OPINION

delivered the opinion of the Court,

The State charged Thomas Kerr Gray, III, with driving while intoxicated due to alcohol consumption. Evidence at trial showed that Gray was taking anti-depressant medication at the time. The jury instructions authorized finding that Gray was intoxicated if he took drugs which made him more susceptible to the influence of alcohol than he otherwise would have been, and therefore became intoxicated from the alcohol. We conclude that this jury charge properly applied the law to the facts of this case.

I. FACTS

The day after Tropical Storm Allison flooded Houston in 2001, Michael White-sides was driving his brand new pick-up truck on a wet, muddy road. The car behind him, he later learned, was driven by 17-year-old Gray. He noticed Gray because Gray “seemed ... unaware” and “was one foot off [Whitesides’s] bumper.” When Whitesides’s tires inevitably sprayed mud on Gray’s windshield, Gray “erratically pulled to the median,” then “cut back into traffic, across the inside lane, cutting people off. [He] got to the outside lane, and approached [Whitesides] from the rear at a high rate of speed.” Gray then rear-ended Whitesides’s truck. When White-sides approached Gray to confront him, Gray’s radio was playing “very loud” and Gray was “not acting normal, at all.”

The police arrived and, noticing Gray’s constricted pupils and the smell of alcohol on his breath, conducted field sobriety tests. Gray failed the tests and was arrested for driving while intoxicated. Gray told the police that he was taking antidepressant medication.

II. PROCEDURAL HISTORY

The State charged Gray by information with misdemeanor DWI, alleging that he “did then and there unlawfully while intoxicated, namely not having the normal use of his mental and physical faculties by the reason of the introduction of ALCOHOL into his body, operate a motor vehicle in a public place.”

Gray’s father testified at trial that Gray had been undergoing treatment for psychological problems for the last three years. He had been prescribed four medications — Respiratol, Zoloft, Klonopin, and Depical — during the general time period of the accident, but on that day he was taking two or three of them. His father indicated that the medications caused “disorientation” and “dizziness.” Gray’s father also testified that Gray was not drinking any alcohol that day — -“he couldn’t have.”

Gray’s stepmother testified that Gray was on three medications at the time of the accident. She also said that he had just started taking Depical three days before the incident and that Depical made him very disoriented.

The State’s chemist testified that the drugs were anti-depressants that have a depressant effect on the central nervous system. Alcohol had the same effect. “And, when you add one depressant medication to another depressant medication, it can have, what’s called, an ‘addictive [sic ] effect.’ Sometimes, what’s described as a ‘synergistic effect.’ Where the two medi*127cations interact in an inappropriate way and accelerate the action of one medication.” She said that effect could manifest itself even if the person had only a small amount of alcohol.

The court’s proposed jury charge stated as follows:

You are further instructed that if a Defendant indulges in the use of a drug, to wit, Respiratol, Zoloft, Klonopin, and/or Depical, to such an extent that he thereby makes himself more susceptible to the influence of alcohol than he otherwise would have been, and by reason thereof becomes intoxicated from recent use of alcohol, he would be in the same position as though his intoxication was produced by the use of alcohol alone.

The application paragraph of the charge instructed the jury to convict if it found that Gray was driving while intoxicated “by reason of the introduction of alcohol into his body, either alone or in combination with Respiratol, Zoloft, Klonopin and/or Depical.”

Gray objected to the charge on the grounds that it did not follow the State’s information. The prosecutor responded that the charge was appropriate under Sutton v. State.2 The judge overruled Gray’s objection.

The jury found Gray guilty of misdemeanor DWI, and the judge assessed punishment at 180 days in the Harris County Jail, suspended for one year.

III. ON APPEAL

On appeal, Gray argued that the trial judge erred in charging the jury on the “synergistic effect” of drugs and alcohol because (1) it was not part of the law applicable to the case, (2) it was a comment on the weight of the evidence, (3) it was misleading and confusing, and (4) Gray was not prosecuted under a combination theory. The Court of Appeals disagreed.3 It concluded, in relevant part, that the charge incorporated the law applicable to the case, relying on our plurality opinion in Sutton.4

Gray petitioned this Court for discretionary review on two grounds. We granted the second ground, which contends that the Court of Appeals erred in holding that the instruction constituted the law applicable to the case when Gray was not charged with intoxication by a combination of alcohol and drugs.

