Appellant, Darius Smith, was convicted after a jury trial of second degree murder while armed and related weapons offenses.1 Over Smith’s objection, the trial court instructed the jury on the meaning of reasonable doubt using portions of FEDERAL Judicial CenteR, PatteRN Criminal JURY Instructions, No. 21 (1988) (FJC instruction) instead of the standard instruction generally used in Superior Court, i.e., Criminal Jury Instructions for the District of Columbia No. 2.09 (4th ed. 1993) (“Redbook” instruction), for which Smith expressed a preference. See Smith v. United States, 687 A.2d 1356, 1359-60 & n. 1 (D.C. 1996). In affirming the judgment of conviction, a division of this court held that “the [trial] court’s instruction, taken as a whole, correctly conveyed the government’s burden of proof under the reasonable doubt standard and did not shift or lessen that burden of proof.” Id. at 1358. We granted rehearing en banc to decide whether to adopt or approve a new instruction on reasonable doubt to replace Redbook instruction 2.09.2 After consideration of the comprehensive briefs from the parties and amicus curiae, we conclude that the standard Redbook instruction should be modified in the manner set forth in this opinion. Our reconsideration does not alter the division’s conclusion that the instruction which the trial court gave in this case did not deprive Smith of due process of law. Therefore, we affirm the judgment of conviction.

I.

The standard of proof beyond a reasonable doubt “is an ancient and honored aspect of our criminal justice system,” although one which “defies easy explication.” Victor v. Nebraska, 511 U.S. 1, 5, 114 S.Ct. 1239, 1242, 127 L.Ed.2d 583 (1994). In a criminal trial, application of that standard of proof is a requirement of due process. See id.; Cage v. Louisiana, 498 U.S. 39, 39-40, 111 S.Ct. 328, 329, 112 L.Ed.2d 339 (1990) (citing In re Winship, 397 U.S. 358, 364, 90 S.Ct. 1068, 1072, 25 L.Ed.2d 368 (1970)(other citations omitted)). The Supreme Court has held that the Constitution neither requires nor prohibits trial courts from defining proof beyond a reasonable doubt. See Victor, 511 U.S. at 5,114 S.Ct. at 1242. However, when a definition of the standard of proof is given to the jury, the instructions, taken as a whole, must convey the concept properly. See Holland v. United States, 348 U.S. 121, 140, 75 S.Ct. 127, 137, 99 L.Ed. 150 (1954).

This court en banc has never considered the issue, but a division has held that the trial court must instruct the jury on reasonable doubt. See Butler v. United States, 646 A.2d 331, 337 (D.C. 1994), cert. denied, 514 U.S. 1009, 115 S.Ct. 1326, 131 L.Ed.2d 206 (1995). Although some courts have held that no attempt should be made to define the concept of reasonable doubt to the jury, we reaffirm the requirement that reasonable doubt be explained to the juries in Superior Court.3 The elimination of the requirement of an explanatory instruction would be a marked departure from the long-standing practice in this jurisdiction. Moreover, it would relegate the most fundamental, and perhaps decisive, principle in a criminal trial to random interpretation by counsel and jurors. Reasonable doubt is “perhaps the most important aspect of the closing instruction to *80the jury in a criminal trial.” Dunn v. Perrin, 570 F.2d 21, 25 (1st Cir.), cert. denied, 437 U.S. 910, 98 S.Ct. 3102, 57 L.Ed.2d 1141 (1978); see also Jackson v. Virginia, 443 U.S. 307, 315, 99 S.Ct. 2781, 2787, 61 L.Ed.2d 560 (1979) (reasonable doubt has “traditionally been regarded as the decisive difference between culpability and civil liability’). Lay jurors should not be left to undertake the important task of deciding whether the government has proven “the guilt of the accused beyond a reasonable doubt without some intelligent statement of its meaning[.]” Egan v. United States, 52 App. D.C. 384, 393, 287 F. 958, 967 (1923).

