IN THE COURT OF APPEALS OF IOWA _______________
No. 25-0412 Filed July 8, 2026 _______________
State of Iowa, Plaintiff–Appellee, v. Kyle John Roberts, Defendant–Appellant. _______________
Appeal from the Iowa District Court for Black Hawk County, The Honorable John J. Sullivan, Judge. _______________
AFFIRMED _______________
Austin Jungblut of Parrish Kruidenier, L.L.P., Des Moines, attorney for appellant.
Brenna Bird, Attorney General, and Nicholas E. Siefert, Assistant Attorney General, attorneys for appellee. _______________
Considered without oral argument by Tabor, C.J., and Chicchelly and Sandy, JJ. Opinion by Chicchelly, J.
1 CHICCHELLY, Judge.
Kyle Roberts appeals his jury conviction for felony stalking. On appeal, he argues (1) there is insufficient evidence that his course of conduct was directed at a specific person and (2) we should adopt plain error doctrine and review a series of unpreserved evidentiary issues. Upon our review, we find the evidence is sufficient and binding supreme court precedent has already rejected plain error review in Iowa, so we affirm Robert’s conviction.
BACKGROUND FACTS AND PROCEEDINGS In 2015, Roberts was convicted of stalking and sentenced to prison. See State v. Roberts, No. 15–1164, 2016 WL 4801382, at *1 (Iowa Ct. App. Sept. 14, 2016). That case was prosecuted by the Black Hawk County Attorney’s Office. While incarcerated, Roberts began to write to the assistant county attorney who handled the case. In two profane and racial-slur-laden letters received by the prosecutor, Roberts detailed his innocence of the stalking charge but also commented on the prosecutor’s physical appearance and expressed an interest in becoming romantically involved with her.
Because the letters concerned her, the prosecutor contacted the Black Hawk County Sheriff’s Office to put a safety plan in place in anticipation of Roberts’s release from prison. Part of that safety plan required law enforcement to contact the prosecutor if they discovered communications from Roberts that concerned the prosecutor.
From 2017 through 2024, the prosecutor became aware of dozens of messages, social media posts, emails, and other communications that had been made by Roberts regarding the prosecutor and other members of the law enforcement community. In one communication Roberts threatened to put a “bullet in the heads” of court officials involved in his case. On another
2 occasion, Roberts messaged a law enforcement officer on Facebook and asked him if he knew the prosecutor. And if he did, Roberts requested the officer shove her face in “dog sh*t.”
Roberts’s conduct became still more concerning. In 2017, the prosecutor attended her high school reunion at a festival. While the festival was a public event, the reunion took place in a private roped-off area. Law enforcement officers who were in attendance became concerned because Roberts had entered the roped-off area and was seen standing right behind the prosecutor.
Then in 2021, the prosecutor went to a local casino with her mother and sister. At the casino, the prosecutor was playing a slot machine when Roberts came up behind her and tapped her on the shoulder. When she turned, Roberts began to ask if she remembered him. Initially not recognizing him, she did not respond immediately, which angered Roberts who then asked if they could sit down and talk. The prosecutor then realized who he was and told him this was inappropriate. She then left the casino with her family.
After this incident, Roberts’s electronic communications amplified. The prosecutor had been appointed as a District Associate Judge. And the communications that referenced her were reported to her on a near daily basis. One of these communications was an email to the Waterloo Police Department that read, “This is a fair warning. If you don’t hold the court [accountable] and your co-worker doesn’t come forward over the bullshit he said and have my innocence acknowledged soon you’ll pay the price, or at least some [of ] you will, at the very least the state will.” Another email accused the prosecutor-turned-judge of misconduct and threatening to “ruin [the prosecutor]’s career, or at least try. I will go further than that.” And in
3 that same email threatened to take “heavy weapons into the police department’s parking lot and causing a lot of damage to innocent people’s vehicles.”
The threats against the judge became more direct. Roberts began posting photos of the judge and included threats. One post included a picture of her and was captioned, “this bitch needs to fix something that she got horribly wrong.” In another post, Roberts made the judge’s judicial branch portrait as his profile picture and then posted a YouTube link to a video about a 20mm anti-material assault rifle.
Roberts’s conduct continued for the next few years. In 2024, in an email to the county treasurer, Roberts professed his innocence and concluded by stating, the judge “is going to get a hard dick in her ass.” Because of her role, the judge was notified of filings made by Roberts in both the original stalking case and his newly filed post-conviction-relief application. These documents included derogatory comments and death threats about the judge. Roberts also began making threats to take his own life if the court system did not give him his preferred result.
The State charged Roberts by trial information with stalking as a class “C” felony in violation of Iowa Code section 708.11(2) and (3)(a) (2021).1 The matter proceeded to a jury trial. Roberts waived his right to be represented by counsel and instead represented himself at trial. The jury convicted Roberts as charged. The district court sentenced Roberts to a ten- year indeterminate term of incarceration. Roberts now appeals.
1 The trial information stated the basis for the stalking charge was Roberts’s conduct “between the period of September 15, 2021, through August 29, 2024.”
4 STANDARD OF REVIEW We review sufficiency-of-the-evidence claims for correction of errors at law. See State v. Jones, 967 N.W.2d 336, 339 (Iowa 2021). “In determining whether the jury’s verdict is supported by substantial evidence, we view the evidence in the light most favorable to the State, including all ʻlegitimate inferences and presumptions that may fairly and reasonably be deduced from the record evidence.’” Id. (citation omitted).