IV. ANALYSIS

Our Legislature has made clear that a trial judge’s charge to the jury must set forth “the law applicable to the case.”5 Relying on that statute, we have held that “[a] trial court is required to fully instruct the jury on the law applicable to the case and to apply that law to the facts presented.” 6 It is not enough for the charge to merely incorporate the allegation in the charging instrument. Instead, it must also apply the law to the facts adduced at trial. This is because “[t]he jury must be instructed ‘under what circumstances they *128should convict, or under what circumstances they should acquit’.”7 Jury charges which fail to apply the law to the facts adduced at trial are erroneous.8

This case involves a specific kind of jury charge known as the “synergistic effect” charge. We begin with the background of this charge.

A. The Synergistic Effect Charge

Originally, our DWI statute did not define “intoxication.” In Heard v. State,9 we addressed a synergistic effect charge under that statute. (Although Heard was decided after the statute had been changed, the defendant’s offense had occurred under the previous version of the statute.) There, the information alleged driving under the influence of liquor. During trial, the defense presented evidence that the defendant had been taking several medications at the time of the offense. The evidence showed that the medications combined with alcohol “might cause an individual to lose his or her mental alertness quicker than if he or she was drinking only alcohol.”10

The jury charge instructed the jury that “if a Defendant indulges in the use of drugs to such an extent that he thereby makes himself more susceptible to the influence of intoxicating liquor than he otherwise would have been and by reason thereof becomes intoxicated from the recent use of intoxicating liquor, he would be in the same position as though his intoxication was produced by the use of intoxicating liquor alone.”11 The application paragraph of the charge then told the jury it could convict if it found that the defendant was “under the influence of intoxicating liquor, either alone or in combination with drugs.”12

We found this charge to be proper because it applied the law to the facts of the case. We explained that the “combination of liquor and drugs which would make an individual more susceptible to the influence of the liquor is in effect equivalent to intoxication by liquor alone.”13

The 68th Legislature amended the DWI statute, effective January 1, 1984.14 After that amendment, the statute defined “intoxicated” as, among other things, “not having the normal use of mental or physical faculties by reason of the introduction of alcohol, a controlled substance, a drug, or a combination of two or more of those substances into the body.”15

We addressed the effect of this statutory change on DWI charging instruments in Garcia v. State.16 There, the information charged the defendant with driving while intoxicated. It did not specify whether the *129intoxication was a result of alcohol, drugs, or a combination of both. As a result, the defendant moved to quash. The trial court denied the motion, but the Court of Appeals reversed, holding that the motion should have been granted.17

We agreed with the Court of Appeals. We concluded that, to provide sufficient notice to a defendant, a charging instrument alleging intoxication under the new DWI statute had to allege the intoxicant.18 Having reached that conclusion, we then added that, since the statute defined different methods of becoming intoxicated, “the type of intoxicant used ... becomes an element of the offense and critically necessary to the State’s proof.”19 In the following and final paragraph of the opinion, we returned to the notice rationale, concluding that the defendant in Garcia had been denied notice and that the trial court erred in denying his motion to quash.20

Our holding in Garcia seemed to call Heard into question. In Heard, we had concluded that a synergistic effect jury charge was permissible even if a combination theory had not been alleged in the charging instrument. But at the time of Heard, the DWI statute did not define intoxication, and the State was not required to allege the intoxicant in the charging instrument to provide the defendant notice. Moreover, at the time of Heard, the intoxicant was not an element of the offense. Now, under Garcia, the State was required to allege the intoxicant in its charging instrument both for notice and because the intoxicant was an element of the offense. And the new statute specifically put forth the option of alleging “a combination” theory of intoxication. Under this new statute and these new pleading requirements, could a jury still be charged on the synergistic effect of more than one intoxicant if the State had not alleged a “combination” theory in its indictment?