In the strongest terms, a division of the court previously approved the definition for reasonable doubt set forth in Redbook instruction No. 2.09.4 See Butler, supra, 646 A.2d at 337. The Redbook instruction on reasonable doubt has been used in this jurisdiction for decades and approved repeatedly, not only by this court but by the D.C. Circuit as well. See, e.g., id.; Foreman, supra note 4, 633 A.2d at 794; Baptist v. United States, 466 A.2d 452, 459 (D.C. 1983); Moore v. United States, 120 U.S.App. D.C. 203, 204 & n. 4, 345 F.2d 97, 98 & n. 4 (1965).5 The Supreme Court also has approved some of the definitional terms from the Redbook instruction. See, e.g., Victor, supra, 511 U.S. at 14-15, 114 5.Ct. at 1247 (“An instruction cast in terms of an abiding conviction as to guilt, without reference to moral certainty, correctly states the government’s burden of proof.”); Wilson v. United States, 232 U.S. 563, 570, 34 S.Ct. 347, 349, 58 L.Ed. 728 (1914) (reasonable doubt defined accurately as an ‘(abiding conviction of defendant’s guilt” and a doubt which would cause a juror to “pause and hesitate” before acting in important personal affairs).6 Thus, although arguably not ideal, the Redbook instruction has withstood over time various constitutional challenges.

The Supreme Court has made clear that a constitutionally deficient reasonable doubt instruction is not subject to harmless error analysis and will require reversal. See Sullivan v. Louisiana, 508 U.S. 275, 279-80, 113 S.Ct. 2078, 2081-82, 124 L.Ed.2d 182 (1993) (citation omitted).7 Therefore, a devi*81ation from the instruction in a way which misdescribes or lessens the government’s burden of proof is severely prejudicial to the fairness of the trial and the administration of justice. See Butler, supra, 646 A.2d at 337. For that reason, we have warned that “[e]f-forts to reformulate [the Redbook] instruction should proceed with extreme caution, for ‘[a]n instruction central to the determination of guilt or innocence may be fatally tainted by even a minor variation which tends to create ambiguity.’ ” Foreman, supra note 4, 633 A.2d at 794 (quoting United States v. Alston, 179 U.S.App. D.C. 129, 135, 551 F.2d 315, 321 (1976) (footnote omitted)). In spite of our strong admonitions to the trial court of the unnecessary risks attendant to deviations from the tried and tested Redbook instruction, some judges have modified the language, no doubt in an effort to achieve greater clarity.8 In turn, this has resulted in a number of appeals challenging the deviations as constitutionally deficient. See, e.g., Proctor, supra note 8, 685 A.2d 735, 738 (trial court revised and deleted parts of the Redbook instruction); Butler, supra, 646 A.2d at 335-36 (same); Foreman, supra note 4, 633 A.2d at 794 (trial court substituted different language for phrase “abiding conviction” in Redbook instruction). In the present ease, the trial court abandoned Redbook instruction 2.09 and gave instead one patterned after the FJC instruction on reasonable doubt. See Smith, supra, 687 A.2d at 1359-60 & n. 2.

This case by case approach creates a lack of uniformity on a fundamental principle applicable to every criminal trial. Approval of a standard reasonable doubt instruction, particularly one which eliminates much of the criticism surrounding the current Redbook mstruction, would promote uniformity and avoid the pitfalls of ad hoc interpretations and repetitive constitutional challenges. Adherence to an approved instruction would relieve the appellate court of “the necessities and niceties—and the concomitant uncertainties—of gauging various ... renditions [of the reasonable doubt instruction] in terms of’ their risk of misdescribing the government’s burden of proof. United States v. Thomas, 146 U.S.App. D.C. 101, 110, 449 F.2d 1177, 1186 (1971) (en banc). A standard instruction on this central principle of a criminal trial would thus conserve judicial resources. See Winters v. United States, 317 A.2d 530, 532 (D.C. 1974) (en banc).9 Furthermore, “[a]llowing varying definitions ... detracts from the goal of a uniform and equal system of justice.” State v. Portillo, 182 Ariz. 592, 898 P.2d 970, 974 (1995). In a matter central to the determination of guilt or innocence, as this is, the appearance of evenhandedness, like the actuality, is important. “Use of a standard definition thus will eliminate confusion and foster fairness for defendants, the [government], and jurors alike.” Id. At the same time, the approval of a single instruction for use in all criminal trials will not intrude unduly into the area of trial court discretion, for the standard of proof beyond a reasonable doubt is applicable to every criminal trial and is not subject to change because of the evidence or legal theories presented. For all of these reasons, we are persuaded that the approval of a uniform, modified reasonable doubt instruction is appropriate.

II.

