When the Jury Convicts of an Offense the Indictment Never Charged

The Verdict Must Conform to the Charging Instrument

The foundational rule is simple: a conviction cannot rest on conduct the indictment never alleged. Texas measures the evidence against a hypothetically correct jury charge that is authorized by the indictment, so a verdict that convicts of a separate, unalleged offense has no valid pleading to stand on.

In Malik v. State, the Court of Criminal Appeals held that the legal sufficiency of the evidence is measured by the elements of the offense as defined by a hypothetically correct jury charge — one that accurately states the law, is authorized by the indictment, does not unnecessarily increase the State’s burden, and adequately describes the offense for which the defendant was tried. The phrase “authorized by the indictment” is the anchor. It means the State is held to the offense it chose to plead; it does not get to convict on a different crime the grand jury never presented.

That principle sounds absolute, but the analysis quickly splits into distinct paths, because “an offense not charged” can describe several very different situations. Sometimes the jury convicted of a genuinely lesser-included offense, which the law permits. Sometimes the proof simply varied from the allegation, which is a sufficiency question. And sometimes the jury charge itself authorized conviction on a theory the indictment never contained, which is charge error reviewed for harm. Each path carries a different remedy, so the first task is to identify which one the case actually presents. The rest of this guide walks through them and shows where a motion in arrest of judgment fits and where it does not.

Notice is the value underneath the rule. The indictment tells the accused what to defend against, so a conviction on an offense it never named would punish conduct the defense had no fair chance to contest. That is why the analysis compares the pleading, the charge, and the verdict rather than simply asking whether the defendant did something wrong. The State is bound to the crime it chose to present to the grand jury, and the trial cannot silently enlarge that choice.

When Conviction of an Uncharged Offense Is Allowed: Lesser-Included Offenses

Not every conviction outside the exact words of the indictment is improper. Texas expressly permits conviction of a lesser-included offense, because the indictment for the greater offense gives notice of the lesser. The question is whether the lesser offense truly is included in the greater as the State pleaded it.

Article 37.09 defines a lesser-included offense, and Hall v. State adopted the cognate-pleadings test to apply it. Under Hall, a court compares the elements of the greater offense as alleged in the indictment with the elements of the potential lesser offense. This first step is a question of law that can be resolved from the pleading itself, before any evidence, because its purpose is notice. If proving the offense as charged necessarily proves the lesser offense, the lesser is included and a conviction on it is permitted. If the lesser offense requires proof of some fact not required by the greater as pleaded, it is not included.

Hall itself illustrates the limit: aggravated assault by threat was not a lesser-included offense of a murder indictment that alleged causing death by shooting, because the assault-by-threat theory required proof of threatening and display that the murder pleading did not. When the jury convicts on an offense that flunks the cognate-pleadings test, it has convicted of something the indictment never charged.

Because the comparison uses the indictment as actually drafted, the same lesser offense can be included in one case and not another, depending on how the State pleaded the greater crime. A murder indictment that alleged a threat or the display of a weapon might make assault by threat a lesser-included offense; one that alleged only a shooting, as in Hall, does not. The wording the State chose, not the evidence that later came in, controls the first step of the analysis.

Variance or a Different Offense Entirely?

A second path arises when the proof at trial differs from what the indictment alleged. Not every discrepancy is fatal. Texas asks whether the difference is a mere variance the law tolerates, a material variance that requires reversal, or a discrepancy so complete that it is a failure of proof.

In Gollihar v. State, the Court of Criminal Appeals held that a variance between the indictment and the proof is fatal only if it is material and prejudices the defendant’s substantial rights — that is, if the pleading failed to give adequate notice to prepare a defense or would not bar a second prosecution for the same offense. Immaterial variances are disregarded. But Byrd v. State drew a sharper line for cases where the mismatch is total: when the State alleges one owner or one item of property and proves an entirely different person or thing, that is not a variance at all — it is a failure of proof that entitles the defendant to an acquittal.

The distinction matters because it dictates the remedy. A material variance and a failure of proof both go to legal sufficiency, which is an acquittal-or-appeal question, not a motion-in-arrest-of-judgment question. The table in the preceding section shows what is permitted; the variance analysis shows what to do when the proof drifts. If the drift is immaterial, the conviction stands. If it is material or amounts to a failure of proof, the answer is a legal-sufficiency challenge aimed at acquittal, litigated in the trial court and on appeal, rather than an arrest of judgment.

The remedy also drives how the claim is preserved. A legal-sufficiency challenge, including a material-variance or failure-of-proof claim, needs no objection at trial and may be raised for the first time on appeal, because it asks whether any rational jury could have convicted of the charged offense. That makes the sufficiency route both powerful and forgiving of trial-level silence, but it aims at acquittal, not at the charging instrument and not at a retrial.

When the Jury Charge Authorized an Unalleged Theory

A third path is charge error. Sometimes the indictment is fine, but the jury charge authorizes conviction on a theory or offense the indictment never alleged. Under current Texas law, submitting a non-lesser-included offense is charge error reviewed for harm — it is not, by itself, a jurisdictional defect that voids the conviction.

This is the crucial modern correction. In Trejo v. State, the trial court submitted an offense that was not a lesser-included offense of the charged crime. The Court of Criminal Appeals held that because a valid indictment had invoked the court’s jurisdiction, the court “had jurisdiction to commit the error” — the mistaken submission was charge error subject to a harm analysis, not a jurisdictional void that could be raised at any time. That reasoning routes most “unalleged theory” complaints into the ordinary framework for jury-charge error, where preservation and harm control the outcome.