We were faced with that question in Sutton v. State,21 and the Court was unable to reach a majority. In that case, the information charged the defendant with driving while intoxicated due to alcohol. At trial, the defendant testified that he had taken two pills of a prescription medication which, when mixed with two beers, put him in a “precarious situation.”22 The defense expert testified that the prescription medication could cause a person to appear intoxicated and that alcohol would enhance the effect of the drug.23 The judge instructed the jury on the synergistic effect of the prescription medication and alcohol, permitting the jury to convict if it found that “the defendant was intoxicated by reason of the introduction of alcohol into his body, either alone or in combination with Klonopin.”24

A four-judge plurality concluded that these facts were identical to Heard and that our opinion in Heard was dispositive.25 The plurality stated that the jury charge did not expand on the allegations in the information; rather, it'“allowed conviction only if the jury found that [the defendant] had been intoxicated with alcohol, either *130alone or in combination with a drug that made him more susceptible to the alcohol. In either case, the jury had to find that [the defendant] had been intoxicated with alcohol, not with the drug.”26 The plurality also noted that “[o]nce [the defendant] introduced this evidence, the trial court was obligated to instruct the jury on the law as explicated in Heard.”27 The plurality did not discuss the fact that the DWI statute had changed since Heard. And it concluded that Garcia was irrelevant because Garcia had involved a motion to quash, and Sutton had not filed such a motion.28

Four judges dissented. In his dissenting opinion, Judge Maloney argued that the plurality failed to “account for changes in the controlling statute”29 and that the plurality’s “reasoning might have made sense at the time of the Court’s opinion in Heard since at that time there was no separate statutory category for a combination of drugs and alcohol, [but] it is not consistent with the current version of the statute.”30

The law was still in this uneasy state when we decided Rodriguez v. State31 several years ago. There, the indictment alleged that the defendant drove while intoxicated due to alcohol. At trial, defense evidence showed that the defendant had been taking Contac, an over-the-counter medication, which made him drowsy. The State’s expert then testified to the effects of Contac as well as the effects of mixing Contac with alcohol. The charge defined “intoxicated” as not having the normal use of mental or physical faculties “by reason of the introduction of alcohol, a drug, or a combination of both of those substances, into the body.”32 The application paragraph of the charge then permitted the jury to convict if it found that the defendant was intoxicated “by the reason of the introduction of a combination of unknown drugs and alcohol into his body.”33

We recognized that this jury charge was a far cry from the ones in Heard and Sutton. In those cases, the charge allowed conviction only if the jury found that the defendant had been intoxicated with alcohol, either alone or in combination with a drug that made him more susceptible to the alcohol. But in Rodriguez, the charge allowed the jury to convict if it found the defendant was intoxicated by a combination of unknown drugs and alcohol. There was “no ‘additional susceptibility’ theory put to the jury.”34 It therefore “improperly expanded on the allegations set forth in the information, and authorized a conviction on a theory not alleged in the charging instrument.”35

We noted in Rodriguez that, “despite its reliance on Heard, the [Sutton] plurality placed no significance on the fact that the type of intoxicant was not an element of the offense under the statute in Heard but was an element of the offense under the statute.”36 We resolved Rodriguez without having to address that issue because “regardless of one’s view of the relevance *131of Garcia to this issue, the jury charge in this case falls short of the requirement established in Sutton.”37

B. Application

In this case, the State alleged that Gray was under the influence of alcohol while driving, and Gray presented evidence that he was under the influence of anti-depressants at the time of the offense. The trial judge’s charge to the jury incorporated both the State’s allegation and the evidence presented at trial by instructing the jury that it should convict if it found that Gray had indulged in the drugs “to such an extent that he thereby makes himself more susceptible to the influence of alcohol than he otherwise would have been, and by reason thereof becomes intoxicated from recent use of alcohol.” Under our caselaw, this charge instructed the jury on the law applicable to the case and applied that law to the facts presented.38 So this was a standard “synergistic effect,” and therefore valid, jury charge.

Gray disagrees, arguing that this charge expanded on the scope of the State’s information. He says that according to Garcia, the intoxicant is an element of the offense of DWI. Since the DWI statute under which Gray was charged lists six possible (though overlapping) intoxicants — “alcohol, a controlled substance, a drug, a dangerous drug, or a combination of two or more of those substances, or any other substance”39 — Gray argues that the State must allege which intoxicant and prove that same intoxicant. By allowing the State to allege “alcohol” but prove “combination,” Gray says, the State is permitted to expand upon the theory alleged in the charging instrument, violating due process.