DISCUSSION I. Sufficiency of Evidence
Roberts first challenges the sufficiency of the evidence supporting his stalking conviction. Unchallenged jury instructions become the law of the case for purposes of our review of sufficiency of the evidence. State v. Mathis, 971 N.W.2d 514, 518 (Iowa 2022). The jury’s instructions required the State to prove the following elements of stalking using a technological device and/or stalking in violation of a protective order: 1. Between September 15, 2021, and August 29, 2024, the defendant purposefully engaged in a course of conduct directed at [M.W.] that would cause a reasonable person to feel terrorized, frightened, intimidated or threatened or to fear bodily injury to [M.W.]
2. The defendant knew or should have known that [M.W.] would reasonably feel terrorized, frightened, intimidated or threatened or be placed in reasonable fear of bodily injury or death to her.
3. The defendant’s course of conduct caused [M.W.] to feel terrorized, frightened, intimidated or threatened or to fear bodily injury to [M.W.]
4. At the time the defendant engaged in the course of conduct[ ] directed at [M.W.]
(a) defendant utilized a technological device while committing stalking; and/or
5 (b) defendant was subject to restrictions contained in a criminal or civil protection order or injunction against any person.
See Iowa Code § 708.11(1)–(3).
Roberts’s sole challenge to the sufficiency is that his conduct was never “directed at a specific person.” See id. § 708.11(1)(b). His position is that because the communications were never sent directly to the judge they were not directed at her. We disagree with Roberts and find on this record that the evidence was sufficient to show that he directed his conduct at the prosecutor-turned-judge.
The jury instruction defines “course of conduct” as repeatedly maintaining a visual or physical proximity to a person without legitimate purpose, repeatedly utilizing a technological device to locate, listen to, or watch a person without authorization or legitimate purpose, or repeatedly conveying oral or written threats, threats implied by conduct, or a combination thereof, directed at or toward a person.
And repeatedly means “on two or more occasions.” Based on that, we find there is no dispute that the incident at the casino where Roberts approached the victim and directly asked her multiple times if she remembered him is part of the “course of conduct” and was “directed at” her.
That brings us to the other instances of conduct that were not as direct as the casino incident. Our supreme court has said that conduct can be directed at a victim even if the conduct was also directed at another party. See State v. Limbrecht, 600 N.W.2d 316, 319–20 (Iowa 1999). In Limbrecht, the defendant argued his course of conduct was directed at the named victim’s husband and not the named victim herself. See id. at 319. The supreme court rejected that argument and found that the victim was the
6 “object of” the defendant’s conduct and that was the guiding principle in finding sufficient evidence of stalking. See id. at 319–20.
Applying that precedent, we find that messages that contained direct or implied threats towards the prosecutor-turned-judge were “directed at” her. Because we are reviewing this case for sufficiency of evidence, we view all of those threats in the light most favorable to the State, which includes making all “legitimate inferences and presumptions that may fairly and reasonably be deduced from the record evidence.” See State v. Williams, 695 N.W.2d 23, 27 (Iowa 2005) (citation omitted). And because these threats were specific and targeted, we find they were “directed at” the prosecutor- turned-judge. Accordingly, we find sufficient evidence supports Roberts’s conviction for stalking.
II. Evidentiary Issues
Then, Roberts argues the district court erred when admitting multiple pieces of evidence that he argues were irrelevant, unduly prejudicial, and prior bad-acts evidence. But Roberts admits he did not object to this evidence before the district court. Instead, Roberts argues that we can bypass this error-preservation problem by adopting plain-error review. But our supreme court rejected that argument just five years ago. See State v. Treptow, 960 N.W.2d 98, 109 (Iowa 2021) (“We have repeatedly rejected plain error review and will not adopt it now.”). As an intermediate appellate court, we are duty-bound to apply binding supreme court precedent, so we decline to reach Roberts’s unpreserved arguments.
AFFIRMED.
Sandy, J., specially concurs; Tabor, C.J., joins special concurrence parts II and III.
7 SANDY, Judge (specially concurring). Roper: So, now you give the Devil the benefit of law!
More: Yes! What would you do? Cut a great road through the law to get after the Devil?
Roper: Yes, I’d cut down every law in England to do that!
More: Oh? And when the last law was down, and the Devil turned round on you—where would you hide, Roper, the laws all being flat? This country’s planted thick with laws from coast to coast—man’s laws, not God’s—and if you cut them down . . . d’you really think you could stand upright in the winds that would blow then? Yes, I’d give the Devil benefit of law, for my own safety’s sake.2
In Robert Bolt’s A Man for All Seasons, Sir Thomas More would give even the Devil the benefit of law, knowing that a country whose laws lie flat offers no shelter from the winds that follow. Error preservation is one of those laws. My concern is not that we have cut it down, but that we have obscurely let it fall—exception by exception—while still insisting it stands upright. I write separately to say plainly what I think we are doing, and to suggest that if we are going to review unpreserved errors, we should do so under a doctrine we can name—plain error review.
There are three reasons I believe adoption of plain error review is the right path forward: (1) predictability, which does not exist under our current scheme; (2) congruence with our state constitution and early case law; and (3) necessity—the legislature’s 2019 amendment to Iowa Code section 814.7, 2019 Iowa Acts ch. 140, § 31, closed the ineffective-assistance safety valve, providing no direct appeal avenue of correction for the district court’s own unpreserved errors.
