concurring and dissenting, in which WILNER, J., joins; BELL, C.J., joins in Part I only; HARRELL, J. joins in Part II only.
I agree with the majority that petitioner’s rape conviction should be reversed, but for different reasons. I agree that the trial court erred in not answering the jury’s note with respect to the effect of a woman’s withdrawal of initial consent post-penetration on the elements of rape. In my view, however, Battle v. State, 287 Md. 675, 414 A.2d 1266 (1980), controlled the trial court’s response, and the court should have told the jury that if a woman consents prior to penetration and withdraws the consent following penetration, there is no rape. Nonetheless, I would expressly overrule that aspect of Battle and hold, prospectively only, “that post-penetration withdrawal of consent negates initial consent for the purposes of sexual offense crimes and, when coupled with the other elements, may constitute the crime of rape.” See State v. Baby, 404 Md. 220, 271-72, 946 A.2d 463, 493 (2008).
I disagree with the judgment of the majority reversing petitioner’s convictions for sexual offenses in the first and third degree. I would hold that the error in the failure to clarify the elements of rape has no bearing whatsoever on the sexual offenses committed by Mike and in which petitioner *273had been convicted as a principal in the second degree. Therefore, I would affirm those two convictions.
The majority holds that a woman may withdraw consent for vaginal intercourse after penetration has occurred and that, after consent has been withdrawn, the continuation of vaginal intercourse by force or the threat of force may constitute rape. Baby, 404 Md. at 260, 946 A.2d at 486. I agree with that conclusion. Because the trial court did not directly address the jurors’ confusion on the effect of withdrawal of consent during the elements of rape, the Court reverses petitioner’s convictions. The Court reverses not only for rape, but for the other convictions for first degree and third degree sexual conduct. Baby, 404 Md. at 260, 946 A.2d at 487.
The majority reasons that “the jury’s questions relating to the timing of withdrawal of consent certainly touched upon an issue central to its ability to determine whether Baby had committed the crime of first degree rape.” Baby, 404 Md. at 263, 946 A.2d at 488-89. The jury was obviously struggling with the legal definition of rape and particularly the issue of post-penetration withdrawal of consent, and I agree that the trial court should have supplemented its instructions to aid the jury.
I part company with the majority view as to whether the language in Battle was a holding or dicta. If it was dicta, as the majority states, then on remand, petitioner may be retried and the jury should be instructed that a woman’s withdrawal of consent, post-penetration, could be rape. If the language in Battle was a holding of this Court, then if petitioner is retried, the jury should be instructed to the contrary and consistent with the Battle language, because that was the law when the acts in question occurred.
As to any effect of the court’s error as to the rape instruction on the other sexual offense convictions, the majority holds that those offenses could have been infected by the trial court’s failure to clarify the instruction and reverses, stating as follows:
*274“It is true that the jury’s instructions specifically mentioned consent ‘to sex’ and ‘rape,’ and did not specifically mention the actions for which Baby was convicted on the sexual offense charges, specifically aiding and abetting Mike in an act of anal penetration and touching J.L.’s breast and vagina without her consent. Lack of consent is an element common to both rape and first and third degree sexual offenses, however. Any clarification which the jury received on the element of consent would have been applicable to its understanding of the first and third degree sexual offense counts, as well as the rape charges. A further instruction on the effect of post-penetration withdrawal of consent may have conceivably affected the jury’s verdict on the charge of rape, so it may have altered the verdicts on the sexual offense charges. Applying the Dorsey standard, we cannot conclude beyond a reasonable doubt that the trial court’s error in not clarifying the effect of post-penetration withdrawal of consent was harmless.
Baby, 404 Md. at 265-66, 946 A.2d at 490. While I agree with the majority that post-penetration withdrawal of consent could be rape, I disagree with the majority’s conclusion that the instruction on rape had an effect on the sexual offense convictions.
I. The Rape Conviction and the Jury Instruction
The rape conviction should be reversed and remanded to the Circuit Court. As the Court of Special Appeals stated, “[t]he jury, in the discharge of its responsibilities to apply the law to the facts as it found them to be, was entitled to a proper response to its inquiries.” Baby v. State, 172 Md.App. 588, 607, 916 A.2d 410, 421 (2007). Whether the Battle language was dicta or holding then becomes highly significant at any retrial of petitioner. In my view, the statement at issue from Battle was a holding.
