d at 514. The court went on to find,

-7-
however, that “[t]here were insufficient facts to show probable cause to search the

premises described” in the search warrant, so “the search was illegal under both

state and federal [law].” Id. The court nonetheless went on to conclude that the

admission of evidence seized pursuant to the warrant—the bullet, clip and

casing—was harmless error. Id.

The state tentatively raised the Leon issue to the Herrera II panel in the

first federal appeal, which did not address the issue. Finally, in its brief before

the Herrera III en banc panel, the government raised Leon as a potentially

dispositive ground for affirmation and set forth its argument in some detail. 301

F.3d 1192, 1196–97. However, since the state had failed to raise this argument in

the district court, we declined to consider it for the first time on appeal. Id.

Nonetheless, in spite of our refusal to hear the state’s Leon argument in Herrera

III, the state again raised the argument before the magistrate judge on remand.

The magistrate judge determined the question was beyond the scope of our

remand instructions in Herrera III and declined to address it.

The law of the case doctrine, although discretionary, holds that prior

judicial decisions on rules of law govern the same issues in subsequent phases of

the same case. See Homans v. City of Albuquerque, 366 F.3d 900, 904 (10th Cir.

2004). When an appellate court rules on an issue of law, that ruling governs

subsequent appeals; after a remand, the “decision of the first appeal establishes

-8-
the law of the case to be followed on the second appeal.” Moore’s Federal

Practice, § 134.22[2][b]. Accordingly, our en banc decision in Herrera III that

Leon does not apply is the law of the case and governs our appeal here. Were we

writing on a clean slate, we would be tempted to remand for a determination of

the application of Leon. But, based on the procedural history before us now, we

again must “decline the state’s invitation to address the applicability of Leon.”

Herrera III, 301 F.3d at 1197.

B. Brecht Analysis

1. Standard of Review

“[W]e consider the question of harmless error de novo under the general

standard established for habeas cases in Brecht.” Willingham v. Mullin, 296 F.3d

917, 931 (10th Cir. 2002). A state court has not previously heard Herrera’s

habeas claim on the merits, so the framework of 28 U.S.C. § 2254 does not apply.

Instead, we examine the district court’s legal conclusions de novo. Turrentine v.

Mullin, 390 F.3d 1181, 1189 (10th Cir. 2004). Since our Herrera III remand was

solely for the purpose of examining the evidence under Brecht, here we need

focus solely on the district court’s record review.

The Supreme Court has told us to start with the wrongly admitted evidence.

The standard of review centers on whether the tainted evidence had a “substantial

and injurious effect or influence [on] the jury’s verdict.” Brecht, 507 U.S. at 637.

-9-
The Supreme Court elaborated in determining the scope of judicial review under

Brecht two years after the decision in O’Neal v. McAninch, 513 U.S. 432 (1995).

There, the Court instructed courts in looking at the record as a whole: If an

appellate court has “grave doubt” about the influence of the offending evidence

on the verdict, i.e., “in the judge’s mind, the matter is so evenly balanced that he

feels himself in virtual equipoise as to the harmlessness of the error,” the court

should treat the error as having a substantial and injurious effect. Id. at 435; see

also Webber v. Scott, 390 F.3d 1169 (10th Cir. 2004) (applying O’Neal).

With these standards in mind, we turn to the evidence at trial.

2. Evidence

Herrera contends the admission of the bullet, clip, and casing had a

substantial and injurious effect on the jury because it was the only physical

evidence introduced in the state’s case in chief and allowed the jury to resolve a

number of factual disputes in its favor. Herrera’s theory of defense attempted to

point the finger at Phillip Arellanes. According to this argument, Herrera claims

that Phillip Arellanes was, in fact, the shooter. Arellanes killed Lovato,

concealed the murder weapon from the other witnesses (or they conspired to lie at

trial), and then arranged with his cousin to plant the bullet evidence in Herrera’s

bedroom. Herrera argues his theory of defense was fatally compromised by the

-10-
fact the bullet casing from Herrera’s home ballistically matched the casing found

at the murder scene.

In reviewing this theory against the evidence, the magistrate judge

dismissed it as lacking “any convincing support” in the record. And the district

court, in reviewing the evidence again, concluded that “a careful de novo review

indicates that Mr. Herrera’s speculation and supposition are just that. Mr. Herrera

presents no record support for his theory, and the reason for that is obvious: there

is none. The recorded admissible testimony presents overwhelming evidence of

Mr. Herrera’s guilt.”

After conducting our own independent review of the record in this case,

including the tape recordings of the trial proceedings, we agree with the district

court. The admission of the bullet, casing, and clip found during the search of

Herrera’s residence do not cause us “grave doubt” as to the harmlessness of the

error. Based on the testimony of five eyewitnesses, three of whom saw Herrera

with a gun at the scene, and all of whom positively identified him shortly after the

murder, we conclude that the tainted evidence did not substantially influence the

jury’s determination that Herrera committed the murder.

We summarize the evidence below.