The State responds that this charge was virtually identical to the one given in Sutton. It also contends that Garcia is poorly reasoned and should be overruled. Gray replies that it is actually Sutton that is poorly reasoned.

1. Intoxicant as Element of Offense

We begin with Garcia’s statement that the intoxicant is an element of the offense of DWI.

Due process provides that no person may be convicted of a criminal offense unless every element of the offense is proved beyond a reasonable doubt.40 Our Penal Code defines the elements of the offense to be the forbidden conduct, the required culpability, any required result, and the negation of any exception to the offense.41 As a practical matter, we define the elements of individual offenses on a case-by-case basis. Under the previous statute, V.A.T.S. Art. 6701 i-1, we held that the elements of misdemeanor DWI were that: (1) a person (2) drives or operates (3) a motor vehicle (4) in a public place (5) while intoxicated.42 The current statute, § 49.04(a), says essentially the same thing as the previous one, and the elements of the offense are: (a) a person (b) is intoxicated (c) while operating (d) a motor vehicle (e) in a public place.

*132In Garcia, we said that an additional element was the type of intoxicant. We made this statement even though- the disputed issue in the case concerned notice. As a result, the statement was dicta, unnecessary to the resolution of the case. Moreover, we did not cite any authority for this conclusion, nor did we explain it. There are several problems with the statement.

First, it contradicts the statutory definitions of what constitutes an element of the offense. The intoxicant is not the forbidden conduct, the required culpability, any required result, or the negation of any exception to the offense.43 The mere fact that the DWI statute separately defines intoxication does not automatically elevate the intoxicant to the status of an element of the offense.

Second, Garcia’s dicta conflicts with our caselaw regarding elements of other offenses. In Ex parte Luna,44 we addressed a similar issue with regard to the theft statute. There we concluded that the definitions of “unlawfully” in § 31.03 were not essential elements of theft, but were evi-dentiary matters that need not be alleged in order to sufficiently charge the offense of theft.45 Like the theft statute, the DWI statute defines “intoxicated,” but that does not mean that the various intoxicants are elements of the offense. Instead, they are descriptive terms.

Third, making the intoxicant an element of the offense constitutes bad public policy. As the State points out, when we concluded that the intoxicant was an element of the offense, the intent of the statute was thwarted, and trials degenerated into ones like this one, where the defense is essentially “Yes, I was intoxicated, but you alleged the wrong intoxicant, therefore I am entitled to an acquittal.” We agree with the State that permitting defendants to “secure an acquittal merely by proving an alternative intoxicant” constitutes “an abuse of the system ... contrary to the plain meaning of the DWI statute, which focuses on the acts of the defendant while intoxicated rather than the act of becoming intoxicated itself.”

We conclude that the substance that causes intoxication is not an element of the offense. It is not the forbidden conduct, the required culpability, any required result, or the negation of any exception to the offense. Instead, it is an evidentiary matter. We disavow the dicta in Garcia to the contrary.

2. Sutton v. State

Doing away with Garcia’s element-of-the-offense language disposes of Gray’s claim that due process was violated due to the State alleging one offense and convicting on another. But it does not refute Gray’s argument entirely. We still must address whether this jury charge expanded on the allegations in the State’s charging instrument. For this, we return to Sutton.

In Sutton, the information alleged alcohol as the intoxicant, and the evidence at trial showed drugs and alcohol. The judge instructed the jury on the synergistic effect of prescription medication and alcohol. The plurality said that the jury charge did not expand on the allegations in the information; rather, it “allowed conviction only if the jury found that [the defendant] had been intoxicated with alcohol, either alone or in combination with a drug that made him more susceptible to the alcohol. In *133either case, the jury had to find that [the defendant] had been intoxicated with alcohol, not with the drug.”46 The charge in this case, being nearly identical to the one in Sutton, should be valid, unless there is some reason that the Sutton plurality erred.

Gray argues that Sutton is “flat-out wrong” because the plurality failed “to come to grips with the fact” that the statute changed after Heard was decided. It is true that the DWI statute was amended after Heard. But we believe the Sutton plurality was still correct.