2 Robert Bolt, A Man for All Seasons: A Play in Two Acts 66 (1960).
8 I. The Erosion of Error Preservation
Judge Buller’s concurrence dubitante in State v. Hunter, No. 24-1999, 2026 WL 892487, at *2–5 (Iowa Ct. App. Apr. 1, 2026), traced the erosion of error preservation through the sentencing context—from State v. Cooley, 587 N.W.2d 752 (Iowa 1998) through State v. Chawech, 15 N.W.3d 78 (Iowa 2024) and State v. Hallock, 31 N.W.3d 36 (Iowa 2026). Whatever the merits of that trajectory, it operated within a discrete doctrinal space with some historical footing. But the recent diminution of error preservation is not merely confined to sentencing cases. See State v. Sharples, ___ N.W.3d ___, ___, 2026 WL 1500818, at *4 (Iowa 2026) (invoking the illegal-sentence doctrine to reach an habitual offender enhancement challenge the defendant affirmatively waived by stipulation, despite the Iowa Code section 902.8 (2024) sequencing requirement being an element requiring proof beyond a reasonable doubt—a stipulated to sufficiency question); State v. Jackson, 35 N.W.3d 345, 349–50 (Iowa 2026) (finding error preserved under Lamasters v. State, 821 N.W.2d 856, 864 (Iowa 2012) where the district court named the weight-of-the-evidence challenge but issued a blanket denial without applying the weight-of-the-evidence standard and determined a bare recitation of a party’s argument followed by a general denial in a posttrial ruling suffices to preserve error); State v. Uranga, 31 N.W.3d 358, 364–65 (Iowa 2026) (reaching a statutory-interpretation sufficiency theory never presented before or during trial by sheltering it under the rule established in State v. Crawford, 972 N.W.2d 189, 198 (Iowa 2022) that general post-trial acquittal motions preserve all sufficiency challenges); Christensen v. Iowa Dist. Ct., 21 N.W.3d 529, 532 (Iowa 2025) (candidly acknowledging the failure to preserve error, reaching an argument raised only by amicus and conceded as waived, and justifying review on the obviousness of the error and
9 the risk of entrenching bad precedent—functionally applying plain error review without the name).
Acknowledgement of the above is not intended as a critique. Rather, it serves to demonstrate that any alleged gap that exists between where we are now and where we would be if we adopted plain error review is not as vast as some may argue. Across all the above decisions, a consistent trend emerges: defendants who fail to preserve error at the trial level have found an increasing number of roads to appellate relief. This trend, collectively, has narrowed the practical force of our error-preservation doctrine. Each individual example can be explained on its own doctrinal terms, but the concern is the cumulative impact. Error-preservation doctrine derives much of its force from predictability. When a series of exceptions and expansions consistently permit review of issues not squarely presented below, the practical operation of the doctrine evolves.
Iowa has no plain error doctrine. But the case law surveyed above reveals the truth: Iowa already reviews unpreserved errors. It simply does so without principled standards for when it will and when it will not. For example, Chawech opened illegal-sentence challenges to essentially unlimited appellate review. See 15 N.W.3d at 86 (“As to illegal sentences, error preservation principles are irrelevant.”). Jackson allowed for a defendant to benefit from a standard the district court never applied nor had the chance to correct. See 35 N.W.3d at 350 n.1 (“[D]ecisions requiring or suggesting that a defendant must file a motion or otherwise bring the mistake to the court’s attention to preserve error on a claim that the district court applied the wrong standard when denying a motion for new trial should not be followed.”). Uranga extended acquittal motions to cover a theory never presented. See 31 N.W.3d at 365. And Sharples allowed a defendant to affirmatively stipulate to
10 a statutory enhancement element—otherwise requiring a sufficiency finding by a jury, see Apprendi v. New Jersey, 530 U.S. 466, 490 (2000), from the face of an indictment—and then challenge it on appeal anyway, shielding the maneuver behind the illegal sentence doctrine. ___ N.W.3d at ___, 2026 WL 1500818, at *4. None of these decisions articulate a consistent standard for when unpreserved errors will and will not be reached. The result is error preservation “whack-a-mole.” Appellate discretion without predictable limiting principles is no doctrine at all.
Plain error review, as the federal courts have applied it since United States v. Olano, 507 U.S. 725 (1993), supplies exactly what Iowa’s current approach lacks: a consistent, four-part framework requiring that (1) there be error, (2) the error be plain, (3) it affect substantial rights, and—critically— (4) leaving it uncorrected would seriously damage the fairness and integrity of judicial proceedings. Adopting plain error review would accomplish three things Iowa’s current doctrine cannot. First, it would restore consistency. Iowa courts are currently reaching unpreserved errors while suggesting—as the supreme court expressly did in Hallock—that “Chawech didn’t make new law.” 31 N.W.3d at 42. That assertion is, as Judge Buller observed in his Hunter concurrence dubitante, “hard to square with the evolution of the case law.” 2026 WL 892487, at *5. Plain error review would require the court to say plainly what it is doing: exercising discretion to correct an unpreserved error because the justice demands it.
Second, it would restore limiting principles. The four-prong Olano framework—error, plainness, effect on substantial rights, and impact on the integrity of the proceedings—gives content to a discretion that Iowa currently exercises without constraint. See 507 U.S. at 731–32. Not every unpreserved error would qualify. A defendant who stipulates to a habitual
11 offender enhancement, as occurred in Sharples, see ___ N.W.3d at ___, 2026 WL 1500818, at *4, would face a steep climb on the third and fourth prongs. A defendant who never raises a weight-of-the-evidence standard with the trial court, unlike in Jackson, see35 N.W.3d at 349–50, would struggle to show the error was plain. A defendant who advances a statutory theory for the first time on appeal, as in Uranga, see 31 N.W.3d at 365, would need to demonstrate that the trial court’s failure to adopt that theory was an obvious error affecting the outcome. The framework structures the analysis but does not guarantee relief.
Third, it would restore the institutional role of the district court. The doctrine of error preservation was not implemented purely for the purpose of procedural formalism. It is meant to ensure that trial courts have a genuine opportunity to correct their own errors before appellate courts intervene. Meier v. Senecaut, 641 N.W.2d 532, 537 (Iowa 2002). Plain error review respects that purpose. By requiring that an error be plain—obvious on the existing record—it ensures that only errors the district court should have caught without prompting qualify for appellate correction. Plain error review does not transfer primary responsibility for error correction from trial courts to appellate courts. Rather, it presumes the opposite. The doctrine’s demanding standards ensure that appellate intervention remains exceptional and that the trial court remains the principal forum for identifying and correcting mistakes. Errors that require argument, development of the record, or legal creativity to identify would not benefit from plain error review.
