Thompson v. State

delivered the opinion of the Court

In this case we consider the scope of one of the more difficult penal code provisions, *789which deals with one variant of the law of “transferred intent” in Texas. Texas Penal Code § 6.04(b)(1) provides:

A person is nevertheless criminally responsible for causing a result if the only difference between what actually occurred and what he desired, contemplated, or risked is that ... a different offense was committed.

We conclude that the provision can be used under certain circumstances to transfer intent from a lesser offense to a greater offense, even when those offenses are contained within the same penal code section. That conclusion comes with caveats, which we will discuss below. Perhaps the most important caveat is that a defendant subject to this type of transferred intent instruction is entitled (upon request) to a mistake of fact instruction. Finding that the trial court’s “transferred intent” instruction was not erroneous, we affirm the judgments of the courts below.

I. BACKGROUND

A. The Incident

Appellant was an associate pastor at a Baptist church. His twin brother, Caleb Thompson, was also active in the church. The victim was an eleven-year-old boy who attended a children’s Bible-study program at the church. On July 3, 2002, the victim’s Bible-study teacher reported to appellant that the child was misbehaving. Appellant drove the child to Caleb’s nearby residence. At some point, Caleb joined them. Appellant beat the child with a tree branch. He struck the victim more than 100 times during a period estimated by the child at one-and-a-half hours. During at least part of that time, Caleb helped hold the child down. As a result of the beating, the victim’s back was one huge bruise from his neck to his buttocks. A paramedic testified that it was the worst bruising he had ever seen. The victim’s blood pressure was low, his heart rate was fast, and he appeared to be undergoing hypovolemic shock, which is an indication that he was losing blood. A doctor testified that the bruising was severe and palpable, indicative of deep tissue bruising, and that the victim’s urine was “coca-cola colored,” indicating collection in the kidneys of a substance called myoglobin, which is released into the blood as a result of the death of muscle cells. The doctor further testified that, as a result of this condition, the child would have died from renal failure if he had not received prompt medical attention.

B. Trial

The injury to a child offenses at issue here are:

(1) intentionally or knowingly causing serious bodily injury, a first-degree felony, and

(2) intentionally or knowingly causing bodily injury, a third-degree felony.1

Appellant was charged with the first-degree felony of injury to a child and the second-degree felony of aggravated assault.2 Appellant’s jury charge contained *790instructions that tracked the language of the charged offenses, along with allegations necessary to support a deadly weapon finding.3

The charge also contained two sets of instructions, to which appellant objected, that applied the doctrine of “transferred intent” found in Texas Penal Code § 6.04(b)(1). First, the charge contained an abstract instruction tracking the language of that provision. Second, with respect to the injury to a child offense, the charge contained an application paragraph that permitted the jury to find appellant guilty of the first-degree felony if he merely intended to cause bodily injury, so long as he actually caused serious bodily injury:

Now, bearing in mind the foregoing instructions, if you believe from the evidence, beyond a reasonable doubt that the defendant ... intending to cause bodily injury to [L.G.], a child 14 years of age or younger, by striking [L.G.] with a stick, branch, or an object unknown to the Grand Jury, did then and there cause serious bodily injury to [L.G.], a child 14 years of age or younger, by striking [L.G.] with a stick, branch or object unknown to the Grand Jury, and [the defendant] did then and there use or exhibit a deadly weapon, to wit: a stick, a branch, or an object unknown to the Grand Jury, during the commission of this offense, in that the manner of its use or intended use was capable of causing death or serious bodily injury, then you will find the defendant ... guilty as alleged in Count I....

Appellant was convicted of both offenses and sentenced to confinement for twenty-six years for the offense of injury to a child and for twenty years for the offense of aggravated assault.

C. Appeal

On appeal, appellant contended that the jury charge improperly allowed the jury to elevate the third-degree offense of injury to a child (intentionally or knowingly causing bodily injury) to the first-degree offense of injury to a child (intentionally or knowingly causing serious bodily injury). The State agreed that the “transferred intent” instructions should not have been given but argued that the error was harmless. The court of appeals, however, relying primarily upon Honea v. State,4 held that no error occurred.5 After addressing the arguments of both parties, the court of appeals held that “appellant’s intent to cause bodily injury to L.G. ‘transferred’ to the serious bodily injury that actually resulted from appellant’s conduct.”6

D. Discretionary Review

Appellant contends that Honea was wrongly decided, and he criticizes the opinion for failing to seriously analyze the issue. He complains that a literal application of § 6.04(b)(1) would result in an extraordinarily broad expansion of criminal liability that the legislature could not possibly have intended.

