v. Delgado

Opinions of the Colorado Supreme Court are available to the
public and can be accessed through the Judicial Branch’s homepage at
http://www.courts.state.co.us. Opinions are also posted on the
Colorado Bar Association’s homepage at http://www.cobar.org.

ADVANCE SHEET HEADNOTE
September 30, 2019

2019 CO 82

No. 17SC29, People v. Delgado—Inconsistent Findings—Mutually Exclusive
Verdicts—Plain Error—Retrial.

In this opinion, the supreme court considers whether the elements of

robbery and theft from a person are inconsistent, such that a defendant cannot be

convicted of both crimes. The supreme court further considers whether, in the

case of mutually exclusive guilty verdicts, the proper remedy is a new trial.

The supreme court holds that when an essential element of one crime

negates an essential element of another crime, guilty verdicts for those two

offenses are mutually exclusive, and the defendant cannot be convicted of both.

Here, the jury convicted the defendant of robbery and theft from a person.

Robbery is the unlawful taking of an item with force, while theft from a person is

the unlawful taking of an item without force. Because an element of

robbery—with force—and an element of theft from a person—without

force—negate one another, the defendant cannot be convicted of both crimes.
 The supreme court concludes that the obvious inconsistency between the

elements of robbery and theft from a person renders the error plain. Because it is

impossible to determine what the jury decided—whether the defendant acted with

or without force—the supreme court further concludes that the proper remedy is

a new trial. Thus, the supreme court affirms the court of appeals’ decision to grant

a new trial.
 The Supreme Court of the State of Colorado
2 East 14th Avenue • Denver, Colorado 80203

2019 CO 82

Supreme Court Case No. 17SC29
Certiorari to the Colorado Court of Appeals
Court of Appeals Case No. 13CA2024

Petitioner:

The People of the State of Colorado,

v.

Respondent:

Johnny Guillermo Delgado.

Judgment Affirmed
en banc
September 30, 2019

Attorneys for Petitioner:
Philip J. Weiser, Attorney General
Kevin E. McReynolds, Assistant Attorney General
Denver, Colorado

Attorneys for Respondent:
Megan A. Ring, Public Defender
Elizabeth Griffin, Deputy Public Defender
Denver, Colorado

JUSTICE HOOD delivered the Opinion of the Court.
JUSTICE BOATRIGHT dissents, and CHIEF JUSTICE COATS joins in the
dissent.
¶1 Unable to gain access to a closed bar by banging on the door, R.B. headed

elsewhere. He didn’t get far. As he began to leave, he was knocked unconscious

from behind. He awoke with his belongings gone and the police in pursuit of a

man seen fleeing the scene. That man was Johnny Delgado.

¶2 The police caught Delgado, who was eventually convicted of both theft from

a person and robbery based on a single taking. But theft from a person is the

unlawful taking of an item without force, and robbery is the unlawful taking of an

item with force. Thus, based on the elements, it appears that Delgado was both

convicted and absolved of taking R.B.’s belongings without force. And he was both

convicted and absolved of taking R.B.’s belongings with force.

¶3 Can these verdicts be legally and logically reconciled? Like the division of

the court of appeals below, we conclude that they cannot. Elements of the two

convictions are mutually exclusive. It’s impossible for Delgado to have unlawfully

taken items from R.B. by force and also by means other than force when there was

only one taking alleged.

¶4 So, what’s the remedy? The People argue that, even if the verdicts are

mutually exclusive, the cure is to maximize the convictions by throwing out the

lesser theft-from-a-person conviction. Delgado counters that double jeopardy

requires striking both convictions. The division below took a middle ground and

concluded that, here, the solution is a new trial.

2
¶5 Again we agree with the division. We hold that, because the elements of

theft from a person and robbery negate each other when they are based on a single

taking, a defendant may not be convicted of both offenses. Because such mutually

exclusive convictions leave us without a meaningful way to discern the jury’s

intent, the proper remedy is a new trial. Accordingly, we affirm the judgment of

the court of appeals.

I. Facts and Procedural History

¶6 Late one evening, R.B. banged on the door of a closed bar, apparently to gain

entry. While unsuccessful, he managed to disturb the staff inside the bar. They

called the police. Shortly after banging on the door, R.B. was attacked from

behind, and he lost consciousness.

¶7 Within minutes, an officer arrived at the bar. As he arrived, that officer saw

a man—later discovered to be Delgado—rummaging through the pockets of R.B.,

who was lying down. Delgado stood and walked away carrying several items.

The officer yelled for Delgado to stop, but Delgado bolted down an alley instead.

Other officers gave chase. During the chase, Delgado dropped several items. The

officers quickly caught and arrested Delgado, and they found R.B.’s phone, wallet,

and keys on the ground nearby.

3
¶8 Delgado was subsequently charged with robbery and theft from a person.

A jury convicted him on both counts. (The jury was unable to reach a verdict as

to a misdemeanor assault charge, so the district attorney dismissed that charge.)

¶9 A division of the court of appeals reversed. Seizing on language from our

decision in People v. Frye, 898 P.2d 559 (Colo. 1995), it reasoned that essential

elements of robbery and theft from a person logically negate each other. People v.