Battle: Dicta or Holding
The majority holds that the statement, “On the other hand, ordinarily if she consents prior to penetration and withdraws *275the consent following penetration, there is no rape,” found in Battle v. State, 287 Md. 675, 684, 414 A.2d 1266, 1270 (1980), is dicta. Whether the statement is dicta or a holding is not merely an academic exercise, but instead, has real significance in the event there is a retrial of this case, for several reasons. If this Court is not now stating new law or changing the law of rape in Maryland, (and the Battle language was dicta), at any retrial, the court should instruct the jury that post-penetration withdrawal of consent may constitute rape. If the Court is announcing a new rule of law, (and the Battle language was a holding and not dicta), the jury should be instructed that if a woman actually consents to sexual intercourse prior to penetration and withdraws the consent following penetration, there is no rape. This new rule of law may be applied prospectively only. Walker v. State, 343 Md. 629, 637-40, 684 A.2d 429, 432-33 (1996).
The dicta /holding argument was presented to the Court of Special Appeals. It seems to me that the Court of Special Appeals found the Battle language to be binding on the Circuit Court, and hence, a holding. The intermediate appellate court stated as follows:
“Although Siering[1], as does the State, denominated the language at issue in Battle as ‘arguably’ dicta, Bunyard[2] refers to it as ‘the holding of Battle.’ As noted, whether the Battle pronouncement is dicta is immaterial to the trial court’s obligation to inform the jury of the current status of Maryland law. It is currently a statement of Maryland law, that has neither been overruled nor commented upon negatively. Whether it should be revisited in light of the weight of authority to the contrary is a matter for the Maryland legislature or the Court of Appeals. Under Battle, no rape occurred if the jury found that the prosecutrix withdrew her prior consent after penetration. The trial judge was obliged to answer the jury’s questions and it should have been *276advised that, under Maryland law, the answer is ‘no’ to the question, ‘If a female consents to sex initially and, during the course of the sex act to which she consented, for whatever reason, she changes her mind and the ... man continues until climax, does the result constitute rape?’ The holding in Battle, of course, would not have been a bar to a conviction for common law assault for any continuation of the sexual act against the complainant’s will after the withdrawal of consent.”
Baby v. State, 172 Md.App. at 620-621, 916 A.2d at 429 (footnote omitted).
Admittedly, it is not always easy to distinguish between dicta and a holding. The distinction between a holding and dicta, while difficult to define, is also difficult to apply. See, e.g., Richard A. Posner, The Federal Courts: Crisis And Reform 252-53 (1985) (“[R]emarkably—considering how fundamental the distinction is to a system of decision by precedent—the distinction [between holding and dictum] is fuzzy not only at the level of application but also at the conceptual level.”); Patricia M. Wald, The Rhetoric of Results and the Results of Rhetoric: Judicial Writings, 62 U. Chi. L.Rev. 1371, 1411 (1995) (“But line drawing between holding and dicta can be blurry; as we have seen, the same language of a prior opinion is often classified differently by future judges dependent on whether they like what it says or not.”). The distinction between dicta and holdings has been described as “famously elusive.” Peter J. Smith, The Marshall Court and the Originalist’s Dilemma, 90 Minn. L.Rev. 612, 637 (2006). See also Michael C. Dorf, Dicta and Article III, 142 U. Pa. L.Rev. 1997, 2003 (1994) (“[N]o universal agreement exists as to how to measure the scope of judicial holdings. Consequently, neither is there agreement as to how to distinguish between holdings and dicta.”)-, Comment, Dictum Revisited, 4 Stan. L.Rev. 509, 512 (1952) (suggesting that the word “dictum ” may mean nothing more than “I do not have to follow this case”).
Most lawyers recall learning in law school that the term “holding” refers “to a rule or principle that decides the case,” *277the ratio decidendi of the case, whereas dicta “typically refers to statements in a judicial opinion that are not necessary to support the decision reached by the court.” Dorf, Dicta and Article III, 142 U. PA. L. REV. at 2000. See also Krupnick v. Hartford Accident & Indem. Co., 28 Cal.App.4th 185, 34 Cal.Rptr.2d 39, 47 (1994) (equating ratio decidendi to holding of a case). This test is often known as the logical necessity test, and is drawn from Black’s Law Dictionary definition. See, e.g., Vogel v. State, 315 Md. 458, 466, 554 A.2d 1231, 1234 (1989); Schmidt v. Prince George’s Hospital, 366 Md. 535, 551, 784 A.2d 1112, 1121 (2001). On the other hand, some scholars and judges have proposed other definitions of dicta. An alternate definition has been proposed by Professors Michael Abramowicz and Maxwell Sterns, as follows:
“A holding consists of those propositions along the chosen decisional path or paths of reasoning that (1) are actually decided, (2) are based upon the facts of the case, and (3) lead to the judgment. If not a holding, a proposition stated in a case counts as dicta.”