*791The State has modified its position somewhat from the position taken before the court of appeals. The State now concludes that § 6.04(b)(1) applies only when the offense intended and the offense committed appear in different statutory sections. The State also suggests the Block-burger7 “same elements” test as a possible alternative method of determining whether the offenses are different.

In addition, the State contends that two other factors may limit the scope of the statute in a way that avoids any unnecessarily harsh effect. First, relying upon language in §§ 6.04 and 6.03, the State suggests that a lesser culpable mental state cannot be transferred to a greater mental state. For instance, a defendant who acted only with the culpable mental state of “recklessness” could never be penalized for a crime that required knowledge or specific intent. Second, the State contends that the statute imposes a foreseeability requirement. Although the State acknowledges that § 6.04(b)(1) contains no explicit language to that effect, it believes that we could read into the statute an implicit requirement that the offense committed be “reasonably foreseeable.”

II. ANALYSIS

A. Precedent

In Honea, we construed § 6.04(b)(1).8 The defendant in that case bound and gagged the victim in a barn and stole $1,200 from his shut pocket.9 As a result of lying bound and gagged on the barn floor, the victim inhaled dust, which caused him to cough, vomit, and eventually suffocate.10 The defendant contended that there was a fatal variance between the allegations in the aggravated robbery indictment and the proof at trial because he did not intentionally and knowingly cause serious bodily injury.11 Relying upon § 6.04, this Court held that, because the defendant clearly intended to rob the victim and his acts resulted in the offense of aggravated robbery, his intent to rob transferred to the aggravated robbery.12 In arriving at this conclusion, the Court also cited former Article 42 (a predecessor to § 6.04(b)) and two cases that construed that statute.13

The defendant also complained of the trial court’s denial of his request for a mistake of fact instruction.14 Paraphrasing the predecessor statute, the Court summarily held that “no mistake of fact issue was raised.”15

Honea stands contrary to the propositions advanced by appellant and some of the propositions advanced by the State. Obviously, the Court held that the transferred intent statute could be used to transfer a defendant’s culpable mental state from a lesser offense to one that carries a greater penalty. Indeed, the transfer in that case occurred between offenses that would be considered greater and lesser-included offenses under the Blockburger test. Also, Honea did not mention any implied foreseeability requirement.

*792In addressing appellant’s contention that Honea was wrongly decided, we turn to principles of statutory construction. In doing so, we keep in mind that the “doctrine of stare decisis indicates a preference for maintaining consistency” with past decisions, especially those that interpret statutory enactments.16

B. Statutory Construction

1. General Principles

In cases of statutory construction, we give effect to the plain meaning of the statutory text, unless the language is ambiguous or the plain meaning leads to absurd consequences that the Legislature could not possibly have intended.17 In such event, we may resort to extratextual factors.18

Honea was decided before Boykin, however, and appears to have relied to some degree on the predecessor statute. So, even without a finding of ambiguity or absurd results, an examination of extratex-tual sources may give insight into the validity of Honea’s holdings.19

2. Statutory Language

With these principles in mind, we begin with the statutory language. Texas Penal Code § 6.04 provides, in its entirety:

(a) A person is criminally responsible if the result would not have occurred but for his conduct, operating either alone or concurrently with another cause, unless the concurrent cause was clearly sufficient to produce the result and the conduct of the actor clearly insufficient.

(b) A person is nevertheless criminally responsible for causing a result if the only difference between what actually occurred and what he desired, contemplated, or risked is that:

(1) a different offense was committed; or

(2) a different person or property was injured, harmed, or otherwise affected.

While the present case involves the construction of subsection (b)(1), the remainder of the statute provides some useful context from which to determine the meaning of that subsection. We make two observations about the statutory language. First, subsection (b) is worded to discount the significance of certain types of differences. That is, the subsection makes the difference legally irrelevant, at least for causation purposes. Second, the statute contains two “transferred intent” provisions that use parallel language. They both turn on the “only difference” being a particular type of fact. In subsection (b)(1), that different fact is a different offense while, in subsection (b)(2), that different fact is a different victim or object of the crime.