Sufficiency review handles the overlap cleanly. Fisher v. State holds that when a jury instruction broadens the basis for conviction beyond the indictment, a reviewing court measures sufficiency against the indictment as properly incorporated, disregarding the language that expanded the theory. In other words, the State cannot bootstrap a conviction on a theory it never pleaded by slipping it into the charge. Between Trejo and Fisher, the message is consistent: an unalleged theory in the charge is a charge-error and sufficiency problem, addressed through objection, harm analysis, and appeal — not through the jurisdictional door.

Preservation changes the harm standard for charge error. A theory-expanding instruction that drew a timely objection is reviewed for “some harm,” while an unobjected-to charge error is reviewed for the tougher “egregious harm” standard. Either way, the inquiry is whether the erroneous charge actually hurt the defense, not whether the court lost its power to act. That is the practical meaning of Trejo’s holding that the court had jurisdiction to commit the error.

Which Remedy Fits: Arrest of Judgment, Acquittal, or New Trial?

Because “an offense not charged” covers several problems, the remedy has to match the flaw. Legal insufficiency and failure of proof point toward acquittal; jury-charge error points toward harm analysis and, if harmful, a new trial or appellate reversal; a verdict substantively defective in relation to the indictment can support a motion in arrest of judgment.

Rule 22.2(b) lists as a ground for arrest of judgment that “in relation to the indictment or information a verdict is substantively defective.” That reaches a verdict that, on the face of the record, does not correspond to the offense the instrument charged. But a motion in arrest of judgment does not weigh evidence, so it is the wrong tool for a pure sufficiency dispute. Under Malik and Byrd, a claim that the State failed to prove the charged offense is a legal-sufficiency claim; if it succeeds, the result is an acquittal, not a do-over. And under Trejo, a charge that authorized an unalleged theory is trial error tested for harm — typically raised by objection, then on appeal, with a new trial as the remedy if the error was harmful.

Practically, the defense often pleads in the alternative to preserve every avenue: a legal-sufficiency challenge for acquittal, a jury-charge complaint for harm review, and, where the record shows the verdict does not conform to the indictment, a motion in arrest of judgment. Sorting the claim correctly is not a formality — filing a sufficiency argument as an arrest-of-judgment motion, or a charge complaint as a jurisdictional attack, can forfeit the strongest relief available. A companion page compares arrest of judgment and the motion for new trial in detail.

Sequencing the alternatives matters. Because a legal-sufficiency win yields an acquittal that bars retrial, it is generally the most valuable outcome and is preserved for appeal regardless of trial objections. Charge-error and record-defect theories are pursued alongside it, not instead of it, so that a court’s rejection of one classification does not sink the others. The strongest posture presents each theory under its own standard rather than collapsing them into a single motion.

Building the Record: Lining Up Indictment, Charge, and Verdict

Every version of this issue is proved from three documents read together: the indictment, the application paragraph of the jury charge, and the verdict form. Laying them side by side shows whether the conviction rests on the charged offense, a permitted lesser-included offense, or something the pleading never contained.

Start with the indictment and identify precisely what offense and theory the grand jury alleged, element by element. Then read the application paragraph — the part of the charge that actually authorizes a guilty finding — and compare its elements to the indictment’s. Finally, check the verdict form to see what the jury actually found. If the application paragraph or verdict names an offense that requires proof of a fact the indictment never alleged, the cognate-pleadings analysis from Hall tells you it was not a lesser-included offense, and the conformity problem is real. If the mismatch is between the allegation and the evidence rather than the pleadings, the variance analysis from Gollihar and Byrd applies instead.

Preservation shapes what is left. A defect in the indictment itself must have been raised before trial under article 1.14(b), or it is forfeited — a point covered on the companion charging-defect page. Jury-charge error should be raised by a timely objection to maximize the harm standard on review, though even unobjected charge error can be examined for egregious harm. Sufficiency needs no trial objection at all and can be raised for the first time on appeal. Because the same underlying facts can present as a pleading defect, a variance, or charge error, the record comparison is what tells counsel which claim — and which remedy — the case truly supports.

The verdict form deserves special attention in multi-count or multi-theory cases. A general verdict that does not say which theory the jury adopted can obscure whether the conviction rested on an alleged or an unalleged basis. Where the record leaves that ambiguous, the disciplined comparison of indictment, application paragraph, and verdict becomes the only reliable way to test whether the jury was authorized to convict on the theory the State actually pleaded.

Where This Fits

This guide is one of four situations that arise under the Motion in Arrest of Judgment. Start with the parent motion for the overall framework, or move to a related fact pattern:

What the Case Law Says

These decisions—verified against primary sources—control how this issue is litigated. Every case still turns on its own facts.

• Malik v. State, 953 S.W.2d 234 (Tex. Crim. App. 1997) — Legal sufficiency is measured against a hypothetically correct jury charge that is authorized by the indictment, not the charge actually given.
• Hall v. State, 225 S.W.3d 524 (Tex. Crim. App. 2007) — Whether an offense is lesser-included is decided by comparing the elements of the greater offense as pleaded with the lesser offense (cognate-pleadings test).
• Gollihar v. State, 46 S.W.3d 243 (Tex. Crim. App. 2001) — A variance between indictment and proof is fatal only if it is material and prejudices the defendant's substantial rights.
• Byrd v. State, 336 S.W.3d 242 (Tex. Crim. App. 2011) — Proving an entirely different owner or property than alleged is a failure of proof entitling the defendant to acquittal, not a mere variance.
• Trejo v. State, 280 S.W.3d 258 (Tex. Crim. App. 2009) — Submitting a non-lesser-included offense is charge error reviewed for harm; a validly invoked court has jurisdiction to commit that error.

General summaries of published opinions for information only — not predictions about any specific case.