Michael Abramowicz & Maxwell Stearns, Defining Dicta, 57 Stan. L.Rev. 953, 961 (2005). Discussing the concept of dicta versus a holding, Judge Alex Kozinski stated as follows:
“Of course, not every statement of law in every opinion is binding on later panels. Where it is clear that a statement is made casually and without analysis, where the statement is uttered in passing without due consideration of the alternatives, or where it is merely a prelude to another legal issue that commands the panel’s full attention, it may be appropriate to revisit the issue in a later case. However, any such reconsideration should be done cautiously and rarely—only where the later panel is convinced that the earlier panel did not make a deliberate decision to adopt the rule of law it announced. Where, on the other hand, it is clear that a majority of the panel has focused on the legal issue presented by the case before it and made a deliberate decision to resolve the issue, that ruling becomes the law of the circuit and can only be overturned by an en banc court or by the Supreme Court.”
*278United States v. Johnson, 256 F.3d 895, 915-16 (9th Cir.2001) (footnote omitted). Another view of the dicta/holding distinction is as follows:
“Courts and commentators sometimes assert that a statement is dictum if it was not essential to the outcome of the case. This formulation focuses on the facts and the outcome in the precedent case, and asks which facts were material to the decision. According to this view, elaborations of legal principle that are broader than the narrowest proposition that could have decided the case given its particular facts are considered dicta. Although frequently deployed in practice, this formulation of the distinction between holdings and dicta has serious potential for judicial manipulation to serve instrumentalist ends, and often results in a patchwork of precedent that provides little guidance to lower courts and potential litigants. The competing explanation of the distinction between holdings and dicta focuses on the rationale of the precedent case. Under this definition, courts are bound by statements that form part of the rationale of the decision in the precedent case, even if, when viewed from a post hoc perspective, they were not technically essential to the result. Although this formulation of the distinction between holdings and dicta raises definitional problems of its own—specifically, there is no neat way to determine the ‘rationale’ of the precedent decision—it is more consistent with the rule of law and a greater obstacle to judicial instrumentalism.
Peter J. Smith, The Marshall Court and the Originalist’s Dilemma, 90 Minn. L.Rev. 612, 638-39 (2006) (footnotes omitted). Finally, commentators and courts have distinguished between obiter dicta and judicial dicta (sometimes referred to as considered dicta), noting as follows:
“The line between dicta and holding is further clouded by the presence of a middle ground, labeled ‘judicial dicta.’ Judicial dictum is generally defined as ‘an opinion by a court on a question that is directly involved, briefed, and argued by counsel, and even passed on by the court, but that is not essential to the decision.’ Statements of judicial *279dicta are technically ‘dicta ’ because they are not necessary to the holding of a case. They do not, however, implicate to the same degree as ordinary dicta the concern of ‘full consideration,’ which is one of the rationales for treating dicta and holdings differently. Unlike ordinary dicta, judicial dicta is, by definition, well-reasoned and stated only after the court has investigated an issue with care. Accordingly, courts afford judicial dicta greater deference than ordinary dicta, treating judicial dicta almost like holdings.”
David Coale & Wendy Couture, Loud Rules, 34 Pepp. L.Rev. 715, 727-28 (2007).