Gray argues that if the statute provides for six different intoxicants, then the jury charge must include only the intoxicant alleged in the State’s charging instrument, otherwise the jury charge expands on the scope of the State’s allegations. But there are two flaws in this position. First, the jury charge in this case did not permit the jury to convict on a completely separate intoxicant. Instead, it permitted conviction if the consumption of one intoxicant made Gray more susceptible, and therefore intoxicated, by the charged intoxicant. So even if Gray’s premise were correct, this particular jury charge would be valid because it still required the jury to find that Gray was intoxicated due to alcohol, which is what the State alleged.

Second, the six possible definitions of intoxication in the statute are not mutually exclusive — indeed, they are patently overlapping. The definitions of intoxication in the DWI statute are in this way unique from other criminal statutes. An allegation of “controlled substance” overlaps with “drug” and “dangerous drug.” An allegation of “combination” overlaps with “alcohol” and “drug.” As a result, these are not really six different intoxicants.

Here, the State alleged intoxication by alcohol, and the jury charge permitted finding intoxication by alcohol if the defendant was more susceptible to the alcohol due to drugs. This charge did not permit conviction on a theory unalleged in the charging instrument. This is still the same theory of committing DWI — the “loss of faculties” theory.47

Gray also argues that the Court of Appeals’ opinion, and therefore, Sutton, conflicts with Rodriguez. We disagree. As explained above, the jury charge in Rodriguez was quite different from the one in Sutton. The charge in Rodriguez did not put forth an “additional susceptibility theory” to the jury. It therefore “improperly expanded on the allegations set forth in the information, and authorized a conviction on a theory not alleged in the charging instrument.”48 The same is not true here or in Sutton. In both this case and in Sutton, the charge permitted conviction only if the drugs made the defendant more susceptible to the alcohol. Both this charge and the one in Sutton still required intoxication due to alcohol.

We conclude that the Sutton plurality was correct to uphold the synergistic effect jury charge, even though the DWI statute changed after Heard. The charge in this case was essentially identical to the one in Sutton and different from the one in Rodriguez. It permitted the jury to convict if Gray’s drug use made him more susceptible to alcohol, but it still required intoxication due to alcohol. Therefore, it did not expand on the allegations in the charging instrument, and it properly applied the law to the facts of the case.

Judge Cochran contends in dissent that we should dispense with synergistic jury *134charges entirely, but we do not believe her arguments rise to our high standard for overruling precedent.49 Additionally, the crux of her argument is really that this charge was a comment on the weight of the evidence. But that issue was presented in Gray’s first ground for review, which we refused. We do not address that claim today. Judge Meyers argues that this charge was erroneous because it permitted the jury to convict Gray on an allegation not within the charging instrument. But as discussed above, this charge still required the jury to find that Gray was intoxicated by alcohol.

Finally, the State argues that Garcia’s notice holding should be overruled. We have already concluded that the charge in this case was valid. Reaching out to address the validity of Garcia’s notice holding would be dicta. And we have rejected this argument before.50 We decline to consider it today.

V. JUDGMENT

We affirm the judgment of the Court of Appeals.

dissenting.

I join Judge Cochran’s dissenting opinion and add these comments regarding the jury charge in this case. A good method for determining whether a charge is correct is to consider the factors enumerated in Malik v. State, 953 S.W.2d 234 (Tex.Crim.App. 1997), in describing a hypothetically correct jury charge. Under the sufficiency analysis in Malik, we stated that the “sufficiency of the evidence should be measured by the elements of the offense as defined by the hypothetically correct jury charge for the case. Such a charge would be one that accurately sets out the law, is authorized by the indictment, does not unnecessarily increase the State’s burden of proof or unnecessarily restrict the State’s theories of liability, and adequately describes the particular offense for which the defendant was tried.” Id. at 240.

The information in this case alleged only intoxication by alcohol. However, the application paragraph instructed the jury to convict the defendant if it found “intoxication by reason of the introduction of alcohol into his body, either alone or in combination with Respiratol, Zoloft, Klono-pin and/or Depical.” This application paragraph was not authorized by the charging instrument. The best that I can discern from the majority opinion is that the trial court can now expand the application paragraph to include additional theories of liability and incorporate evidence on how the defendant became intoxicated even though that intoxicant was not pled in the information. I am afraid that I have not come across this authority before. Because this is error in the charge and Appellant timely objected, I would hold that the error was not harmless and reversal is required. See Almanza v. State, 686 S.W.2d 157 (Tex.Crim.App. 1985).