The objection to plain error review has always been that it would flood appellate dockets with unpreserved claims and undermine the discipline that error preservation supplies. That concern may have merit in the abstract. But
12 it has little force as applied to Iowa today. As demonstrated above, the discipline is already gone. And decades of federal case law suggest that adoption of plain error review would not unleash a torrent of successful unpreserved claims.3 Although federal appellate courts routinely entertain plain error arguments, relief remains comparatively rare because the third and fourth Olano prongs impose substantial hurdles. 507 U.S. at 731–32. Further, I would argue that forcing plain error questions to dubious ineffective-assistance-of-counsel claims via postconviction relief clogs both the district and appellate dockets more than the alternative.
The question is no longer whether Iowa will review unpreserved errors. It already does. The question is whether it will do so within a principled framework that gives consistent guidance to courts and litigants— or whether it will continue doing so case by case without the limiting principles that protect the integrity of the process. What plain error review offers is what Iowa currently lacks: a principled, consistent framework for deciding when unpreserved errors merit appellate attention. It acknowledges that the preservation requirement, like all procedural rules, occasionally produces unjust results—and it provides a structured, disciplined response to that reality rather than an ad hoc expansion of exceptions that cannot be predicted or consistently applied. A plain error doctrine, properly
3 For instance, after United States v. Booker, 543 U.S. 220 (2005), the federal circuit courts of appeals had to review thousands of pre-Booker sentences. See Michael W. McConnell, The Booker Mess, 83 Denver U. L. Rev. 665, 665, 669 (2006). One study tracked the Tenth Circuit’s review of those sentences. Id. at 669. It found that plain error was successfully invoked in those cases fifteen percent of the time. Id. But if the error involved was not a constitutional question, then only seven percent of cases were reversed under plain error review. Id.
13 understood, is not a retreat from the values that animate Iowa’s error preservation rules. It is the most faithful way left to administer it.
II. Our State Constitution & History
The supreme court has appellate jurisdiction only in cases in chancery and “shall constitute a court for the correction of errors at law, under such restrictions as the general assembly may, by law, prescribe”. Iowa Const. art. V, § 4. Some contend that plain error review is incompatible with correction of errors at law, and thus, our constitution. See Olson v. BNSF Ry. Co., No. 22-0587, 2023 WL 386709, at *4 (Iowa Ct. App. Jan. 25, 2023) (Buller, J., concurring) (“Preserving error is likely constitutionally required.”), majority vacated 999 N.W.2d 289 (Iowa 2023); State v. Tidwell, No. 13-0180, 2013 WL 6405367, at *2 (Iowa Ct. App. Dec. 5, 2013) (“If a litigant fails to present an issue to the district court and obtain a ruling on the same, it cannot be said that we are correcting an error at law.”); State v. Harrington, 893 N.W.2d 36, 42 (Iowa 2017) (“Error preservation is a fundamental principle of law with roots that extend to the basic constitutional function of appellate courts.”). I disagree.
One of our supreme court’s recent decisions, State v. Gomez Medina, never analyzed the constitutionality of plain error review. 7 N.W.3d 350 (Iowa 2024). Presumably, the contention that it did originates from the following excerpt: We are an error-correction court. Iowa Const. art. V, § 4 (stating the supreme court “constitute[s] a court for the correction of errors at law”); Iowa Code § 602.4102(1) (same). If an issue was never presented to the district court to rule on, and if the district court did not in fact rule on it, we lack any “error” to correct. Had the defendant raised the argument, we don’t know how the district court would have ruled on it.
Id. at 355 (alteration in original).
14 Two points. First, I do not read the Gomez Medina court’s citation to the Iowa Constitution as holding that the precepts of plain error review and correction for errors at law are constitutionally incompatible. The passage quoted above is merely the court’s explanation of the ordinary rule of error preservation—not a constitutional adjudication of whether plain error review could ever coexist with our jurisdictional principles. The opinion never mentions plain error. No party briefed it. The court was never asked whether article V, section 4, forbids it. Cf. Webster v. Fall, 266 U.S. 507, 511 (1925) (“Questions which merely lurk in the record, neither brought to the attention of the court nor ruled upon, are not to be considered as having been so decided as to constitute precedents.”). To read Gomez Medina as constitutionally forbidding plain error review is to commit the very sin the error-preservation doctrine exists to prevent: treating an issue as decided that was never raised.
Second, the missing error in Gomez Medina is missing for an epistemic reason: no ruling exists to evaluate. But plain error applies to the opposite situation—error apparent on the existing record, where the court isn’t guessing. So, while Gomez Medina sets the floor of error correction; I would argue it doesn’t bar reaching a manifest unpreserved error. Tellingly, the court treats preservation as prudential, not jurisdictional—it reviews de novo and reaches the statutory merits anyway. Gomez Medina, 7 N.W.3d at 356–57.
Our supreme court did reject adopting plain error in State v. Treptow— but as an unexplained institutional choice rather than a constitutional impossibility: “We have repeatedly rejected plain error review and will not adopt it now.” 960 N.W.2d 98, 109 (Iowa 2021). The opinion refutes incompatibility itself: for fifty years Iowa reached unpreserved trial errors on direct appeal “under the rubric of ineffective assistance of counsel.” Id. at
15 118 (Appel, J., dissenting). If an error-correction court could correct unpreserved errors for half a century under another label, reaching those errors is not beyond its function or capacity. Justice Appel’s dissent confirms the point: plain error rests within the court’s supervisory power, id. at 120 n.4 (Appel, J., dissenting), and both State v. Rutledge, 600 N.W.2d 324 (Iowa 1999) and State v. Graves, 668 N.W>2d 860 (Iowa 2003) illustrate that the same unpreserved error went uncorrected as “plain error” but “couched as a claim based on ineffective assistance of counsel.” Treptow, 960 N.W.2d at 118–19 (Appel, J., dissenting).