Delgado, 2016 COA 174, ¶¶ 19–21, 410 P.3d 697, 700. More specifically, the division

reasoned that because “force” has the same definition in both the robbery and

theft-from-a-person statutes, Delgado couldn’t have taken R.B.’s belongings both

“by the use of force” (as required for robbery) and “by means other than the use

of force” (as required for theft from a person). See id. at ¶ 21, 410 P.3d at 700. The

division additionally concluded that the error was plain because the rule

preventing mutually exclusive convictions is “well-established” and there could

“be no clearer example of inconsistency” than between these elements of theft

from a person and robbery. Id. at ¶ 23, 410 P.3d at 701.

¶10 As for the remedy, the division disagreed with both the People and Delgado.

See id. at ¶¶ 24, 27, 410 P.3d at 701–02. The People asked the division to maximize

the jury’s verdicts by sustaining the robbery conviction, id. at ¶ 27, 410 P.3d at 701,

while Delgado requested that the division throw out both convictions, id. at ¶ 34,

410 P.3d at 702. The division reasoned that it couldn’t possibly know what the

4
jury was thinking because it “expressed irreconcilable findings,” making

maximization inappropriate. Id. at ¶ 32, 410 P.3d at 702. And it equally disagreed

with Delgado’s contention that he was acquitted of both crimes because each

verdict contained an affirmative finding that necessitated acquittal on the other

count. Id. at ¶ 34, 410 P.3d at 702. Rather, the division noted that such reasoning

could just as easily lead it to conclude that both verdicts included findings of

conviction, rather than acquittal. Id. So, it concluded that the correct remedy is a

new trial. Id. at ¶ 33, 410 P.3d at 702.

¶11 We granted the People’s petition for certiorari.1

II. Analysis

¶12 After briefly discussing the standard of review, we address the proper

framework for dealing with mutually exclusive guilty verdicts. We conclude that,

when essential elements of two guilty verdicts logically negate each other, they are

mutually exclusive. Here, the robbery and theft-from-a-person convictions do

1 We granted certiorari to review the following issues:
1. Whether the court of appeals erred in concluding the jury’s
verdicts finding defendant guilty of both robbery and theft were
inconsistent under People v. Frye, 898 P.2d 559 (Colo. 1995).
2. Whether the court of appeals erred by reversing for a new trial for
inconsistent jury verdicts, instead of maximizing the jury verdicts
by affirming the most serious conviction.

5
exactly that. Therefore, Delgado may not be convicted of both. We then analyze

whether this error was plain. We conclude that it was, given the stark differences

between the elements of theft from a person and robbery. Finally, we examine the

proper remedy for mutually exclusive guilty verdicts. Because it is impossible to

know exactly what the jury intended, the remedy is a new trial.

A. Standard of Review

¶13 Whether verdicts are mutually exclusive is a question of law. Therefore, we

review this issue de novo. See Lucero v. People, 2012 CO 7, ¶ 19, 272 P.3d 1063, 1065

(citing Montez v. People, 2012 CO 6, ¶ 7, 269 P.3d 1228, 1230). Unpreserved

constitutional errors, such as the one here, are reviewed for plain error. See

People v. Miller, 113 P.3d 743, 748–49 (Colo. 2005). An error is plain when it is

“obvious and substantial,” id. at 750 (quoting People v. Stewart, 55 P.3d 107, 120

(Colo. 2002)), and “so undermine[s] the fundamental fairness of the trial itself as

to cast serious doubt on the reliability of the judgment of conviction,” Hagos v.

People, 2012 CO 63, ¶ 18, 288 P.3d 116, 120 (quoting Wilson v. People, 743 P.2d 415,

420 (Colo. 1987)).

B. Mutually Exclusive Guilty Verdicts

¶14 This court has not yet directly confronted how to analyze mutually

exclusive guilty verdicts. We made it clear in Frye that it isn’t necessary for a guilty

verdict and a not-guilty verdict to be consistent. See 898 P.2d at 571 (holding that

6
“consistency in [guilty and not-guilty] verdicts is unnecessary”). But as the parties

and division below note, in Frye, we at least suggested how cases such as the one

before us now might be handled. See id. at 569 n.13 (observing that “courts are

generally uniform in their agreement” that mutually exclusive guilty verdicts

“should not be sustained”).

¶15 The facts of Frye, though disturbing, are straightforward. Ronald Frye

raped the victim, referred to as Doe, while pointing a gun at her head. Id. at 561.

Frye was charged with first degree burglary, second degree burglary, sexual

assault in the first degree, and menacing with a deadly weapon. Id. at 560–61.

Before the jury deliberated, the trial court instructed the jury on the lesser included

offense of second degree sexual assault. Id. at 564. The jury ultimately found Frye

guilty of that offense and menacing with a deadly weapon, which implied that

Frye was not guilty of first degree sexual assault. Id.

¶16 The problem was that sexual assault in the first degree and menacing with

a deadly weapon included similar elements. Sexual assault in the first degree

required causing the submission of the victim “through the actual application of

physical force, physical violence, or . . . by threat of imminent death, serious bodily

injury, [or] extreme pain.” Id. at 563. Menacing with a deadly weapon required

placing the victim in “fear of imminent serious bodily injury.” Id. We concluded

7
“that only by a most strained analysis could the two guilty verdicts be held to be

consistent.” Id. at 565.