For these reasons, I respectfully dissent.

joined by MEYERS, J., dissenting.

I respectfully dissent. I would overrule *135both Garcia v. State1 and Sutton v. State,2 reverse the court of appeals’ judgment, and remand this case to that court for a harm analysis because the trial court’s application paragraph was technically improper and the “synergistic effect” jury instruction is not part of the law applicable in a DWI case.

I.

The evidence at trial showed that the 17-year-old appellant was arrested for DWI after he rear-ended another driver’s truck with his blue Lexus. When a police officer arrived at the accident scene, he noticed that appellant smelled of alcohol, had constricted pupils, and did not appear to have the normal use of his mental or physical faculties. The officer administered an HGN test which indicated intoxication, a portable breath test which indicated that appellant had alcohol in his system, and field sobriety tests which appellant could not perform successfully. Appellant told the officer that he was taking anti-depressant medications, including Respiratol, Zoloft, Klonopin, and Depical. According to appellant’s father, these medications caused “disorientation” and “dizziness.” Appellant’s stepmother said that one of the drugs, Depical, caused appellant to become very disoriented. In rebuttal, the State called a chemist who testified that these anti-depressant medications could accelerate the action of alcohol, causing mental and physical impairment in a driver with only a small amount of alcohol.

As a part of the jury charge in this case, the trial judge included the following general instruction:

You are further instructed that if a Defendant indulges in the use of a drug, to wit, Respiratol, Zoloft, Klonopin and/or Depical, to such an extent that he thereby makes himself more susceptible to the influence of alcohol than he otherwise would have been, and by reason thereof becomes intoxicated from recent use of alcohol, he would be in the same position as though his intoxication was produced by the use of alcohol alone.

Although appellant was charged only with intoxication by means of alcohol, the trial judge also included, in the application paragraph of the charge, an instruction that the jury could convict appellant of DWI if it found that he was intoxicated “by reason of the introduction of alcohol into his body, either alone or in combination with Respiratol, Zoloft, Klonopin and/or Depical.”

During closing argument, the defense attorney agreed that appellant was “obviously” impaired at the time he rear-ended the other driver’s truck. His defense was that this mental and physical impairment was not caused by the alcohol appellant had consumed, but rather by the antidepressant drugs he was taking. The jury, by its verdict, rejected this defense and convicted appellant of DWI.

The issue before us is whether the trial court’s “synergistic effect” jury instruction and its application paragraph, permitting conviction upon proof of intoxication by means of a combination of alcohol and drugs, were proper when appellant was charged only with intoxication by means of alcohol.

II.

Appellant argues that the trial court erred in charging the jury on intoxication by means of a combination of drugs as well as by means of alcohol alone. According to appellant, the application paragraph broadened the allegations in the informa*136tion and permitted conviction for DWI on a basis which was not specifically pleaded.

I think that appellant is technically correct. However, I would take this opportunity to overrule Garcia which held that, in a DWI prosecution, “the type of intoxicant used, i.e., alcohol, a controlled substance, a drug, or a combination of two or more of these substances, becomes an element of the offense and critically necessary to the State’s proof.”3 Garcia is no longer good law because the DWI statute has been amended so that the type of intoxicant that a person consumed to become physically or mentally impaired is of no legal significance.

When Garcia was decided in 1988, there were only three possible “intoxicants” that were subject to prosecution under the DWI statute: alcohol, a controlled substance, or a drug. A driver who was impaired because of the ingestion of any one of these three substances, either alone or in combination, could be convicted of DWI. In 1993, the Legislature amended the definition of “intoxicated” to include “alcohol, a controlled substance, a drug, a dangerous drug, a combination of two or more of those substances, or any other substance.,”4 Any other substance at all. Under current law, if a person eats too many M & Ms either alone or in combination with alcohol, drugs, water, or whatever such that his mental and physical faculties are impaired, he may be prosecuted for DWI. However, “eating M & Ms” is clearly not an element of the offense of DWI. “Intoxicated” is an element of DWI, and it is a physical state of being, regardless of the specific substance which caused the impairment.