Gomez Medina and Treptow demonstrate that Iowa has not adopted plain error and now lacks a direct-appeal safety valve outside of that provide by section 814.7 (2025). Neither rest on constitutional incompatibility. Gomez Medina requires a reviewable ruling, see 7 N.W.3d at 356, which plain error supplies, and Treptow corrected unpreserved error under another name, see 960 N.W.2d at 109, which proves the function tolerates it.
I contend that plain error review is compatible with correction of errors at law and thus our constitution. I start from the premise that “correction of errors at law” describes the appellate court’s function, not a catalog of what errors it may reach. See Iowa R. App. P. 6.907. The standard tells us what the court is doing—reviewing legal rulings for correctness rather than reweighing facts or substituting discretion—and it tells us the posture is deferential to findings of fact but plenary as to law. It does not, by its own terms, define the universe of correctable errors. Error preservation is a separate doctrine, prudential in origin, that ordinarily limits what errors the court will reach. To treat preservation as built into the meaning of “errors at law” is to conflate the scope of review with the standard of review. They are distinct, and once distinguished, the supposed contradiction dissolves. On
16 this view, plain error is not an exception that suspends legal-error review; it is legal-error review applied to a particular class of error—one so clear and so consequential that the prudential reasons for the preservation rule fall away.
I find that the anti-sandbagging rationale against plain error review’s adoption unpersuasive. Most concerningly, “while encouraging one party to prevent errors, it encourages the other to commit errors, or at least to be careless about doing so.” Darryl K. Brown, Does It Matter Who Objects? Rethinking the Burden to Prevent Errors in Criminal Process, 98 Tex. L. Rev. 625, 642 (2020). This is the “mirror image of an injured party sandbagging.” Id. at 645. Sandbagging occurs when the injured party silently sits by while the other side makes an error in hopes that an appellate court will overturn their conviction later. Id. The party presentation rule, of course, seeks to prevent sandbagging. See Crawford, 972 N.W.2d at 199. But without plain error review, the erring party can also idly sit by while committing error, and if the error slips by defense counsel, it will be unreviewable on appeal. Plain error review ensures that prosecutors do not inject error into a trial in hopes that defense counsel—or a pro se defendant—may miss it. See United States v. Young, 470 U.S. 1, 7 (1985). Finally, the anti-sandbagging rationale assumes a litigant could have raised the issue and strategically withheld it. But a plain error, by definition, is one the trial court should have caught on its own. Thus, the error is the court’s.
Structural objections—that one posture presupposes preservation and the other presupposes its absence—are equally unpersuasive. Preservation is not a jurisdictional bar; it is a rule of orderly adjudication. And as addressed above, Iowa courts already reach certain unpreserved questions: subject- matter jurisdiction, illegal sentences under Iowa Code section 814.10, and
17 other issues of an exceptional character the court has chosen to address. If legal error review was genuinely incompatible with review of unpreserved errors, none of these settled practices could exist. They do exist. And plain error would be one more entry in a list the doctrine already tolerates, not a foreign body the doctrine must reject.
Adopting plain error review would not break faith with our appellate tradition—it would recover a principle that animated our judiciary from its earliest days. Long before the modern error preservation orthodoxy hardened—before the late Justice Harris could write that “[n]othing is more basic in the law of appeal and error than the axiom that a party cannot sing a song to us that was not first sung in trial court,” State v. Rutledge, 600 N.W.2d 324, 325 (Iowa 1999), the Iowa territorial and early state courts understood their reviewing function more capaciously. For example, in David v. Ransom, our supreme court held that it will not be confined exclusively to an examination of the errors assigned, and that “[t]he writ of error may be supported by any error appearing on the face of the record,” because it is “the province of all courts for the correction of errors to act upon the whole bill without distinction, even though counsel do not specially refer, by assignment or argument, to the particular matter upon which the decision turns.”4 1 Greene 383, 388 (Iowa 1848).
4 One might object that David v. Ransom addresses the scope of assigned error review (waiver), not error preservation as we now conceive it. The point is fair but misplaced. David is not offered as authority that early Iowa recognized our modern preservation doctrine—a doctrine, and a vocabulary, that postdate the decision. It is offered for what it reveals about how this court once understood its office: as a tribunal “for the correction of errors” whose province was the “whole bill” and whose review was “not confined exclusively to” the matters counsel chose to assign. The assigned error rule and the preservation rule are cousins, not strangers; both answer the same question— whether a reviewing court is a passive instrument of the parties’ framing or a guardian of
18 In Redman v. Williamson, the supreme court made the corollary explicit: the absence of a bill of exceptions did not necessarily preclude the possibility that there was error in the record, and where error appears on the face of the record, there need be no bill of exceptions at all. 2 Iowa 488, 488– 90 (Iowa 1856). These cases reflect a settled premise—visible too in Dunham v. Benedict, 1 Greene 74, 74–75 (Iowa 1847), and Chapman v. Arnold, 1 Greene 368, 368–69 (Iowa 1848)—that the reviewing court’s duty to correct manifest legal error did not rise or fall on the parties’ procedural exactitude.