Appellant appears to concede that his desired interpretation finds no support in the language of subsection (b)(1), but he argues that adherence to the literal text leads to absurd results. On the other hand, the State’s proposed limitation of the statute’s applicability to offenses with the same mental state is consistent with the language of the statute and is inherent in the notion that the defendant’s culpable *793mental state transfers.20 But we do not agree that the State’s other proposed limitations follow from the statutory language.

As for the State’s proposed foreseeability requirement, the language of the statute does not support such a requirement at all, even by implication. That fact becomes clear when one considers the effect of applying the State’s reasoning to “different persons” under the parallel provision codified by subsection (b)(2). The phrase “the only difference ... is that ... a different person ... was injured” plainly does not require the State to prove that injury to a third person was reasonably foreseeable. Substituting the word “offense” for “person” should not yield a different result.

The State reads the language “different offenses” as creating a double-jeopardy-like relationship issue, which means that whether offenses are “different” depends upon their relationship to each other. But “different offenses” could be understood, on a more basic level, to mean different legal theories upon which a conviction could be procured, regardless of how those legal theories might be related to each other.21 Under the double-jeopardy-type analysis, greater and lesser-included offenses would not be “different.” But under the more basic approach, even a lesser-included offense would be a “different” offense from the offense charged. This latter, more basic, understanding would seem most consistent with the structure of the statute. It seems a little odd to say that a provision whose purpose is to discount the significance of an offense being “different” would attribute significance to the fact that the offense is not so different after all. But a “same statute” or “same elements” test would do just that: the perpetrator could not escape responsibility on the ground that the offense committed is “different” from the one intended but could do so on the ground that the two offenses are really the “same.”

Although the parties have not raised an issue with respect to the mistake of fact defense found in Texas Penal Code § 8.02, consideration of that section is necessary to our interpretation of § 6.04(b)(1). The mistake of fact defense provides:

(a) It is a defense to prosecution that the actor through mistake formed a reasonable belief about a matter of fact if his mistaken belief negated the kind of culpability required for commission of the offense.

(b) Although an actor’s mistake of fact may constitute a defense to the offense charged, he may nevertheless be convicted of any lesser included offense of which he would be guilty if the fact were as he believed.22

In this case, a mistake of fact regarding the seriousness of L.G.’s injury would seem to negate the charged mental state of intent to cause serious bodily injury. Because appellant intended to commit bodily injury, which is also an offense, the intentional culpable mental state would “transfer” to the serious bodily injury actually committed, but he would have a defense, so long as his mistaken belief about the type of injury he was inflicting was reasonable. Of course, he would still be guilty of the lesser-included offense of intending to cause bodily injury.

*794So, a plain meaning construction of the relevant statutes seems to support Ho-nea’s interpretation of the transferred intent provision but appears to conflict with its related holding that the mistake of fact defense is inapplicable.

3. Extratextual Sources

Before the enactment of the current Penal Code in 1973, Texas was governed by the Penal Code of 1948, which contained the precursors to the statutory provisions at issue today. A defense of mistake of fact23 and three “mistake” provisions that imputed liability from one offense to another24 were codified within the same chapter in immediate succession. An understanding of the development of the law on mistake of fact is critical to an understanding of the law on transferred intent.

The 1948 mistake of fact defense provided:

If a person laboring under a mistake as to a particular fact shall do an act which would otherwise be criminal he is guilty of no offense, but the mistake of fact which will excuse must be such that the person so acting under a mistake would have been excusable had his conjecture as to the fact been correct, and it must also be such mistake as does not arise from a want of proper care on the part of the person so acting.25

Three important observations can be made about this defense. First, the mistake did not have to negate the culpable mental state required for the offense; all that had to be shown was that no offense would have been committed if the mistaken belief had been correct. Second, the defense contained a requirement that the mistake “not arise from a want of proper care,” which is roughly equivalent to the current statutory requirement that the defendant’s mistaken belief be “reasonable.” Third, the defense did not specify that a mistake could result in liability for a lesser-included offense; rather, the statutory language suggested that the defense applied only when the mistake would completely exonerate the defendant of any offense.