Eyewitness Testimony. The state’s case-in-chief was anchored by the five

eyewitnesses to the events immediately surrounding the shooting. As we

-11-
discussed above, Phillip Arellanes, the state’s first witness, saw the shooting from

point blank range, and two witnesses—Farrell and Robert Arellanes—saw Herrera

holding a gun within seconds of the shooting. Although Phillip Arellanes was the

only person who witnessed the actual shooting, his testimony was uncontroverted

at trial. Farrell and Martinez, moreover, testified regarding the events leading up

to and after the shooting. Together, the three witnesses’ testimony supplies a

motive for the shooting (a bar room encounter), the opportunity for the shooting

(Herrera was the only one to confront Lovato), and the consequences of the

shooting (a bleeding Lovato and a fleeing Herrera). Each positively identified

Herrera the day of the crime. The witnesses’ statements were uncontroverted.

In addition to the two women accompanying the victim, the other two

eyewitnesses were within sight of the shooting and testified regarding the events

immediately after the shooting. Those witnesses, Bustamante and Robert

Arellanes, testified to seeing Herrera acting suspiciously when leaving the scene

of the shooting. They both were able positively to identify him. Robert Arellanes

saw him with a gun. Their testimony was also uncontroverted at trial.

Herrera argues in the face of this evidence that Phillip Arellanes was the

actual gunman. To support this theory, at trial Herrera’s counsel cross-examined

Phillip Arellanes and accused him of being the shooter, despite no testimony that

Arellanes ever had a gun in his possession. Herrera also offered the testimony of

-12-
a friend, Richard Roybal, who claimed he was parked outside of Herrera’s

residence waiting to pick up Herrera at dawn on the day of the shooting. He

claimed that, while waiting for Herrera, he witnessed Robert Arellanes’s cousin

get into a car parked in the alley next to Herrera’s residence and leave the area.

The implication of this testimony was that Arellanes’s cousin planted the bullet

evidence in Herrera’s residence. Complicating this argument, however, was the

fact that, prior to trial, Roybal never relayed his story to the police or to anyone

involved in the case.

Finally, Herrera tried to create reasonable doubt by pointing out

inconsistencies and bias in the testimony of the eyewitnesses. For example, trial

counsel sought to demonstrate that it was impossible for Herrera to shoot Lovato,

turn and threaten Arellanes, and flee across the street all in the few seconds

between when Farrell and Martinez heard the shot and when they turned around.

Herrera also points to several inconsistencies between statements witnesses gave

prior to the trial and their testimony during the trial. Farrell, for instance,

testified at a preliminary hearing that she saw Herrera put something down his

pants as he crossed the street, but she testified at trial she saw him put a gun

down his pants. None of these examples, however, seriously undermine the

common story told by all of the eyewitnesses. As the district court observed, and

-13-
we agree, these discrepancies are the normal inconsistencies that flow from

witness recollections of a dramatic nighttime murder.

Expert Testimony. Herrera attempted to counter the eyewitness testimony

with expert testimony showing he had not recently fired a gun. Anticipating this

evidence, the state presented expert testimony to explain the result of the gunshot

residue analysis conducted on samples taken from Herrera’s hands after the

shooting. According to the state’s expert, the test was inconclusive—he could not

say whether Herrera had or had not fired a gun recently. The expert, however, did

testify that the absence of gunshot residue on Herrera’s hands could be explained

by the fact that such residue wears off easily and the time lapse between the

shooting and the test.

Herrera’s expert agreed that the samples taken from Herrera’s hands three

hours after the shooting showed no sign of gunpowder residue. In addition, he

explained how Herrera’s hands could not have been washed in close temporal

proximity to the test. However, in the end, both experts testified that gun shot

residue wears off rapidly. And both agreed the residue test here was at best

inconclusive.

Lack of Substantial, Injurious Effect. In short, Herrera has not presented a

plausible interpretation of the evidence that would allow us to find that admission

of the bullet, clip, and casing had a substantial, injurious effect by allowing the

-14-
jury to resolve these factual disputes against him. To the contrary, these

arguments lack record support and can be readily resolved against him without

reference to the bullet, clip, and casing.

We have held in other contexts that the emphasis given at trial and at

closing to evidence admitted in violation of the Constitution is relevant to

harmless error analysis. United States v. Summers, 414 F.3d 1287, 1304 (10th

Cir. 2005); United States v. Lauder, 409 F.3d 1254, 1261–62 (10th Cir. 2005).

Contrary to Herrera’s claims, the physical evidence was a relatively small part of

his trial. It was briefly mentioned in the prosecutor’s opening statement but only

after a lengthy exposition of what the anticipated eyewitnesses would say. The

casing evidence came in on the fourth day of a five-day trial through a police

witness. Similarly, during closing, as the district court noted, the references to

the evidence “amounted to only a small fraction” of the prosecutor’s final

argument.

On the other hand, the prosecution’s case was a powerful collection of

eyewitness testimony: it was unchallenged that five witnesses placed Herrera at

the scene of the murder, one witness saw him pull the trigger, and two other

witnesses saw him with a gun. While we do not dismiss the physical evidence as

having no significance to a jury, we cannot conclude the evidence itself and the

emphasis placed on it by the state in this case “tip the scales” such that its

-15-
admission had a substantial, injurious effect on the jury’s verdict. Brecht, 507

U.S. at 637. In sum, the evidence of guilt in this case was overwhelming.

CONCLUSION

We therefore AFFIRM the decision of the district court.

Entered for the Court

Timothy M. Tymkovich
Circuit Judge

-16-