That tradition is consistent with the Iowa Constitution. Article V designates our supreme court as “a court for the correction of errors at law,” exercising “a supervisory and administrative control overall all inferior judicial tribunals.” Iowa Const. art. V, § 4. As Cooper v. Armstrong 5 recognized, that constitutional charge defines the court’s essential character: it exists to correct legal error. 3 Greene 120, 122 (Iowa 1851). A court constitutionally charged with the correction of errors possesses, as an attribute of that charge, the authority to notice and correct a clear error that would otherwise work an injustice—precisely the function plain error review serves. The early decisions confirm this: even while insisting that error be shown, the court read its mandate as reaching errors apparent on the face of
the judgment’s correctness. Thus, a court historically unwilling to be confined to assigned error is difficult to recast as one now powerless to notice plain error not raised below in modernity. 5 Cooper quotes article V, section 3, of the 1846 Iowa Constitution, which gave our supreme court “appellate jurisdiction only in all cases in chancery, and shall constitute a court for the correction of errors at law.” 3 Greene at 121. The aforementioned “supervisory and administrative control” language and the “§ 4” designation come from the present corollary constitutional provision. Iowa Const. art. V, § 4.
19 the record regardless of assignment, and in criminal matters directed courts “to overlook slight technical objections” so that substance might prevail over form. Zumhoff v. State, 4 Greene 526, 530–31 (Iowa 1854).
“We are an error-correction court,” Gomez Medina, 7 N.W.3d at 355, describes a practice; it does not delimit a power. If article V, section 4, truly stripped appellate courts of power over unpreserved errors, the rule would admit no exceptions, because jurisdictional limits cannot be waived, forgiven, or excused for good behavior. See State ex rel. Vega v. Medina, 549 N.W.2d 507, 508 (Iowa 1996). Yet our law is honeycombed with exceptions, many of ancient vintage and all of them tolerated by the same constitution: void sentences may be corrected at any time, State v. Ohnmacht, 342 N.W.2d 838, 843 (Iowa 1983), a defendant need not object mid-sentencing to the court’s exercise of discretion, State v. Cooley, 587 N.W.2d 752, 754 (Iowa 1998), and recently, our supreme court reached a concededly waived argument because the law was clear and the precedent at stake, Christensen, 21 N.W.3d at 532. If the incompatibility thesis were sound, each of these would be a constitutional violation rather than a doctrinal choice. The supreme court has never argued that. And the supreme court’s decisions forgiving error- preservation in certain scenarios is intelligible only if preservation is a prudential rule of judicial administration, not a constitutionally forged jurisdictional command.
The phrase “correction of errors at law” in article V, section 4, draws the line our framers cared about in 1857: the line between law and equity.6 See
6 The framers of our current 1857 constitution inherited the distinction between law and equity from the Iowa Organic Act of 1838 that established the territory of Iowa. The act gave the territorial supreme court “a chancery as well as common law jurisdiction,” then allowed “writs of error, bills of exception, and appeals in chancery causes. . . under such regulations as may be prescribed by law.” Organic Law of Iowa § 9,
20 In re Est. of Custer, 295 N.W. 848, 851 (Iowa 1941). Cases in chancery were heard de novo; cases at law came up on a record for the correction of legal error. Id. at 851. Thus, the clause allocates modes of review. “Correction of errors at law” describes the character and scope of appellate review not a rule about who must object to review the error. Id. (“Rights substantive are embalmed in the constitution—rights adjective, or the modes of asking for the former, are accidental—have been always flexible—never have been, in many particulars, well defined. . . . We can not consent to return again to the dead mumblements of the past, and we know that neither the people, nor the bar, nor the bench would allow it.” (quoting Rep. of the Comm’rs of Civ. Prac., reprinted in footnote to Iowa Revision of 1860 § 2608)). Thus, the flaw
1838 Iowa Acts 37. The writ of error only reviewed issues raised and adjudicated at the district court. See, e.g., Wallis v. Sparks, Morris 20, 20–21 (Iowa 1839) (“Writs of error will only lie where there has been a final judgment in the court below.”). Appeals in chancery were heard de novo, allowing the court to readjudicate both the facts and the law of the case. Blake v. Blake, 13 Iowa 40, 42 (Iowa 1862). Appeals by writ of error were restricted to matters of law, Brazelton v. Jenkins, Morris 15, 16 (Iowa 1843), while appeals in chancery were limited to cases in equity. Bills of exception did not fall under a mode of review. Rather, it was the era’s vehicle for lodging an objection on the record. The 1844 constitutional convention compressed that scheme into the binary constitutional formula in the never-ratified 1844 constitution, and the 1857 constitutional convention carried that provision forward untouched. See 2 The Debates of the Constitutional Convention of the State of Iowa 1067–96 (W. Blair Lord rep., 1857) (parallel texts). Its grammar was the grammar of the age: “An appeal is a process of civil law origin, and removes a cause entirely; subjecting the fact as well as the law, to a review and re-trial: [but] a writ of error is a process of common law origin, and it removes nothing for re-examination but the law.” See Wiscart v. D’Auchy, 3 U.S. 321, 327 (1796). The first General Assembly read it the same way: the Code of 1851 gave civil cases an “appeal,” Iowa Code ch. 114 (1851); criminal cases a “writ of error,” id. chs. 184–185; and kept “bills of exceptions” in a chapter of their own, id. ch. 179. Objection practice in 1851 was, as it was in 1838, a creature of procedure—never of the grant. The proviso says as much: a restriction the general assembly “may, by law, prescribe” is, by definition, one the constitution does not itself impose. See Iowa Const. art. V, § 4.
21 in the syllogism; the premise—“if the issue was never presented, we lack any error to correct”—conflates the existence of an error with the preservation of a claim.