The purpose of DWI laws, both here in Texas and across the nation, is to prevent and punish the “carnage caused by drunk drivers.”5 These laws are not focused upon the type of substance that caused the driver to become “drunk” and then get behind the wheel of a car. The law does not differentiate between the drunk driver who was intoxicated because he consumed alcohol, or injected heroin, or sniffed glue, or took prescription medicine. It is the act of driving while one’s mental or physical faculties are impaired that is inherently dangerous and the target of our DWI law.

This case is a prime example of the Dickensian hair-splitting that we have allowed ourselves to fall into. Appellant makes no claim that he was not intoxicated at the time he rear-ended a fellow driver. He admits that he was a physically and mentally impaired driver. His contention is simply that he was not intoxicated because of the alcohol he drank; instead, he was intoxicated because of the anti-depressants he was taking. This is a “defense” that is condoned, if not encouraged, by our decision in Garcia which requires the State to allege the precise substance that it thinks caused an impaired driver’s intoxication, and then permits that driver to defend against the charge by claiming that he was intoxicated on some other substance.

Nonetheless, the State, in this case did allege only alcohol as the means of intoxication in its information. I certainly cannot fault the State for doing so, as it was *137conscientiously following the dictates set down , by this Court in Garcia. But our law is also clear that if the State alleges a specific intoxicant, such as alcohol, it must prove what it pleads because that substance is descriptive of an element.6 Generally, when an element or term is defined by statute, such as the element “intoxicated,” it need not be further alleged in the indictment.7 The specific name of the substance that caused a person’s intoxicated state is not an element of the offense of DWI, nor is there now a discrete statutory list of specific intoxicating substances that the State must rely upon in obtaining a DWI conviction. Thus, the name of the substance is an evidentiary matter, not an element of the offense or a specifically defined statutory manner and means of committing the offense of DWI.8

The defendant is, of course, entitled to due process notice of the charge against him, but as we recently held in Kellar v. State>9 such notice need not always come from the face of the indictment or information. It may come as well from formal or informal discovery procedures. Furthermore, in ' many instances, the defendant may be the only person who has knowledge of whether he consumed substances in addition to those alleged by the State. In any event, it should not be a defense to DWI that the defendant was intoxicated on one substance instead of another substance.10

Although I reluctantly conclude that, based upon our prior law, the trial court technically erred in expanding the application paragraph to include the potential for finding appellant guilty of DWI based upon his intoxication due to substances other than alcohol, I would take the opportunity to overrule Garcia and hold that henceforth the State need allege, in its charging instrument, only that the defendant drove or operated a motor vehicle in a public place while intoxicated. It need not allege any specific substance that caused the defendant to become intoxicated, but if it does do so, it must prove that the specifically alleged substance was the cause of the intoxication.

III.

I also agree with appellant that the “synergistic effect” instruction given in *138this case is not part of the “law” applicable to a criminal case and should not be given as a jury instruction. The attorneys may argue this very statement, in these very words, as a common sense proposition if it is based upon the evidence presented. However, the judge may not set out in his jury instructions statements that are subject to proof by evidence, but which do not form part of the statutory law of this state. Once again, the trial judge, as well as the State, acted in complete good faith in following this Court’s prior decisions which had approved this very instruction. I cannot fault him at all, and I cannot fault the majority for following precedent, but I think that we should take the opportunity to correct our own past mistakes.

We recently held that “Texas courts are forbidden from instructing the jury on any presumption or evidentiary sufficiency rule that does not have a statutory basis.”11 And we based that holding on our discussion in Browning v. State:

In any given case the jury could make any number of reasonable inferences. But when the trial court, the only source of law the jury has, picks out only one such inference and instructs the jury that that one, though rebuttable, is a presumption provided by law, the court gives the force of law to that one possible inference. In fact neither statute nor caselaw provides such a “presumption” at the trial level. Instructing the jury that it does constitutes, in effect, a comment on the weight of the evidence. We do not hold that the jury may not make such an inference, nor that an appellate court in reviewing the sufficiency of the evidence may not assume that the jury made such a reasonable inference. The error lies in instructing the jury that they may apply such an inference.12

The “synergistic effect” instruction in this case is just like the instructions this Court condemned in Brown and Browning. It is-not part of the law applicable to the case, and it is a comment on the weight of the evidence. I think that it is error.

