concurring in part and dissenting in part.
I disagree with the majority’s conclusion that it is per se reversible error when a trial judge instructs a jury, prior to deliberations, that it cannot have testimony read back. Although I agree that the instruction is in error, I believe that in some cases the error can be harmless. Therefore, I would quash the Third District’s decision and remand for a harmless error analysis under State v. DiGuilio, 491 So.2d 1129 (Fla.1986). Accordingly, I respectfully concur in part and dissent in part.
The majority holds that the read-back error here is per se reversible “because it is impossible to determine the effect of the erroneous instruction on the jury without engaging in speculation.” Majority op. at 1005. I disagree. In State v. Schopp, 653 So.2d 1016 (Fla.1995), this Court receded from a similar per se reversible error ruling based upon presumed harm from the possibility of uncertainty. Specifically, in Schopp, this Court receded from its holding in Smith v. State, 500 So.2d 125 (Fla.1986), wherein this Court held that the failure to conduct a Richardson8 hearing was per se reversible error, assuming that “a reviewing court is in no position to determine from a cold record whether a discovery violation is harmless.” Schopp, 653 So.2d at 1019 (describing Smith, 500 So.2d 125). The Smith court had reasoned that “an appellate determination as to whether a Richardson violation is harmless is impossible in light of the fact that ‘[t]he purpose of a Richardson inquiry is to ferret out procedural, rather than substantive, prejudice.’ ” Id. (quoting Smith, 500 So.2d at 126). But the Schopp Court determined that there are exceptions to that stringent rule.
In Schopp, the State sought to admit the testimony of a police officer who was not included on the State’s original witness list but was added to an amended witness list shortly before trial. Id. The trial court overruled defense counsel’s objection and concluded that a Richardson hearing was not required, in part because of Schopp’s request for a speedy trial. Id. The Fourth District Court of Appeal reversed based on this Court’s opinion in Smith, although it was “ ‘absolutely convinced that the admis*1014sion of the testimony of the undisclosed witness and the failure to conduct a Richardson inquiry were harmless,’ under the harmless error analysis set out in [DiGuilio ].” Id. (quoting Schopp v. State, 641 So.2d 141, 142 (Fla. 4th DCA 1994)). The Fourth District was convinced the error was harmless because “1) the undisclosed witness presented testimony that was known to the defendant, was cumulative to other testimony, and concerned facts openly admitted by the defendant; 2) defense counsel admitted during opening statement that Schopp committed the offenses of which he was ultimately convicted; and 3) Schopp was acquitted of the charged offenses and thus ‘won’ his case.” Id.
This Court concluded that the circumstances in Schopp “demonstrate[d] that there are cases in which a reviewing court can determine that a discovery violation is harmless beyond a reasonable doubt, absent an adequate Richardson inquiry” because in that case it was clear “beyond a reasonable doubt that neither the discovery violation nor the trial court’s failure to inquire into whether corrective sanctions were warranted materially hindered the defendant’s trial preparation or strategy.” Id. at 1019-20.
This Court reiterated that “a per se rule is appropriate only for those errors that always vitiate the right to a fair trial and therefore are always harmful.” Id. at 1020 (citing DiGuilio, 491 So.2d at 1134— 35) (emphasis added). We then explained that “we have never held that the failure to conduct a Richardson hearing always results in an unfair trial. Rather, we established the per se reversal rule based on our assumption that ‘no appellate court can be certain that errors of this type are harmless.’ ” Id. (quoting Cumbie v. State, 345 So.2d 1061, 1062 (Fla.1977)). But Schopp presented a case where a review of the record revealed that this type of error could be found harmless.
As in Smith, the majority has held here that the trial court’s erroneous instruction on read-backs constitutes per se reversible error based upon uncertainty in determining whether, in some cases, this procedural error could be harmful — the same reasoning that was proved to be unworkable in Schopp. In Schopp, however, the district court performed a thorough review of the record and described the circumstances present in that case. Here, the Third District’s opinion does not discuss its review of the record but simply states that the evidence against Johnson was “overwhelming.” Johnson v. State, 10 So.3d 680, 681 (Fla. 3d DCA 2009). But finding that the evidence is overwhelming is not the test for harmless error in Florida.
In Florida, an error is harmless if there is no reasonable possibility that the error affected the verdict. DiGuilio, 491 So.2d at 1139. As this Court has explained,
[Hjarmless error analysis must not become a device whereby the appellate court substitutes itself for the jury, examines the permissible evidence, excludes the impermissible evidence, and determines that the evidence of guilt is sufficient or even overwhelming based on the permissible evidence....
[[Image here]]
The harmless error test ... places the burden on the state, as the beneficiary of the error, to prove beyond a reasonable doubt that the error complained of did not contribute to the verdict or, alternatively stated, that there is no reasonable possibility that the eiror contributed to the conviction. Application of the test requires not only a close examination of the permissible evidence on which the jury could have legitimately relied, but an even closer examination of the impermissible evidence which might have possibly influenced the jury verdict....
*1015[[Image here]]
... The test must be conscientiously applied and the reasoning of the court set forth for the guidance of all concerned and for the benefit of further appellate review. The test is not a sufficiency-of-the-evidence, a correct result, a not clearly wrong, a substantial evidence, a more probable than not, a clear and convincing, or even an overwhelming evidence test. Harmless error is not a device for the appellate court to substitute itself for the trier-of-fact by simply weighing the evidence. The focus is on the effect of the error on the trier-of-fact. The question is whether there is a reasonable possibility that the error affected the verdict. The burden to show the error was harmless must remain on the state. If the appellate court cannot say beyond a reasonable doubt that the error did not affect the verdict, then the error is by definition harmful.