The culpability and reasonableness aspects of the defense (the first two observations made about the statute) were discussed by the Court in Green v. State.26 In that case, the defendant was charged with stealing hogs.27 The jury was instructed on the defense of mistake of fact, with the instruction tracking the statutory language.28 The defendant contended that the “proper care” language did not apply in a theft prosecution when the mistake was that he believed he owned the property, and he argued that the language placed a greater burden on the defense than required by law.29 This Court agreed, holding that a finding of proper care was not required when intent was an element of the offense and the mistake negated that intent.30 Some of the authority quoted by the court indicated that the mistake of fact defense applied only “to acts ‘otherwise criminal,’ or acts in themselves criminal if unexcused” such as the selling of liquor without a license or the commission of a homicide under a mistaken belief that *795would excuse the act.31 “In order that no confusion may arise,” the Court emphasized that the rule that the jury should not be instructed on the “proper care” element of mistake of fact applied “only to those crimes where the unlawful intent is an essential element without which the offense does not arise.”32

Our third observation about the mistake of fact statute — that it applied only when the mistake would completely exonerate the defendant of any offense — is further borne out by the three “mistake” provisions, mentioned earlier, that immediately followed the mistake of fact defense:

Art. 42. Act done by mistake of felony. One intending to commit a felony and who in the act of preparing for or executing the same shall through mistake or accident do another act which, if voluntarily done, would be a felony, shall receive the punishment affixed to the felony actually committed.

Art. 43. Act done by mistake a misdemeanor. One intending to commit a felony and who in the act of preparing for or executing the same shall through mistake or accident do another act which, if voluntarily done would be a misdemean- or, shall receive the highest punishment affixed to such misdemeanor.

Art. 44. Felony done by mistake. One intending to commit a misdemeanor and who in the act of preparing for or executing the same shall through mistake commit a felony shall receive the lowest punishment affixed to the felony.33

These provisions did not mention any culpable mental state lower than “intent.” Aside from that, however, the language of Article 42 was broad, imposing punishment for the “felony actually committed” regardless of whether that felony carried greater punishment than the felony intended. Indeed, at the time, that provision was the basis for the felony-murder doctrine.34 Under Article 42, intent was “transferred” from an intended, but lesser, felony (such as robbery or injury to a child) to murder, an unintended, but greater, felony.35 The remaining two provisions transferred intent between a felony and a misdemeanor, and vice versa, and carried a kind of proportionality limitation. All of this is contrary to the way that appellant wants us to construe the current transferred intent statute.

In 1962, the American Law Institute issued its official draft of the Model Penal Code, which became highly influential throughout the United States, including in Texas. Under the heading of causation, the Model Penal Code set forth its proposed law of transferred intent in § 2.08:

(2) When purposely or knowingly causing a particular result is an element of an offense, the element is not established if the actual result is not within the purpose or the contemplation of the actor unless:

(a) the actual result differs from that designed or contemplated, as the case may be, only in the respect that a different person or property is injured or affected or that the injury or harm designed or contemplated would have *796been more serious or more extensive than that caused; or

(b) the actual result involves the same kind of injury or harm as that designed or contemplated and is not too remote or accidental in its occurrence to have a [just] bearing on the actor’s liability or on the gravity of his offense.

(3) When recklessly or negligently causing a particular result is an element of an offense, the element is only established if the actual result is not within the risk of which the actor is aware or, in the case of negligence, of which he should be aware unless:

(a) the actual result differs from the probable result only in the respect that a different person or property is injured or affected or that the probable injury or harm would have been more serious or more extensive than that caused; or

(b) the actual result involves the same kind of injury or harm as the probable result and is not too remote or accidental in its occurrence to have a [just] bearing on the actor’s liability or on the gravity of his offense.36

The Model Penal Code thus recognized that all culpable mental states were subject to being transferred, and it contained “different person or property” provisions similar to the ones in the current Texas Penal Code. But § 2.03 contained no provision transferring culpability between offenses. It did contain a provision extending liability to “the same kind of injury” so long as that injury was not “too remote or accidental” to have a “just bearing” on the actor’s liability or the gravity of the offense. But the Model Penal Code contained another provision that more explicitly addressed liability for the commission of an unintended offense. That provision was found in the section titled “Ignorance or Mistake,” § 2.04:

(1) Ignorance or mistake as to a matter of fact or law is a defense if:

(a) the ignorance or mistake negatives the purpose, knowledge, belief, recklessness or negligence required to establish a material element of the offense;

(b) the law provides that the state of mind established by such ignorance or mistake constitutes a defense.