If the jury had believed that Frye used a gun to cause Doe’s
submission, it would logically have found him guilty of first degree
sexual assault. Therefore, a verdict of not guilty of first degree sexual
assault but guilty of second degree sexual assault suggests that the
jury found that Frye did not use a gun to cause Doe to submit. Such
a finding would be at odds with the guilty verdict for menacing with
a deadly weapon because there is no evidence that Frye used the gun
other than in the course of the sexual assault.

Id. (citation omitted). To find consistency in the two verdicts, we noted that we

would have had to determine that the jury didn’t believe Doe’s testimony that Frye

used a gun during the commission of the sexual assault, but did believe that Frye

used it at some other point. See id. at 566. We rejected such an “implausible

explanation” and concluded that the verdicts were inconsistent. Id. Despite this,

we followed the federal rule and allowed both verdicts to stand so long as the

evidence supported both convictions because it was impossible to determine why

the jury convicted on one count and acquitted on the other. Id. at 570–71.

¶17 The People argue that we should analyze the inconsistency here the same

way. But Frye arose in a different setting: inconsistent guilty and not-guilty

verdicts, not mutually exclusive guilty verdicts. See id. at 564–66. In fact, we stated

explicitly that Frye was not a case where a defendant had been convicted of two

crimes, certain elements of which are mutually exclusive. See id. at 569 n.13.

8
Instead, we noted a “general agreement” that legally and logically inconsistent

guilty verdicts “should not be sustained.” Id.

¶18 This dicta has taken root. We have at least alluded to it in a subsequent

opinion, see Candelaria v. People, 148 P.3d 178, 183 (Colo. 2006), and the court of

appeals has applied it, see, e.g., People v. Weare, 155 P.3d 527, 529 (Colo. App. 2006).

Moreover, a majority of other states still apply the same or a similar rule. See, e.g.,

State v. Chyung, 157 A.3d 628, 636, 643 (Conn. 2017) (holding that convictions of

reckless manslaughter and specific intent murder were inconsistent and thus had

to be vacated); State v. Owens, 766 S.E.2d 66, 71 (Ga. 2014) (concluding that felony

murder and felony involuntary manslaughter were mutually exclusive and

remanding for a new trial); Commonwealth v. Medeiros, 921 N.E.2d 98, 103 (Mass.

2010) (remarking that Massachusetts sets aside mutually exclusive verdicts);

State v. Speckman, 391 S.E.2d 165, 167–68 (N.C. 1990) (reasoning that convictions

for embezzlement and false pretenses were “mutually exclusive” and required a

new trial).

¶19 Still, the People contend that in Frye we simply conducted a factual analysis

to assess whether two verdicts are irreconcilable. They ask us to do the same here.

Because Delgado allegedly knocked R.B. unconscious before he took R.B.’s

personal belongings, the People contend that the jury could have reasonably

concluded that there was both a theft from a person and a robbery. But, because

9
both parties agree that there was only one taking, this seems, at best, implausible.

We would have to assume that the jury believed that Delgado knocking R.B.

unconscious was somehow the robbery, and the taking of R.B.’s items was the

theft. Yet, the robbery could not have occurred without the taking, and the theft

could not have occurred through the use of force. And, in any event, this would

be exactly the type of “speculation into a jury’s thought processes that courts

generally eschew.” Frye, 898 P.2d at 566. So, the People’s argument fails on its

own terms.

¶20 Moreover, Frye suggests an elemental approach to evaluating whether the

remaining guilty verdicts are irreconcilable. There, we stated that, for mutually

exclusive guilty verdicts, “where the existence of an element of one of the crimes

negates the existence of a necessary element of the other crime . . . [,] the verdicts

are legally and logically inconsistent and should not be sustained.” Id. at 569 n.13

(emphases added).

¶21 But why, exactly, is it that mutually exclusive guilty verdicts aren’t

permitted? The answer lies in a defendant’s right to have each element of a crime

proven beyond a reasonable doubt.

¶22 The prosecution’s responsibility to prove each element of a charged offense

beyond a reasonable doubt is embedded within the Fifth and Fourteenth

Amendments’ due process protections and the Sixth Amendment’s right to trial

10
by jury. See Sullivan v. Louisiana, 508 U.S. 275, 278 (1993) (“[T]he jury verdict

required by the Sixth Amendment is a jury verdict of guilty beyond a reasonable

doubt.”); In re Winship, 397 U.S. 358, 364 (1970) (“[T]he Due Process Clause protects

the accused against conviction except upon proof beyond a reasonable doubt of

every fact necessary to constitute the crime with which he is charged.”).

¶23 When a defendant is convicted of crimes featuring elements that are

mutually exclusive, the defendant hasn’t been convicted of each crime beyond a

reasonable doubt. There’s an explicit finding of doubt in the contradictory

element. Delgado can’t have been convicted of each element of robbery beyond a

reasonable doubt when robbery explicitly requires that he use force, which he was

specifically found not to have used.