The idea that the distinction between holding and dicta is not easily discerned is not a new concept. Indeed, in Carstairs v. Cochran, 95 Md. 488, 52 A. 601 (1902), our predecessors opined on the subject, stating as follows:
“We cannot agree that the expression of opinion referred to was an obiter dictum. All the constitutional objections which are here urged, were urged in that case by the late Judge Fisher with all the ability and force for which he was so justly distinguished, and the ruling upon the prayers, which there constituted the single exception, required the consideration by the Court of all those objections. It 'may be difficult to frame a concise definition of an obiter dictum applicable to every such expression of opinion, and some Courts incline to the rule that the most deliberate expression of opinion, upon a question distinctly raised in the record, and fully argued by counsel, may nevertheless be regarded as a dictum, unless essential to the actual disposition made of the case. But as Bouvier well says: ‘It is difficult to see why, in a philosophic point of view, the opinion of the Court is not as persuasive on all the points which were so involved in the cause that it was the duty of counsel to argue them, and which were deliberately passed on by the Court, as if the decision had hung upon but one point;’ and in Maryland the rule is in accord with this view. In Alexander v. Worthington, 5 Md. 489, it is said: ‘All that is necessary in Maryland to render the decision of the Court of Appeals authoritative on any point decided, is to show *280that there was an application of the judicial mind to the precise question adjudged;’ and in Michael v. Morey, 26 Md. 261, it was said that a decision there cited, could not be said to be obiter dictum, ‘as the question was directly involved in the issues of law raised by the demurrer to the bill, and the mind of the Court was directly drawn to, and distinctly expressed upon the subject.’ ”
Id. at 499, 52 A. at 602.
I do not believe that the language at issue in Battle was obiter dicta or judicial dicta but instead believe that it was part of the ratio decidendi of the case and therefore a holding. The Battle statement was hardly the expression of a court on a collateral question or a mere illustration originating with the author of the opinion. The question was directly involved in the issues raised in the case. The statement was hardly a “by the way” statement, incidental to the question in the case. The Battle jury sought to determine when, in point of time, a withdrawal of consent would sustain a conviction for rape.
Significantly, a review of the briefs filed in the Battle case indicate that this issue was raised and argued in the brief of the appellant. The record reflects that appellant’s counsel objected to the court’s jury instruction, arguing “that the instruction was wrong because it incorrectly told the jury that consent could be withdrawn during the sex act thus converting consensual intercourse to rape.” Brief for Appellant at 6, Battle v. State, 287 Md. 675, 414 A.2d 1266 (1980) (No. 159). Appellant then argued as follows:
“The jury may have been unsure about whether consent could be withdrawn while sexual intercourse (penetration) was taking place. The answer to that would have been ‘no.’ Hazel v. State, 221 Md. 464, 469, 157 A.2d 922 (1960).”
Id. at 12.
The trial court should have instructed the jury in the language of Battle.
*281II. The First Degree and Third Degree Sexual Offenses
Baby’s convictions for first degree and third degree sexual offense, based on his conduct as a principal in the second degree to the criminal, sexual conduct of his friend, Mike, should be affirmed. These charges and convictions were unrelated in any manner to the question posed by the jury. The sexual offense charges arise from separate and discrete acts, that of aiding and abetting Mike in an act of anal penetration and the touching of J.L.’s breasts and vagina without her consent, which are entirely unrelated to the issue of post-penetration withdrawal of consent during the separate act of sexual intercourse.
Even though lack of consent is an element to the rape and sexual offenses, there was absolutely no evidence that the victim consented to Mike’s acts of first degree and third degree sexual offense, offenses in which the jury found petitioner to be a principal in the second degree. The majority’s statement that “[a] further instruction on the effect of post-penetration withdrawal of consent may have conceivably affected the jury’s verdict on the charge of rape, so it may it have altered the verdicts on the sexual offense charges ” is a bald conclusion, with no reasoning or facts to support it. Baby, 404 Md. at 266, 946 A.2d at 490. A plain reading of the jury question makes it crystal clear that the question related only to the rape charges against Baby. As the Court of Special Appeals said:
“The plain meaning of the jury’s words, ‘during the sex act,’ leads one ineluctably to conclude that the reference was to the act of intercourse.... [I]f there had been initially any cause for confusion, it certainly should have been cleared up when the jury submitted the second note the following morning: ‘If at any time, the woman says stop, is that rape?’.... A fair interpretation of the jury’s question is that it was an inquiry as to the legal effect of a withdrawal of consent subsequent to penetration, and prior to climax.”
Baby, 172 Md.App. at 605-06, 916 A.2d at 420. Moreover, the absence of any evidence of the victim consenting to Mike’s *282acts makes the Court’s ipse dixit conclusion is too great a leap. The first degree and third degree sexual offense convictions should be affirmed.
Judge WILNER authorizes me to state that he agrees with and joins this concurring and dissenting opinion. Chief Judge BELL authorizes me to state that he agrees with and joins Part I only. Judge HARRELL authorizes me to state that he agrees with and joins Part II only.
. State v. Siering, 35 Conn.App. 173, 644 A.2d 958 (1994).
. State v. Bunyard, 31 Kan.App.2d 853, 75 P.3d 750 (2003).