The parties in this case and amicus all agree that some modification of the Red-*82book instruction is warranted, although they disagree on its formulation. After reviewing various alternatives, ably presented by the counsel for the parties and amicus curiae, we have formulated an instruction which we believe fairly and accurately conveys the meaning of reasonable doubt and responds to the most substantial criticisms of the Red-book instruction made by some Superior Court judges and by the institutional litigants representing the parties in this case. The instruction retains proven language from the Redbook instruction, but also borrows helpful language from the FJC instruction. The instruction which we adopt is as follows:

The government has the burden of proving the defendant guilty beyond a reasonable doubt. In civil cases, it is only necessary to prove that a fact is more likely true than not, or, in some cases, that its truth is highly probable. In criminal cases such as this one, the government’s proof must be more powerful than that. It must be beyond a reasonable doubt.

Reasonable doubt, as the name implies, is a doubt based upon reason—a doubt for which you have a reason based upon the evidence or lack of evidence in the case. If, after careful, honest, and impartial consideration of all the evidence, you cannot say that you are firmly convinced of the defendant’s guilt then you have a reasonable doubt.

Reasonable doubt is the kind of doubt that would cause a reasonable person, after careful and thoughtful reflection, to hesitate to act in the graver or more important matters in life. However, it is not an imaginary doubt, nor a doubt based on speculation or guesswork; it is a doubt based upon reason. The government is not required to prove guilt beyond all doubt, or to a mathematical or scientific certainty. Its burden is to prove guilt beyond a reasonable doubt.

This instruction maintains the essential concepts required for defining proof beyond a reasonable doubt, but it eliminates expressions from earlier versions of the Redbook instruction which may be viewed as outdated in favor of simpler ones. The formulation “hesitate to act in the graver or more important matters in life,” has been approved by the Supreme Court. See Holland, supra, 348 U.S. at 140, 75 S.Ct. at 137; see also Victor, supra, 511 U.S. at 20-21, 114 S.Ct. at 1250-51.10 The first paragraph is taken from the FJC instruction, with modifications to dispel any claim that jurors will not grasp that the reasonable doubt standard is the highest burden of proof required in a court of law. We have opted to include this language in spite of some suggestion that it introduces legal concepts about civil burdens of proof with which the jurors in criminal trials need not concern themselves. However, even with our one-day/one-trial system, jurors often are called to serve many times and may have had experience on a civil jury. This explanation should serve to avoid any confusion about different standards of proof with which they may be familiar, and at the same time reflect by comparison the very heavy burden of proof required in a criminal trial.

III.

There is ample precedent for the court’s exercise of its supervisory authority to develop a jury instruction to assure fairness and to conserve judicial resources by eliminating uncertainties which result in multiple challenges on appeal.11 Given the great risks to the integrity of the trial which attend a deficient reasonable doubt instruction, the uncertainties and controversies generated by varying definitions, and the importance of fairness and the appearance of fairness in our justice system, the greater part of wisdom would dictate that the trial court give the standard instruction approved here, which has been determined to be faithful to *83the constitutional meaning of reasonable doubt. Therefore, we state, in the strongest terms, that the trial court should “resist the temptation to stray from, or embellish upon, that instruction.” Wills v. State, 329 Md. 370, 620 A.2d 295, 304 (1993)(McAuliffe, J., concurring). As the Supreme Court of New Jersey has stated:

We believe the trial courts will better serve the interest of justice if they do not attempt additional definitions of “reasonable doubt.” ... The failure to adhere to the definition, over objection, runs the risk of reversible error.

State v. Medina, 147 N.J. 43, 685 A.2d 1242, 1252 (1996), cert. denied, — U.S. -, 117 S.Ct. 1476, 137 L.Ed.2d 688 (1997). We adopt this revised uniform instruction prospectively only. See Waits, supra note 11, 362 A.2d at 711.

IV.

Finally, as stated at the outset, the division correctly held that the reasonable doubt instruction given in this ease did not deprive Smith of due process, and our reconsideration en banc does not alter the division’s opinion, which we hereby reinstate. The judgment of conviction is, therefore,

Affirmed.

concurring in the judgment:

“[S]o long as the court instructs the jury on the necessity that the defendant’s guilt be proven beyond a reasonable doubt, ... the Constitution does not require that any particular form of words be used in advising the jury of the government’s burden of proof.” Victor v. Nebraska, 511 U.S. 1, 5, 114 S.Ct. 1239, 1243, 127 L.Ed.2d 583 (1994) (citations omitted). There are thus many constitutionally acceptable ways in which the judge can “correctly convey the concept of reasonable doubt to the jury.” Id. (citation omitted).