¶24 But couldn’t the same be said for inconsistent guilty and not-guilty

verdicts—an inconsistency that both this court and the Supreme Court have

permitted? See, e.g., United States v. Powell, 469 U.S. 57, 69 (1984); Frye, 898 P.2d at

571. We think not.

¶25 Take Powell, for example. There, the defendant was acquitted of possession

with intent to distribute cocaine, but he was convicted of using a telephone to

possess and distribute the same. See Powell, 469 U.S. at 59–60. These findings were

clearly inconsistent, but the Supreme Court let them stand, see id. at 69, because

the verdicts could have been the result of “mistake, compromise, or lenity,” see id.

11
at 65. In other words, “[t]he most that can be said . . . is that the verdict shows that

either in the acquittal or the conviction the jury did not speak their real

conclusions, but that does not show that they were not convinced of the

defendant’s guilt.” Id. at 63 (quoting Dunn v. United States, 284 U.S. 390, 393

(1932)).

¶26 Not so with mutually exclusive verdicts. We know the People failed to meet

their burden on both charges, because the jury explicitly found that they failed to

meet their burden. A guilty verdict is a specific finding, encompassing all the

elements of the crime. An acquittal isn’t. In other words, while an acquittal has

various explanations, a guilty verdict has but one.

¶27 So, due process prevents a defendant from being convicted of crimes with

mutually exclusive elements. But to analyze whether the verdicts are mutually

exclusive, we need to look to the elements of both crimes that the state alleged.

First, theft from a person: “Theft from the person of another by means other than the

use of force, threat, or intimidation is a class 5 felony without regard to the value of

the thing taken.” § 18-4-401(5), C.R.S. (2019) (emphasis added). Now, robbery: “A

person who knowingly takes anything of value from the person or presence of

12
another by the use of force, threats, or intimidation commits robbery.” § 18-4-301(1),

C.R.S. (2019) (emphasis added).2

¶28 The two verdicts are legally and logically inconsistent. Sustaining these

convictions would mean that Delgado was found guilty beyond a reasonable

doubt of taking R.B.’s items by force. But it would also mean that Delgado was

found guilty beyond a reasonable doubt of taking those same items, from the same

victim, during the same occurrence, without any force.

¶29 Our plain language resolution is bolstered by precedent. In People v. Warner,

801 P.2d 1187, 1191 (Colo. 1990), we analyzed the Legislative Council’s comment

to the proposed “theft from the person” subsection. That comment stated that

theft from a person “is an act of ‘stealing,’ without circumstances of force or

violence as would constitute robbery.” Id. (quoting Colo. Legis. Council, 45–98, at

36 (1964), http://www.law.du.edu/images/uploads/library/CLC/98a.pdf

[https://perma.cc/MUW2-Z4RE]). Unlike robbery, “theft from the person of

another is intended to cover those thefts involving an invasion of the victim’s

person of which the victim is unaware, but which are not accomplished through

the use of force, threats, or intimidation.” Id. Warner therefore confirms what the

2We focus our attention on force because the People have not argued that Delgado
used threats or intimidation.

13
plain language already indicates: Elements of robbery and theft are mutually

exclusive when they are predicated on a single taking.

¶30 This raises the question of what the trial court should do to avoid mutually

exclusive guilty verdicts. The answer is that the trial court should instruct the jury

that a defendant may not be convicted of multiple crimes when the elements of

those crimes are mutually exclusive. Thus, here, the court should have instructed

the jury that Delgado could be convicted of robbery or theft, but not both. The

failure to give such a carrying instruction constituted error.

¶31 The trial court compounded this error by telling the jury to do just the

opposite. Instruction 14 explicitly stated that the jury should consider “each count

. . . separately, uninfluenced by [its] decision as to any other count.” 3 This gave

the jury license to consider the theft-from-a-person charge without pausing to

consider the elemental conflict as to the robbery charge. In addition, the court told

3 Jury Instruction 14 states in its entirety:
In this case a separate offense is charged against the defendant in each
count of the complaint. Each count charges a separate and distinct
offense and the evidence and the law applicable to each count should
be considered separately, uninfluenced by your decision as to any
other count. The fact that you may find the defendant guilty or not
guilty of one of the offenses charged, should not control your verdict
as to any other offense charged against the defendant.
The defendant may be found guilty or not guilty of any one or
all of the offenses charged.

14
the jury that it could find Delgado guilty of “any one or all of the offenses

charged.” This then culminated in the court erroneously entering judgment on

mutually exclusive guilty verdicts.

C. The Error Was Plain
¶32 Even so, the error (whether in improperly instructing the jury or in receiving

and entering judgment on the verdicts) wasn’t preserved. Thus, the error must

have been plain to warrant any remedy.4 As we stated above, an error is plain

when it’s “obvious and substantial,” Miller, 113 P.3d at 750 (quoting Stewart,

55 P.3d at 120), and “so undermine[s] the fundamental fairness of the trial itself as

to cast serious doubt on the reliability of the judgment of conviction,” Hagos, ¶ 18,

288 P.3d at 120 (quoting Wilson, 743 P.2d at 420). The “error must generally be so

obvious that a trial judge should be able to avoid it without the benefit of an

objection.” Scott v. People, 2017 CO 16, ¶ 16, 390 P.3d 832, 835 (citing

People v. Pollard, 2013 COA 31M, ¶ 39, 307 P.3d 1124, 1133).