A district court errs when it misapplies the law, whether or not anyone points that out. Forfeiture affects the remedy a party may demand, not the existence of the mistake. The federal framework makes the distinction explicit: the first prong of plain error review asks whether there was “error” at all—a deviation from a legal rule—and treats the failure to object as bearing on the standard of correction, not on whether anything went wrong. Olano, 507 U.S. at 732–34. Plain error review does not invent errors the district court never made. It corrects errors the district court did make. To adopt plain error review, is not to graft a federal innovation onto Iowa soil but to restore something the territorial and early state courts already practiced— a reviewing function broader than the preservation rule we later imposed on ourselves. The Iowa Supreme Court—mindful of its founding charge to correct errors at law—remains constitutionally free to reclaim that heritage.7
Correction of errors at law exists because the legal correctness of the judgment matters. A regime that lets a clear, outcome-determinative legal error stand—solely because no one objected at the right moment—elevates procedure over the very correctness the standard is meant to vindicate. Plain error review reconciles the two: it preserves the preservation rule for the ordinary case while ensuring that the rare, grievous legal mistake does not
7 Because Iowa Code section 602.5103 (2024) simply mirrors the language in article V, section 4, the interpretive conclusion reached above is no different in substance. See Iowa Code § 602.5103(1) (“The jurisdiction of the court of appeals is coextensive with the state. The court of appeals has appellate jurisdiction only in cases in chancery, and constitutes a court for the correction of errors at law.”). Thus, correction of errors at law is not incompatible with plain error review in constitutional text, case law, or statute.
22 survive appeal merely on a technicality. Not only is plain error review compatible with our Iowa Constitution, it honors article V, section 4’s animating purpose.
III. Plain Error Review Complements Section 814.7
Iowa Code section 814.7 states: An ineffective assistance of counsel claim in a criminal case shall be determined by filing an application for postconviction relief pursuant to chapter 822. The claim need not be raised on direct appeal from the criminal proceedings in order to preserve the claim for postconviction relief purposes, and the claim shall not be decided on direct appeal from the criminal proceedings.
Section 814.7 speaks to one question only: the forum in which claims against counsel must be brought. It says nothing about appellate review of the court’s error on direct appeal. Adopting plain error review would not evade the legislature’s judgment; it would respect it, by giving each doctrine its proper object.
The case against plain error review in Iowa has rested on an illusory substitution: that the ineffective-assistance route through postconviction relief is an equivalent remedy for unpreserved trial error, merely housed in a different proceeding. It is not. The two doctrines address fundamentally different actors. A Strickland claim indicts counsel. See Strickland v. Washington, 466 U.S. 668, 685–86 (1984) (“[T]he right to counsel is the right to the effective assistance of counsel.”). Plain error review corrects the court. When a district court delivers a self-defeating jury instruction or admits plainly inadmissible evidence, the judicial system itself has erred—and routing the defendant’s only recourse through an attack on his own lawyer’s competence misattributes who is responsible for the error. Collapsing the two does violence to both doctrines. It forces Strickland to carry institutional
23 freight it was never built for, and it leaves judicial error formally uncorrectable in the very forum designed to correct it.
This doctrinal mismatch has been addressed by both the Colorado and Michigan Supreme Courts. The Colorado Supreme Court held that plain error and Strickland prejudice are distinct standards measuring different things, with plain error demanding a greater impairment of the judgment’s reliability than Strickland requires. Hagos v. People, 288 P.3d 116, 119–22 (Colo. 2012) (en banc). Plain error review reflects a “careful balancing of our need to encourage all trial participants to seek a fair and accurate trial the first time around against our insistence that obvious injustice be promptly redressed.” Plain error review allows the opportunity to reverse convictions in cases presenting particularly egregious errors, but reversals must be rare to maintain adequate motivation among trial participants to seek a fair and accurate trial the first time.
Ineffective assistance of counsel claims, however, do not require the same balancing of interests.
Id. at 121–22 (internal citation omitted). The standards diverge because the claims diverge. To reach a court’s error through Strickland, the defendant must first prove a constitutionally deficient performance by counsel—a showing wholly unnecessary to the question whether the court erred—and must do so years later, on a cold record, against the strong presumption of strategic judgment. The United States Supreme Court itself has recognized that some claims of trial error are fully visible on the direct- appeal record and need no collateral factual development. See Massaro v. United States, 538 U.S. 500, 508 (2003).
The Michigan Supreme Court has explained why the two cannot substitute for one another. An obvious judicial error—say, a trial court
24 permitting the prosecutor to comment extensively on a defendant’s post- arrest, post-Miranda silence—does nothing, without more, to carry the defendant’s burden under Strickland, “because Strickland’s two prongs ask different questions.” People v. Randolph, 917 N.W.2d 249, 254 (Mich. 2018). Plain error review asks whether the court erred and whether the error was clear; ineffective-assistance review asks whether counsel’s silence was constitutionally deficient and prejudicial. And because “counsel may decide, for strategic reasons, not to object to an obvious error,” a plausible strategy defeats the deficiency prong no matter how plain the underlying error. Id. (cleaned up). The result is a perverse inversion: the defendant whose lawyer made a defensible tactical choice in the face of a glaring judicial error has the weakest ineffective-assistance claim, precisely because the strategy was reasonable—yet the judicial error remains, uncorrected and uncorrectable. Michigan offers this reasoning to explain why plain error review and Strickland operate side by side, each catching what the other cannot. See id. at 256 (“[A] court . . . cannot simply plug in the plain-error prejudice analysis for the ineffective-assistance prejudice analysis.”). Iowa has removed one of the two nets entirely and then moved the remaining one out of the appellate courts altogether. See Iowa Code § 814.7. The error that slips through is not caught later in a different forum. It is simply never caught.
Until 2019, Iowa could plausibly claim the gap was theoretical, because ineffective-assistance claims supplied a safety valve on direct appeal when the record was adequate. See State v. Fountain, 786 N.W.2d 260, 263 (Iowa 2010). The legislature closed that valve. See Iowa Code § 814.7. And our supreme court has continued to hold the other door shut. See Treptow, 960 N.W.2d at 109 (“We have repeatedly rejected plain error review and will not adopt it now.”). The combination—no ineffective-assistance claims on direct appeal and no plain error review—has made Iowa nationally singular.