[[Image here]]

(2) Although ignorance or mistake would otherwise afford a defense to the offense charged, the defense is not available if the defendant would be guilty of another offense had the situation been as he supposed. In such case, however, the ignorance or mistake of the defendant shall reduce the grade and degree of the offense of which he may be convicted to those of the offense of which he would be guilty had the situation been as he supposed.37

The Model Penal Code’s mistake defense contained no requirement that the mistake be “reasonable.” Indeed, the explanatory note indicated that the mistake “defense” was simply an application of the general principles of culpability inherent in any offense containing a culpable mental state:

The matter is conceived as a function of the culpability otherwise required for commission of the offense. Such ignorance or mistake is a defense to the extent that it negatives a required level *797of culpability or establishes a state of mind that the law provides as a defense. The effect of this section therefore turns upon the culpability level for each element of the offense, established according to its definition and the general principles set forth in Section 2.02.38

The Model Penal Code commentary viewed “mistake of fact” as being a mere evidentiary issue:

In other words, ignorance or mistake has only evidential import; it is significant whenever it is logically relevant, and it may be logically relevant to negate the required mode of culpability or to establish a special defense.39

In his treatise on criminal law defenses, Robinson explains: “§ 2.04(1) does not provide a general mistake defense but simply states the obvious: If a culpable state of mind is required by an offense definition and cannot be proven because of the defendant’s ignorance or mistake, then the defendant cannot be convicted of the offense.” 40

The Model Penal Code § 2.04(2) exception to the defense was based on the idea that “the defendant should not go free, for on either view — the facts as they occurred or as the defendant believed them to be — a criminal offense was committed.” 41 Given the relationship in the Model Penal Code between mistake of fact and the elements of an offense, the § 2.04(2) exception to that “defense” was essentially an affirmative imputation of criminal liability. Robinson called it the “doctrine of substituted mental elements” and observed that the doctrine is broad enough to swallow completely the transferred intent provisions of Model Penal Code § 2.03.42

In 1970, the Texas State Bar Committee on Revision of the Penal Code submitted a proposed draft of what would eventually become our modern Penal Code. In many areas, the influence of the Model Penal Code was evident and included the proposed sections addressing causation and mistake of fact. The causation section contained transferred intent provisions that closely paralleled those articulated in Model Code:

(b) If the offense requires that the actor intentionally or knowingly cause a result, he is criminally responsible for the result if the result that actually occurred:

(1) was desired or contemplated, whether the desire or contemplation extended to natural events or the conduct of another; or

(2) was desired or contemplated and occurred in a manner not too accidental, or by a means not too dependent on another’s volitional act, to have a just bearing on the actor’s criminal responsibility or the gravity of his offense.

(c) If the offense requires that the actor recklessly or with criminal negligence cause a result, he is criminally responsible for the result if the result that actually occurred:

(1) was within the risk perceived or that which should • have been perceived, whether the risk extended to natural events or the conduct of another; or

*798(2) was within the risk perceived or that which should have been perceived and occurred in a manner not too accidental, or by a means not too dependent on another’s volitional act, to have a just bearing on the actor’s criminal responsibility or the gravity of his offense.

(d) An actor is nevertheless criminally responsible for causing a result if the only difference between what actually occurred and what he desired, contemplated, or risked is that a different person or property was injured, harmed, or otherwise affected.43

The drafters commented that their proposed penal code “for the most part emphatically rejected” the doctrine of “constructive or implied malice” that was then codified in Penal Code articles 42 through 44.44 The drafters believed that the only features of the doctrine that were retained were the felony-murder rule and the portion of the transferred intent provision found in § 6.07(d).45

But as we now know, most of § 6.07 never became part of the Penal Code. Only subsection (d) survived to become part of the current code. Despite the changes from the proposed code to the code that was enacted, the practice commentary to the current provision suggested that § 6.04(b)(1) was unnecessary and “was apparently added out of an abundance of caution.”46 Although the practice commentary is often helpful because it was drafted by two individuals who were part of the Penal Code revision project, we do not find the commentary to be instructive in this instance because the language of the current statute does not reflect the revision committee’s recommendations. The Legislature rejected the proposed provisions that made liability contingent upon whether an accidental occurrence had “a just bearing on the actor’s criminal responsibility or the gravity of his offense.” Instead, the Legislature created § 6.04(b)(1), which extended liability where the “only difference” with respect to the actor’s culpability was that a “different offense was committed.” This language is reminiscent of, and in fact broader than, the language found in articles 42 through 44. and is also similar to language found in MPC § 2.04(2), the “substituted mental elements” portion of the “mistake” section of the Model Penal Code.