Let us look at the logic involved. Do trial judges always give this jury instruction as part of the law applicable to a DWI case? No. This instruction has nothing to do with the statutes relating to intoxication or driving while intoxicated. It has to do with the specific evidence admitted at trial.

Suppose the defendant testified that he ate three M & Ms with his two glasses of beer. Would the trial judge then be required or permitted to give this same instruction but substitute the phrase “M & Ms” for the phrase “Respiratol, Zoloft, Klonopin, and/or Depical”? No. Would the attorneys be able to argue “synergistic effect”? No. There is no evidence that M & Ms cause a “synergistic effect” when combined with alcohol. It takes scientific evidence to support a possible conclusion that X substance, when combined with Y substance, causes a synergistic effect upon the body, making the effect of Y substance more pronounced than it otherwise would have been.

*139Suppose an expert witness testifies, as the State’s chemist did in this case, that Respiratol (or M & Ms or whatever), when combined with alcohol, creates a synergistic effect. Now there is evidence in the record that supports a possible inference that the defendant experienced greater effects from the alcohol that he consumed than he would have otherwise. Does that mean when any expert testifies to a scientific “fact” or “opinion” that the trial judge is required or permitted to instruct the jury that the expert’s testimony is the law applicable to the case? No. If the Defendant should call an expert who testifies that Respiratol has a reverse synergistic effect — i.e., a person who takes Respiratol is less susceptible to the effects of alcohol than one who is not taking Respiratol— should the trial judge instruct the jury in the converse? No. What is the logic of singling out one portion of one witness’s testimony and instructing the jury that a specific portion of that witness’s testimony is “the law applicable to the case”? I see no logic in this position, but there is considerable mischief lurking here.

Why, then, do we condone it? It’s been done before. In Sutton v. State, a plurality of this Court said the trial court did not err in giving this “synergistic effect” instruction because ... it had been done before in Heard v. State.13 And in Heard v. State, this Court held that the instruction was proper because .... it had been done before in Kessler v. State.14, But, in fact, it had not been done before in Kes-sler. Instead, the DWI defendant in that case had asked for the reverse instruction: “that if she was intoxicated on the night in question from the combined use of amytal and whiskey to acquit her.”15 This Court appropriately rejected her claim for such an instruction, reasoning that

If she indulged in the use of amytal to such an extent that she thereby made herself more susceptible to the influence of intoxicating liquor than she otherwise would have been and by reason thereof became intoxicated from the recent use of ardent spirits, she would be in the same position as though her intoxication was produced by the use of whisky alone. A person who gets himself in a condition whereby he may become intoxicated from a lesser quantity of whisky than it would ordinarily take to produce intoxication is nevertheless intoxicated from the use of whisky.

We note, however, that the court in his main charge instructed the jury that if they believed from the evidence that appellant was drunk from the use of amytal, or if they had a reasonable doubt thereof, to acquit her. This was all she was entitled to.16

Somehow, in the forty-five years between Kessler and Heard, this Court’s discussion and logical reasoning in the former case became transmogrified into a legal jury instruction in the latter case. But we should not continue to accept, in perpetuity, the illogical proposition that a snippet of an expert witness’s testimony — or the common-sense conclusion that can be drawn from that testimony — becomes the “law applicable to the case” simply because one case went wrong a long time ago.

*140We recently explained in Brown that under our Texas adversarial system, “the judge is a neutral arbiter between the advocates; he is the instructor in the law to the jury.”17 But he cannot comment on the weight of the evidence, or instruct the jury on conclusions that they may draw from the evidence, regardless of how eom-mon-sensical or logically obvious those conclusions might be. That is the job of the jurors, aided by the testimony of the witnesses, and the arguments of the attorneys.

Therefore, I would overrule Sutton and hold that the trial court, although acting in good faith reliance upon our prior precedent, erred in this case in giving a jury instruction that is not part of the statutory law applicable to DWI. Although the attorneys for the State and defense may certainly argue the logical proposition of a “synergistic effect” (or lack thereof) of various substances upon one’s mental or physical faculties, a trial judge errs by instructing the jury on the significance of any witness’s opinion testimony as if that testimony were the law applicable to the case.

I would remand this case to the court of appeals to assess the harmfulness of the improper jury instruction and application paragraph.