Ventura v. State, 29 So.3d 1086, 1089-90 (Fla.2010) (quoting DiGuilio, 491 So.2d at 1138-39). ' I believe this Court should refrain from rejecting a harmless error standard in favor of creating this per se reversible error rule when a harmless error analysis under DiGuilio was not performed or reviewed by this Court.
In Galindez v. State, 955 So.2d 517, 518 (Fla.2007), this Court held that violations of Apprendi and Blakely are subject to a harmless error analysis.9 In Galindez, a jury found the defendant guilty of “two counts of lewd and lascivious assault on a minor and one count of child abuse by impregnating the victim.” Galindez, 955 So.2d at 519-20. Yet Galindez’s score-sheet assessed 240 points for penetration. Id. at 520. This Court concluded that any Apprendi error was harmless. Id. at 523. This Court stated that “[i]n light of the clear and uncontested record evidence of penetration ... we hold that no reasonable jury would have returned a verdict finding there was no penetration.” Id. at 524. In reaching its conclusion, this Court relied on the harmless error analysis applied by the United States Supreme Court in Washington v. Recuenco, 548 U.S. 212, 126 S.Ct. 2546, 165 L.Ed.2d 466 (2006) (reversing the Washington Supreme Court’s holding that harmless error analysis does not apply to an Apprendi error). See 955 So.2d at 521-22. This Court noted that it had “long applied” the harmless error test announced in Chapman v. California, 386 U.S. 18, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967), and outlined in DiGuilio, which we described as questioning whether it is “clear beyond a reasonable doubt that a rational jury would have found the defendant guilty absent the error.” 955 So.2d at 522 (quoting Neder v. United States, 527 U.S. 1, 18, 119 S.Ct. 1827, 144 L.Ed.2d 35 (1999)).
In Galindez, this Court applied a harmless error analysis even though the fact that led to the enhanced sentence had never been submitted to the jury. Although we did not know what the jury would have actually determined or what questions the jury would have asked if the fact enhancing the sentence had been submitted to the jury, we held that a harmless error analysis was appropriate. Similarly, here we do not know what the jury would have done absent the erroneous read-back instruction. We do not know what, if any, questions the jury would have had regarding the facts submitted to it for its consideration. And just as we perform a harmless error analysis when faced with the uncertainty surrounding an Apprendi violation, we should perform a harmless error analysis when faced with an erroneous read-back instruction.
*1016I believe that, just as with violations of Apprendi, there are circumstances under which an erroneous read-back instruction can be harmless. At least one federal court has concluded that the specific read-back error at issue here can be harmless. In United States v. White, 23 F.3d 404, 1994 WL 177280 (4th Cir.1994) (unpublished), the Fourth Circuit concluded that an erroneous read-back instruction was harmless based upon its observations that the appellant’s trial lasted just one-and-one-half days, and he was the only defendant; the announcement prohibiting read-backs was made after all the evidence was presented; four incriminating witnesses told consistent stories; the defense was not based on fine distinctions; and the appellant “point[ed] to nothing, either in general or in particular, that might have generated confusion among the jurors.” White, 1994 WL 177280 at *3. The Fourth Circuit explained:
Of course we have no way of knowing whether the jury in White’s trial would have asked for a read-back of any testimony, just as a reviewing court can never know with absolute certainty what weight a jury put on an erroneously admitted piece of evidence. It is difficult, and no doubt sometimes nigh impossible, to gauge the effect on a jury’s verdict of, say, a coerced confession, but we are bound to do so when presented with such a case. The difficulty of applying the harmless error test in some (or even most) cases, however, is an inadequate basis for declaring a per se rule for all cases.
Id.
As in White, a proper review of the record here may have revealed the error was harmless, that is, that there was no reasonable possibility that the error affected the verdict. For example, it is possible that several witnesses gave consistent testimony, or that Johnson or his counsel admitted that he fled from a police officer, or that all the facts in the case were undisputed, or any number of other possible scenarios which could have led the Third District to conclude beyond a reasonable doubt that the read-back error did not contribute to the verdict.
On the other hand, a proper analysis could have led to the conclusion that the error was harmful. Maybe a review would have revealed the presence of conflicting testimony, or an exceptionally lengthy jury deliberation, or some questionable admission of evidence, or any number of other circumstances which could have led the Third District to conclude that the error was harmful. But even finding the error harmful in this case does not mean that the error could not be found harmless in another case. As in Smith, the majority is assuming that “no appellate court can be certain that errors of this type are harmless.” Schopp, 653 So.2d at 1020 (quoting Cumbie, 345 So.2d at 1062). But “a per se rule is appropriate only for those errors that always vitiate the right to a fair trial and therefore are always harmful.” Id. (emphasis added) (citing DiGuilio, 491 So.2d at 1134-35). That this type of error is always harmful has not been proven here. Therefore, this case does not present a sufficient basis for creating a per se reversible rule, especially considering that this Court has not been apprised of all the facts. Instead, in my view, this case should be reversed and remanded so that the Third District can perform a harmless error analysis under the standards adopted by this Court in DiGuilio.
Accordingly, I respectfully concur in part and dissent in part.
LABARGA, J., concurs.
. Richardson v. State, 246 So.2d 771 (Fla.1971).
. Apprendi v. New Jersey, 530 U.S. 466, 120 S.Ct. 2348, 147 L.Ed.2d 435 (2000); Blakely v. Washington, 542 U.S. 296, 124 S.Ct. 2531, 159 L.Ed.2d 403 (2004).