The mistake of fact section of the 1970 proposed code also reflected the influence of the Model Penal Code, but with one important difference. Like the Model Code, the proposed code did not require that the actor’s mistaken belief be reasonable, and the proposed code limited the mistake of fact defense to situations that negated the culpable mental state required for the offense:

It is a defense to prosecution that the actor was honestly ignorant or mistaken about a matter of fact or law if his ignorance or mistake negated the intent, knowledge, recklessness, or criminal negligence required to establish an element of the offense charged47

But the proposed code’s exception to the defense was narrower than that prescribed by the Model Penal Code. While the Model *799Code made the defense unavailable whenever the actor’s mistaken belief would have made him liable for any different crime (although reducing the punishment range, if it were less), the proposed code recognized only that the defendant could be “convicted of a lesser included, offense of which he would be guilty” if the facts were as he believed.48 The commentary to the proposed Texas provision explained that, if the defendant’s mistaken belief would result in committing “an offense other than one included in the offense charged, he must be charged and tried again.”49

The drafters of the proposed penal code viewed the mistake of fact defense as essentially redundant of the requirement that the State prove the mental element of an offense,50 but they included the defense as a method of placing upon the defendant “the burden of producing evidence” so that a mistake of fact is something “the prosecution does not have to negate unless raised.”51 That view was maintained in the practice commentary to the current provision.52 Again, we do not find the practice commentary to be instructive because the enacted statute differs significantly from the proposed version. Although the proposed version simply required that the defendant’s mistake be an “honest” one, the enacted version requires that the mistaken belief be “reasonable.” This requirement reflects language found in the prior version of the statute. As with the transferred intent provision, then, the mistake of fact defense appears to incorporate elements from both the Model Penal Code and the prior version of the Texas Penal Code.

The history of these two provisions reveals that the law of transferred intent with respect to offenses has been entwined with the law of mistake. Given that history, it seems probable that the Legislature intended in its enactment of the current Penal Code that these two aspects of the law go hand-in-hand. A codified, specialized example of this interaction may be found in the current Penal Code in the offense of murder. The murder statute describes three methods of committing the offense: (1) intentionally or knowingly causing the death of an individual, (2) intending to cause serious bodily and committing an act clearly dangerous to human life that causes the death of an individual, and (3) felony murder.53 The first method is the traditional form of murder, while the third is the new (in 1974) offense of felony murder. But the second method is a straightforward application of the doctrine of the transferred intent of offenses, along with conditions that negate a mistake of fact defense. A person who “intends to commit serious bodily injury” intends thereby to commit the offense of aggravated assault.54 By actually causing death, the person commits the unintended offense of murder. And committing “an act clearly dangerous to human life” negates the reasonableness requirement found in the mistake of fact defense. By contrast, the offense of ca/pital murder contains what appears to be a sort of anti-transfer element: the offense specifically requires that *800a murder be committed “as defined under Section 19.02(b)(1),” the provision proscribing intentional and knowing murders.55

We also observe that applying the mistake of fact defense to the transfer of intent between offenses mitigates greatly the concern that a person could be penalized far beyond his actual culpability. At the same time, it has the salutary effect of placing some onus on the person who intends to commit a lesser offense to exercise care that a greater one is not in fact committed. Our interpretation seems to be the only one that gives meaning and effect both to § 6.04(b)(1) and to § 8.02.

4. Conclusion

Given the plain language and the history of the provisions at issue, we conclude that § 6.04(b)(1) does indeed authorize the transfer of a culpable mental state between offenses contained in the same statute and also between greater and lesser included offenses. That authorization may be overridden by language defining a particular offense, as in the offense of capital murder, but no such impediment arises with respect to the injury-to-a-child offense. Where § 6.04(b)(1) permits the transfer of a culpable mental state, mistake of fact may be raised as a defense. The mistake must be reasonable for it to constitute a circumstance that exculpates the defendant of the offense charged, and of course, the defendant would still be guilty of any lesser included offense that would be applicable if the facts were as the defendant believed. Given our holding, we overrule Honea to the limited extent that it held that a defendant is not entitled to a mistake-of-fact instruction.

C. Jury Charge

The trial court correctly submitted the law of transferred intent in the jury charge. The charge contained no instruction regarding the mistake of fact defense, but appellant failed to request its submission, and as a result, no error occurred with respect to the absence of that defensive instruction.56 Although our holding regarding the mistake of fact defense constitutes a new proposition of law, that does not relieve appellant of the obligation to preserve such a claim. Even new rules that are held to be retroactive can be forfeited by a party’s failure to complain at trial.57

We affirm the judgment of the court of appeals.