¶33 Typically, the alleged error must violate “(1) a clear statutory command;

(2) a well-settled legal principle; or (3) Colorado case law.” Id. (quoting Pollard,

4We reject Delgado’s invitation to find that the error at issue here is structural
without any precedent from the Supreme Court or this court stating as much.
Furthermore, our conclusion that the error here was plain renders structural error
analysis superfluous.

15
¶ 40, 307 P.3d at 1133). The People argue that there isn’t a case directly on point.

That is, there isn’t a case that says in no uncertain terms that “robbery and theft

from a person are mutually exclusive.” But even if Frye doesn’t satisfy the

command of “Colorado case law” or “a well-settled legal principle” because our

statement there was dicta, the court of appeals has applied Frye in several cases

like this one reaching back decades. See Weare, 155 P.3d at 529–30 (applying the

Frye rule); People v. Beatty, 80 P.3d 847, 852–53 (Colo. App. 2003) (same); People v.

Lee, 914 P.2d 441, 447–48 (Colo. App. 1995) (same).

¶34 Perhaps even more importantly, the error here was obvious on its face. The

difference between theft from a person and robbery as to force couldn’t be more

stark: They are diametrically opposed.

¶35 The error was also substantial. The jury essentially found Delgado not

guilty of the robbery, through its guilty finding on the theft-from-a-person charge,

while simultaneously finding him guilty on the robbery charge. And, it “cast[s]

serious doubt on the reliability of the judgment of convictions” because it’s

impossible to know whether the jury believed that Delgado did or did not use

force. Hagos, ¶ 18, 288 P.3d at 121 (quoting Wilson, 743 P.2d at 420).

16
¶36 Thus, the error was plain.5

D. The Proper Remedy Is Retrial

¶37 If we reject the People’s argument regarding inconsistency, the People ask

us in the alternative to maximize the verdicts by sustaining the class-four-felony

robbery conviction and erasing the class-five-felony theft-from-a-person

conviction. Delgado argues that the prohibition against double jeopardy

necessitates that we acquit him on both charges. Both parties miss the mark. Our

jurisprudence on maximization doesn’t apply to the situation before us. And

while double jeopardy in other cases might require acquittal, it doesn’t here.

¶38 The People direct us to a line of court of appeals cases that have maximized

mutually exclusive guilty verdicts. See, e.g., Beatty, 80 P.3d at 853; People v. Atkins,

844 P.2d 1196, 1201–02 (Colo. App. 1992), abrogated by Candelaria, 148 P.3d at

183–84. We disagree with the divisions’ approach in those cases for several

reasons.

¶39 First, the precedent on which those cases rely didn’t analyze mutually

exclusive guilty verdicts. Instead, they rested on notions of lenity and duplicity,

as a defendant shouldn’t be convicted of murder twice for murdering one victim

5The People do not argue that the error here was invited, and we do not address
that issue.

17
and shouldn’t be convicted of a lesser offense when he’s already been convicted of

the greater offense. See People v. Bartowsheski, 661 P.2d 235, 245–47 (Colo. 1983);

People v. Lowe, 660 P.2d 1261, 1268–69 (Colo. 1983), abrogated by Callis v. People,

692 P.2d 1045 (Colo. 1984).

¶40 Second, the reasoning behind those court of appeals cases makes little sense

in cases such as this one, where the jury effectively acquitted the defendant on one

of the charges. For example, in Atkins, the defendant was charged with and

convicted of first degree murder “based upon both extreme indifference murder

and murder after deliberation.” 844 P.2d at 1198. As the division there discussed,

“both verdicts were necessarily based upon” the defendant shooting and killing

the victim. See id. at 1200–02. As a result, no matter whether the defendant

committed the murder with specific intent or extreme indifference, he would still

be guilty of first degree murder. See id. at 1202.

¶41 Not so here. The two crimes, robbery and theft from a person, are

necessarily based upon different actions, rather than “differing characterizations

that may be placed upon that conduct.” Cf. id. Moreover, they’re in fact two

different crimes, whereas in cases like Atkins, the murder statute “simply

outline[d] two alternative methods of committing the singular crime of first degree

murder.” Id. Thus, even if we did agree that cases like Atkins properly maximized

the verdicts, their rationale doesn’t reach this case.

18
¶42 Third, because an element of each crime negates the other, it’s impossible to

know what exactly the jury intended. We can’t assume that the jury intended to

find Delgado guilty of the “maximized” crime when the jury effectively acquitted

him of that crime through its theft-from-a-person verdict. The only finding that

we can be sure of is that Delgado unlawfully took items, since that’s a requirement

of both convictions. That might imply that we should “minimize” the convictions.

Yet, we can’t do that either because the jury made an explicit finding through its

verdict that Delgado used force.