When several permissible alternatives are available, “[ajppellate judges ought not to substitute their ... notions of apt phraseology [in jury instructions] for the experience of trial judges in such matters; our more limited responsibility is to ensure that the law is correctly stated for jurors to apply.” United States v. Yunis, 288 U.S.App. D.C. 129, 139, 924 F.2d 1086, 1096 (1991). Substantially for the reasons stated in my separate opinion in Proctor v. United States, 685 A.2d 735, 744-45 & n. 6 (D.C. 1996), I continue to believe that this court ought not to prescribe which particular definition of reasonable doubt a trial judge must or should utilize in instructing the jury. Instead, like our colleagues in Maryland, we should “refuse[] to adopt a boilerplate explanation of the term so long as, “when an explanation is given to the jury, it does not tend to confuse [or] mislead [the jurors] or prejudice the accused.’ ” Hunt v. State, 345 Md. 122, 691 A.2d 1255, 1269 (1997), cert. denied, — U.S. -, 117 S.Ct. 2536, 138 L.Ed.2d 1036 (1997) (ellipsis omitted) (quoting Wills v. State, 329 Md. 370, 620 A.2d 295, 301 (1993)).

If a ipodel decree is to be adopted, I should think that it ought to have an experiential basis founded upon field-tested information regarding the kind of phraseology which is most readily understandable to lay jurors. As appellate judges, our expertise is in the law, rather than in communicating with citizens who are not trained in our craft.1 The disciplines of law and communication are quite different from one another, and when a choice is being made between legally acceptable formulations of reasonable doubt, the most useful thing to know is which one will best enlighten the jury. The differences between the two disciplines reinforce my view that we should follow Chief Judge Mikva’s counsel in Yunis, swpra2

*84My colleagues do not share my reservations, however, and I acknowledge that the model instruction that they have collectively composed is constitutionally unassailable. In terms of understandability for the jurors, it is doubtless an improvement upon the somewhat stilted prose of the old Redbook instruction which was deemed de rigueur until today. See, e.g., Butler v. United States, 646 A.2d 331, 333-35 (D.C. 1994), cert. denied, 514 U.S. 1009, 115 S.Ct. 1326, 131 L.Ed.2d 206 (1995).3 I also agree with the court that the trial judge’s instruction in the present case did not deprive Smith of liberty without due process of law. Accordingly, I concur in the judgment.

concurring:

I fully join the court’s endorsement of a revised reasonable doubt instruction. As the majority opinion notes, in view of the central importance of that instruction to criminal trials and in the interest of uniformity and conservation of judicial resources, pursuant to the appellate court’s supervisory power over the trial court, it is proper for the Court of Appeals to invest its time and authority in developing a revised reasonable doubt instruction for use in all criminal trials. It is correspondingly incumbent on the trial court to conform its practice accordingly. In view of our prior decisions endorsing the Redbook instruction on reasonable doubt, see, e.g., Butler v. United States, 646 A.2d 331, 337 (D.C. 1994), the full court appropriately announces the revised rule sitting en banc. In the unanimous opinion of the Court of Appeals, the revised instruction we have formulated is not only constitutional but also well-designed to address the deficiencies that have been identified in the Redbook instruction.

I disagree with Judge Schwelb that we should refrain from acting beyond deciding the immediate case. We not only have the power to endorse an instruction pursuant to our supervisory authority, but the responsibility to do so in order to respond in a comprehensive manner to systemic issues even if they come to our attention in the context of individual eases.1 When we so act, however, we must be mindful of the limitations inherent in our adjudicative proceedings. If we are going to develop and announce an instruction for general application, as we do here, we should act on as informed a basis as we possibly can. Accordingly, we sensibly sought advice not only from the parties, but also from appointed amicus curiae, who in turn sought information from the trial judges who daily instruct juries on reasonable doubt, observe their reactions, and deal with their questions. The revised rule reflects their comments and incorporates a number of their suggestions.

My sole reservation is that we have not gone far enough to seek the views of jurors *85themselves in order to understand what terms in the current instruction are problematic, and become better informed as to what substitute language would be most helpful in conveying the core but slippery concept, “beyond a reasonable doubt.” In truth, it is difficult for an appellate court, in the context of an appeal, to obtain this kind of information. It would have been possible in this case, however, just as we requested that amicus consult with trial judges, to request that jurors be consulted, or that data and recommendations be obtained from expert individuals or organizations with the appropriate juror testing and sampling techniques to provide us with necessary information on what jurors think and would prefer. My point is not that we should not act—we should—but that we tailor our proceedings consistent with the task at hand.