25 In the federal courts and forty-nine states, a defendant may raise an unpreserved error on direct appeal through one route or the other. Iowa alone categorically bars both. See id.at 117–18 (Appel, J., dissenting); Commonwealth v. Holmes, 79 A.3d 562, 563–64 (Pa. 2013) (recognizing exceptions to postconviction-relief requirements for record-apparent, meritorious claims where immediate review serves the interests of justice, or good cause is shown with express postconviction relief waiver); Commonwealth v. Delgros, 183 A.3d 352, 360–61 (Pa. 2018) (adopting a third exception to postconviction-relief requirements when the defendant is statutorily ineligible for postconviction-relief review). Even Pennsylvania, our supposed companion in the plain review minority, permits record-based ineffective-assistance claims on direct appeal—and has been relaxing, not tightening, its preservation rules. Commonwealth v. Bradley, 261 A.3d 381, 401 (Pa. 2021) (“[W]e hold that a [postconviction relief ] petitioner may, after a [postconviction relief ] court denies relief, and after obtaining new counsel or acting pro se, raise claims of [postconviction relief ] counsel’s ineffectiveness at the first opportunity to do so, even if on appeal.”).
Postconviction relief does not deliver the same remedy later. It delivers a harder remedy slower. The defendant whose trial was infected by obvious judicial error must wait through years of collateral litigation. Chief Judge Tabor, in her special concurrence in State v. Young, catalogued postconviction-relief actions pending for three, twelve, even thirty-four years—while bearing the added burden of proving counsel’s constitutional inadequacy. See State v. Young, No. 23-1924, 2025 WL 1452559, at *8 (Iowa Ct. App. May 21, 2025) (Tabor, C.J., specially concurring). “It’s cliché to say that justice delayed is justice denied. But it’s a cliché for a reason[:] it’s true.” Id. (citation omitted). Calling the two routes equivalent ignores the
26 differences in both the burden and delay between the two and does so on the back of Iowa’s already overburdened criminal defense bar.
Finally, the familiar objection—that plain error erodes the contemporaneous-objection rule—is answered by the very architecture of the doctrine Iowa would adopt. The Olano framework is a gauntlet. The defendant bears the burden of showing (1) error, (2) that the error is plain, and (3) it affects substantial rights. Even then, the appellate court retains discretion to correct the error only where the error seriously affects the fairness, integrity, or public reputation of judicial proceedings.. Olano, 507 U.S. at 732–36; Johnson v. United States, 520 U.S. 461, 466–67 (1997); Fed. R. Crim. P. 52(b); see also Rosales-Mireles v. United States, 585 U.S. 129, 137 (2018). This framework was built precisely to preserve the incentives to timely object while simultaneously preventing the preservation rule from becoming an instrument of injustice.
A smattering of our sister states’ experiences confirm the framework travels well. Minnesota adopted the federal three-prong test plus the discretionary fourth. State v. Griller, 583 N.W.2d 736, 740 (Minn. 1998); Minn. R. Crim. P. 31.02. North Dakota expressly imported the Olano framework. State v. Olander, 575 N.W.2d 658, 666 (N.D. 1998). Arizona reaches unpreserved fundamental error. State v. Escalante, 425 P.3d 1078, 1085 (Ariz. 2018). Oregon recognizes a plain-error exception to preservation. State v. Vanorum, 317 P.3d 889, 897 (Or. 2013). Montana exercises an inherent power of plain error review for constitutional claims. State v. Finley, 915 P.2d 208, 215 (Mont. 1996). Rhode Island and Virginia reach unpreserved error in extraordinary circumstances or where the ends of justice require. State v. Williams, 432 A.2d 667, 670 (R.I. 1981); Redman v. Commonwealth, 487 S.E.2d 269, 272 (Va. Ct. App. 1997). Forty-nine
27 laboratories have run this experiment. None reports that the contemporaneous-objection rule collapsed.
What remains then, is not a doctrinal obstacle but an institutional choice—and the choice now carries a cost it did not carry before 2019. A judicial system that cannot correct its own plain errors on direct appeal does not preserve its authority. It diminishes it. It diminishes that authority by announcing that the court’s mistakes will be answered, if at all, by litigating the competence of the defense bar in another forum, years later. See generally Jon M. Woodruff, Note, Plain Error by Another Name: Are Ineffective Assistance of Counsel Claims a Suitable Alternative to Plain Error Review in Iowa?, 102 Iowa L. Rev. 1811 (2017); Rhoades v. State, 848 N.W.2d 22, 33 (Iowa 2014) (Mansfield, J., concurring specially) (“In some respects, we are using ineffective assistance as a substitute for a plain error rule, which we do not have in Iowa.”). The legislature has spoken to counsel’s errors. It falls to the courts to speak to our own.
CONCLUSION As an intermediate appellate court judge, I am duty bound by Treptow, 960 N.W.2d at 109, which squarely rejected plain error review. Cf. State v. Hastings, 466 N.W.2d 697, 700 (Iowa Ct. App. 1990) (“We are not at liberty to overturn Iowa Supreme Court precedent.”). I therefore concur. But I would not do so if writing on a clean slate. Were the question open, I would accept Kyle Roberts’s invitation to adopt plain error review to review his evidentiary claims of error. Plain error review, properly understood, would supply exactly what the current framework lacks: predictability and consistent limiting principles. Its adoption would honor article V, section 4’s text and animating purpose. Its adoption would harmoniously complement
28 Iowa Code section 814.7. Its adoption exists in our cases; it awaits only a court with the authority to give it a name.