¶43 Delgado maintains that the proper outcome is acquittal on both charges

based on double jeopardy. Delgado argues that, since the jury decided beyond a

reasonable doubt that the taking was committed by means other than force, he was

acquitted on the robbery charge. And, additionally, since the jury decided beyond

a reasonable doubt that the taking was committed by force, he was acquitted on

the theft-from-a-person charge. But the jury didn’t actually acquit Delgado, so

double jeopardy doesn’t come into play. Indeed, acquittal isn’t preclusive when a

jury convicts on one charge and acquits on another, even if both convictions are

based on the same factual circumstances. Bravo-Fernandez v. United States,

137 S. Ct. 352, 356–57 (2016). Therefore, double jeopardy doesn’t bar retrial here.

¶44 That leaves us with one logical option: a new trial. Retrial avoids trying to

peer into the minds of the jurors to determine what they could have meant by the

19
conflicting verdicts. And it also ensures that Delgado doesn’t gain a windfall from

the error. This is by no means a radical remedy, as other courts have remanded

for retrial when a jury finds the defendant guilty of mutually exclusive verdicts.

See, e.g., Chyung, 157 A.3d at 632–33; Owens, 766 S.E.2d at 71; Speckman, 391 S.E.2d

at 168.

¶45 The proper remedy for mutually exclusive verdicts is retrial.

III. Conclusion

¶46 Therefore, we hold that because the elements of theft from a person and

robbery negate each other when they are based on a single taking, a defendant

may not be convicted of both offenses. Because such mutually exclusive

convictions leave us without a meaningful way to discern the jury’s intent, the

proper remedy is a new trial. Accordingly, we affirm the judgment of the court of

appeals.

JUSTICE BOATRIGHT dissents, and CHIEF JUSTICE COATS joins in

the dissent.

20
 JUSTICE BOATRIGHT, dissenting.

¶47 The majority asks the question, “what’s the remedy?” Maj. op. ¶ 4. But that

question is premature. The real question is, why is the remedy warranted when

the error was not plain? In my view, the majority only makes a passing reference

to the instructional error, and it fails to meaningfully articulate why the trial

court’s instructions rise to the level of plain error. Instead, the majority focuses its

plain error analysis on explaining how the two convictions are “mutually

exclusive.” But that misses the point. This case turns on the instructions the trial

court did, and did not, provide to the jury. The purported inconsistency in the

verdicts is not the error itself; rather, the existence of the two verdicts is the product

of an instructional error committed by the trial court. The fact that neither party—

throughout the entire appellate process—correctly identified the instructional

error as the critical error demonstrates that it was not obvious. And it follows that,

if the error was not obvious, then it cannot be considered plain. Hence, I would

affirm the convictions. Additionally, to the extent that a remedy would be

warranted, I disagree with the majority’s remedy to vacate both convictions and

order a new trial because it incentivizes defendants to remain silent at the trial

court when they perceive that an instructional error such as this is about to occur.

Accordingly, I respectfully dissent.

1
¶48 Although the majority accurately recites the facts and procedural history,

I provide my own synopsis and include some additional procedural history to

illustrate that the error committed by the trial court was not plain.

¶49 When R.B. knocked on the exit door of a closed bar, a member of the bar’s

staff called the police to report a disturbance. An officer arrived only a few

minutes later to find Delgado digging through the pockets of R.B., who had been

attacked from behind and lost consciousness. There were no eyewitness accounts

of who struck R.B.

¶50 After Delgado saw the officer, he began walking away from R.B. while

carrying several items. The officer ordered him to stop, but Delgado began

running down an alley and a chase quickly ensued. As several officers chased

Delgado, they observed him throw several items aside. When the officers caught

and arrested Delgado, they also recovered R.B.’s wallet, phone, and keys nearby.

¶51 The People initially charged Delgado with robbery 1 and misdemeanor

assault. However, shortly before trial, the People appear to have recognized that

the identity of who struck R.B. was the weakest link in their case against Delgado.

1 In Colorado, someone has committed robbery when he or she applies physical
force “at any time during the course of a transaction culminating in the taking of
property from the victim’s person or presence.” People v. Bartowsheski, 661 P.2d
235, 244 (Colo. 1983).

2
 As a result, one week before trial, the People added an additional count of theft

from a person. Because that offense eliminates the force element, it gave the

prosecution a plan B in the event that the jury had a reasonable doubt about who

knocked R.B. unconscious. Defense counsel objected to the additional count on

the ground that it was too close to trial to add a felony charge. The trial court

overruled the objection and allowed the People to add the charge.

¶52 The matter proceeded to trial. Although defense counsel vigorously cross

examined R.B. about his inability to describe his attacker to the police immediately

after the incident, he identified Delgado as his attacker.

¶53 After the close of the evidence, the trial court held a jury instruction

conference with the parties to discuss the proposed instructions. During the

conference, the trial court concluded that theft from a person is not a lesser

included offense of robbery, and stated that Delgado “can be convicted of both of

the[] offenses.” Neither party raised an objection about the potential for

conflicting verdicts. Instead, defense counsel only renewed the objection to the

late addition of the theft from a person charge. Importantly, no one asked for a

carrying instruction, nor did anyone object to Instruction 14, which directed the

jury to consider each charge independently. As a result, the trial court did not

instruct the jury that it could only find Delgado guilty of either robbery or theft

from a person, but it did instruct the jury that a verdict on one count should not

3
 influence its verdict on the others. Following the trial court’s instructions, the jury

convicted Delgado of both robbery and theft from a person but hung on the

misdemeanor assault charge. Delgado then appealed.

¶54 On appeal, Delgado’s primary argument was that the result of the jury’s

verdicts—the inconsistent convictions—constituted structural error based on his

assertion that the jury’s verdicts negated essential elements of each offense.

Additionally, Delgado argued in the alternative that, if the court declined to find

the error structural, then the existence of the inconsistent convictions themselves

amounted to plain error. The defendant concedes that these arguments were not

preserved at the trial court.

¶55 As the majority noted, we review unpreserved constitutional claims for

plain error. For an error to be “plain,” it must be both “obvious and substantial.”

Hagos v. People, 2012 CO 63, ¶ 14, 288 P.3d 116, 120. The error must be obvious to

the point that it “contravene[s] (1) a clear statutory command; (2) a well-settled

legal principle; or (3) Colorado case law.” Scott v. People, 2017 CO 16, ¶ 16, 390 P.3d

832, 835 (quoting People v. Pollard, 2013 COA 31M, ¶ 40, 307 P.3d 1124, 1133

(citations omitted)).

¶56 While the majority makes a passing reference to the instructional error at

issue here, it primarily focuses its time and attention on discussing how the two

convictions are “mutually exclusive.” And it goes to lengths to explain that the

4
 theft from a person and robbery verdicts are logically and legally inconsistent

based on the proposition that a taking cannot simultaneously be accomplished

with force and with means other than force. However, the jury’s verdicts are not

the error, but rather they are the product of the error. Therefore, where I depart

from the majority is in its decision to rest its plain error analysis almost exclusively

on the arguably inconsistent verdicts rather than the instructional error itself.

¶57 In focusing predominately on the verdicts, the majority speeds past the

instructional error. In so doing, it merely nods at that error and fails to provide

any meaningful analysis as to why that error is plain. Maj. op. ¶¶ 30, 31. Indeed,

the majority purports to reject Delgado’s “invitation to find that the error at issue

here is structural.” Id. at ¶ 32 n.4. But, in my view, it does just that; it treats the

result of the verdicts as some sort of quasi-structural error2 by concluding that “it’s

impossible to know whether the jury believed that Delgado did or did not use

force.” Id. at ¶ 35. The majority correctly identifies the three types of errors that

would constitute plain error: (1) errors contravening a clear statutory command;

2Structural errors are few and far between and “includ[e] errors concerning rights
protecting some interest other than the defendant’s interest in not being
erroneously convicted; errors the effects of which are too hard to measure, in the
sense of being necessarily unquantifiable and indeterminate; and errors that can
be said to always result in fundamental unfairness.” James v. People, 2018 CO 72,
¶ 15, 426 P.3d 336, 339.

5
 (2) errors contravening a well-settled legal principle; and (3) errors contravening

Colorado case law. Scott, ¶ 16, 390 P.3d at 835 (quoting Pollard, ¶ 40, 307 P.3d at

1133). But it fails to indicate into which category the instructional error falls. That

is because it cannot. Instead, it simply pronounces that it is plain error because it

is “obvious on its face.” Maj. op. ¶ 34. That rationale, in my opinion, fails. How

can the majority call the error “obvious on its face” when neither party identifies

it as the critical mistake?3

¶58 The majority is correct when it mentions that the trial court should instruct

the jury that it cannot convict a defendant of mutually exclusive verdicts. In fact,

the trial court should give the jury a specific roadmap in a carrying instruction. It

should instruct the jury to first consider the greater offense, in this case the

robbery, and decide whether each and every element of that offense has been

proven. If the jury determines that the prosecution has proven that offense beyond

a reasonable doubt, then it should not consider the offense of theft from a person

and instead leave that verdict form blank. Next, the jury should be told that if it

3It should be noted that the defendant did assert that the prosecution injected the
error by, among other things, requesting Instruction 14. But the defendant made
that point only when discussing an appropriate remedy—namely, why this court
should reject the People’s recommended remedy of maximizing the jury’s
verdicts.

6
 determines that the prosecution failed to prove one or more elements of robbery

beyond a reasonable doubt, then, and only then, should it consider the offense of

theft from a person.

¶59 Instead, the trial court told the jury to do just the opposite here. Instruction

14 explicitly stated that the jury should consider each count “separately,

uninfluenced by [its] decision as to any other count.”4 This directed the jury to do

exactly what it did: consider the theft from a person charge separate and apart and

uninfluenced by its decision on the robbery charge. In addition, the instruction

informed the jury that it could find Delgado guilty of “any one or all of the offenses

charged.” So, it did; the jury found him guilty of robbery and theft from a person. 5

¶60 The trial court’s failure to provide the jury with a carrying instruction here,

coupled with the instruction to consider each count separately, constitutes

4 Instruction 14 states in its entirety:
In this case a separate offense is charged against the defendant in each
count of the complaint. Each count charges a separate and distinct
offense and the evidence and the law applicable to each count should
be considered separately, uninfluenced by your decision as to any
other count. The fact that you may find the defendant guilty or not
guilty of one of the offenses charged, should not control your verdict
as to any other offense charged against the defendant.
The defendant may be found guilty or not guilty of any one or
all of the offenses charged.
5 The jury hung on the misdemeanor assault charge, and the prosecutor

subsequently dismissed that count.

7
 instructional error. But, in my view, this error is not obvious for two reasons. First,

throughout this entire process, neither party has correctly identified the

instructional issue as the pertinent error in this case. Again, if an error is not

properly identified by the parties throughout the entire process, it cannot be plain

error. And second, while the majority makes a persuasive case for the mutually

exclusive nature of the verdicts, there is a reasonable explanation for both of the

jury’s verdicts.

¶61 By unpacking what happened here, it is apparent what occurred. In

considering each count separately, the jury could, and obviously did, decide that

Delgado used force to knock the victim out and steal the items; thus, the robbery

conviction. Then, following Instruction 14, it turned to the theft from a person

offense. In considering that offense—uninfluenced by its verdict on the robbery

count—the jury then decided that the prosecution had proven beyond a

reasonable doubt that Delgado took items from an already unconscious man

without force and intended to permanently deprive the victim of those items. The

jury did not consider that one offense was “with force” and the other was “without

force” because Instruction 14 told the jurors not to do so. If the jury had, in fact,

compared the conflicting elements, it would have explicitly violated Instruction

14. In other words, the jury did exactly what it was told to do. That is why the

error here lies in the instructions the jury did and did not receive and not in the

8
 verdicts themselves. And, in my opinion, because both parties fail to argue that

this is the critical error, this error was not plain.

¶62 For an error to be plain, it must be obvious and substantial. Hagos, ¶ 14,

288 P.3d at 120. Indeed, “[p]lain error review allows the opportunity to reverse

convictions in cases presenting particularly egregious errors, but reversals must

be rare.” Id. at ¶ 23, 288 P.3d at 121–22. In practical terms, for something to rise to

the level of plain error, the judge should observe something so flagrant and unfair

that he or she feels compelled to raise both hands and yell, “stop.” See Pollard,

¶ 39, 307 P.3d at 1133 (noting that such errors “must be so clear-cut, so obvious,

that a trial judge should be able to avoid [them] without the benefit of an

objection”). The instructional error here simply does not rise to that level. In fact,

the judge said just the opposite, declaring that Delgado “can be convicted of both

of the[] offenses.” That declaration, which the majority now says is plain error

because it is so obvious, went unchallenged and unnoticed through trial and

through direct appeal. To be sure, the jury instructions were only identified as the

critical error—not by the parties, but rather by this court—at this late stage in the

proceedings. That, almost by definition, means that the error was not obvious.

¶63 Not only must an error be obvious to be deemed plain, it must also be

substantial. Hagos, ¶ 14, 288 P.3d at 120 (stating that the error must “so

undermine[] the fundamental fairness of the trial itself so as to cast serious doubt

9
 on the reliability of the judgment of conviction”). Here, the majority concludes

that the error is substantial because by finding Delgado guilty of the theft from a

person charge, “the jury essentially found [him] not guilty of robbery.” Maj. op.

¶ 35. But that misses the mark. Under these facts, with the instructions given, a

conviction on one offense does not necessarily acquit Delgado of the other. When

viewed through the lens of what the jury was instructed to do, there is not serious

doubt about the conviction of theft from a person.

¶64 In analyzing the second prong of our plain error framework, the majority

declares that the error is substantial because the two verdicts conflict. I agree in

part. Focusing exclusively on the verdicts, it is fair to say that the verdict of

“without force” in the theft from a person count negates the element of “with

force” in the robbery count. But in this instance, the converse is not true.

¶65 In my view, the element of “without force” is not a material element of the

crime of theft from a person. When the statute states that the theft was

accomplished “by means other than the use of force, threat, or

intimidation,” § 18-4-401(5), C.R.S. (2019), it is, in effect, saying that it was not a

robbery. The “without force” element is unusual in that it requires a jury to find

something that a person did not do. Here, what the jury found beyond a

reasonable doubt satisfies the material elements of theft from a person:

10
 • The defendant

• exercised control over

• anything of value

• which was the property of another,

• without authorization,

• with the intent to permanently deprive the other person of the use or

benefit of the thing of value, and

• the thing was taken from the person of another.

That is theft from a person. All of those elements were proven beyond a

reasonable doubt. The fact that the jury verdicts make it unclear whether Delgado

accomplished the taking by force or without force does not exonerate him, nor

does it cast serious doubt on the reliability of the theft conviction. Hence, even if

the error is obvious, it is not substantial.

¶66 While I would not afford Delgado a remedy, I feel compelled to comment

on the majority’s remedy—it goes too far. By requiring retrial in this instance, the

majority incentivizes defendants to remain silent when a potential instructional

error such as this arises. Thus, under the majority’s formulation, the worst

possible outcome for a defendant facing inconsistent convictions would merely be

retrial based on the remedy the majority adopts today. In other words, by staying

quiet, he gets a do-over. With this as the remedy, why would competent defense

11
 counsel say anything if they perceive the potential for inconsistent convictions?

Instead, in order to cure the inconsistent verdicts, I would affirm the conviction on

the theft from a person charge and I would vacate the robbery conviction.

¶67 For the foregoing reasons, I respectfully dissent.

I am authorized to state that CHIEF JUSTICE COATS joins in this dissent.

12