Opinion
No. 1D98-4280.
May 17, 2001.
Appeal from the Circuit Court, Gulf County, Glenn L. Hess, J.
Amok Hill, Pro Se.
Robert A. Butterworth, Attorney General, and Douglas T. Squire, Assistant Attorney General, Tallahassee, for Appellee.
The appellant challenges the order by which the trial court denied his Florida Rule of Criminal Procedure 3.850 motion for postconviction relief. Only one of the claims presented in the motion requires discussion, the claim that the appellant's trial counsel provided ineffective assistance by failing to request an instruction as to a lesser included offense. Although we reverse the trial court's denial of this claim, we certify a question to the supreme court in this regard. We affirm the order in all other respects.
The appellant was convicted as charged of the offense of possession of contraband (cannabis) in a state correctional institution. Although the information and the proofs at trial would have supported an instruction as to the one-step-removed permissive lesser included offense of possession of cannabis, see generally Wilcott v. State, 509 So.2d 261 (Fla. 1987), the appellant's trial counsel did not request the instruction and it was not given.
A valid claim of ineffective assistance of counsel is presented when a defendant shows (1) that the acts or omissions of counsel fell below the standard of reasonably effective assistance, and (2) that there is a "reasonable probability" that, but for counsel's ineffectiveness, the result of the proceeding would have been different. Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). A reasonable probability is a probability sufficient to undermine confidence in the outcome. Id. at 694, 104 S.Ct. 2052.
A competent defense attorney will sometimes decline to request an instruction on a lesser included offense as a matter of reasonable trial tactics. When this occurs, a subsequent finding of deficient performance under the first Strickland prong (the performance prong) will be foreclosed. But trial counsel in the present case testified at the hearing on the 3.850 motion that her failure to request the instruction was not a trial tactic. She testified that it was instead motivated by her belief that the appellant was not entitled to the instruction because it was a permissive, rather than necessary, lesser included offense. Because this testimony indicated that trial counsel's failure to request the instruction was rooted in her failure to conduct the basic research that would have revealed the Wilcott decision, it provided a sufficient evidentiary basis upon which the trial court might have found the performance prong of the Strickland test established. But the trial court failed to recognize the potential evidentiary significance of this testimony, accepting instead the appellant's trial counsel's incorrect recitation of the relevant law.
Our decision in Bateson v. State, 516 So.2d 280 (Fla. 1st DCA 1987), provides authority for the proposition that the second prong of the Strickland test (the prejudice prong) was also established in the present case. Bateson involved review of the trial court's denial of Bateson's 3.850 claim that his trial counsel had been ineffective by failing to request instructions as to lesser included offenses. After recognizing that a defendant's entitlement to an instruction on lesser included offenses is grounded upon his right to an instruction that affords the jury an opportunity to exercise its "pardon power," we concluded that Bateson's claim was colorable because it was conceivable that defense counsel's failure to request a jury instruction as to the lesser included offenses had deprived Bateson of the possibility of a jury pardon. This reasoning has been applied in many cases from this and other Florida courts over the years since Bateson was decided. See, e.g., Peffley v. State, 766 So.2d 418 (Fla. 4th DCA 2000); McClendon v. State, 765 So.2d 247 (Fla. 1st DCA 2000); Critton v. State, 668 So.2d 242 (Fla. 1st DCA 1996); Oehling v. State, 659 So.2d 1226 (Fla 5th DCA 1995); Kennedy v. State, 637 So.2d 987 (Fla. 1st DCA 1994); Newton v. State, 527 So.2d 876 (Fla. 2d DCA 1988). See also Bethea v. State, 767 So.2d 630 (Fla. 5th DCA 2000). But the trial court in the present case did not consider these authorities.
Accordingly, because the trial court did not properly consider the evidentiary and decisional support for the appellant's claim, we must reverse the order to the extent that it reflects a denial of the claim and remand this case to the trial court for further consideration in light of the foregoing discussion. However, we do so with reservation.
Although the decisional law from the district courts, beginning with the decision in Bateson, indicates that the prejudice prong of the Strickland test may be satisfied in the circumstances presented here, we confess our inability fully to embrace the rationale for these decisions. Strickland indicates that prejudice will exist in conjunction with an ineffective assistance of counsel claim only where it can be said that there is a "reasonable probability" that counsel's deficient performance affected the outcome of the proceeding. But Bateson seems to address the question of prejudice from a very different perspective. It contains no discussion of whether there is a reasonable probability that, given the opportunity, the jury would have declined to follow the law and granted Bateson a jury pardon. Instead, the Bateson court concluded that prejudice could be found because it was "conceivable" that defense counsel's failure to request a jury instruction as to the lesser included offenses had deprived the appellant of the "possibility" of a jury pardon. Subsequent decisions have not articulated any other rationale for a finding of prejudice in these circumstances.
When a trial court improperly denies a defense request for an instruction as to a lesser included offense, the trial court commits error which may be corrected on direct appeal. Wilcott is one of many decisions standing for this proposition. But the appellant's motion in the present case was not offered in conjunction with a direct appeal. It was offered in the context of a collateral proceeding. Significantly, the test for prejudicial error in conjunction with a direct appeal is very different from the test for prejudice in conjunction with a collateral claim of ineffective assistance. There are different tests because, once a conviction becomes final, a presumption of finality attaches to the conviction. See Brecht v. Abrahamson, 507 U.S. 619, 633, 113 S.Ct. 1710, 123 L.Ed.2d 353 (1993); Goodwin v. State, 751 So.2d 537, 546 (Fla. 1999). Thus, as Goodwin explains, the test for prejudice on direct appeal is the harmless error test of Chapman v. California, 386 U.S. 18, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967), under which trial court error will result in reversal unless the prosecution can prove "beyond a reasonable doubt" that the error did not contribute to the verdict obtained. Conversely, however, as explained in Strickland, prejudice may be found in a collateral proceeding in which ineffective assistance of counsel is claimed only upon a showing by the defendant that there is a "reasonable probability" that counsel's deficient performance affected the outcome of the proceeding.
It is entirely reasonable that the Chapman test is not satisfied where a trial court has improperly failed to honor a defense request for an instruction on a lesser included offense. Because we know that jury pardons are occasionally awarded by aberrant juries, it would be difficult for an appellate court to conclude beyond a reasonable doubt that a jury in a particular case, given the opportunity, would not disobey the law and grant a pardon. And if a trial court's refusal to grant a defense request for an instruction as to an applicable lesser included offense were considered harmless upon a finding of guilt as to the greater offense, there would be no remedy for a trial court's failure to comply with the law relating to the giving of instructions for lesser included offenses. A finding of prejudice in the direct appeal context is therefore appropriate both as a proper application of the Chapman test and as a necessary enforcement mechanism.
But these considerations do not apply in the context of a collateral proceeding in which ineffective assistance of counsel is claimed for failure to request an instruction as to a lesser included offense. In this context, there has been no trial court error. And, although it is "conceivable" that a jury in a given case might decline to follow the law and grant a jury pardon, this does not seem to us a reasonable probability. We recognize that a finding of reasonable probability under Strickland does not require a finding that it is more likely than not that the deficient performance of counsel affected the outcome of the proceeding. It requires only a finding that the deficient performance put the whole case in such a different light as to undermine the court's confidence in the outcome of the proceeding. See, e.g., Robinson v. State, 770 So.2d 1167, 1171-73 (Fla. 2000) (Anstead, J., specially concurring). But we have difficulty accepting the proposition that there is even a substantial possibility that a jury which has found every element of an offense proved beyond a reasonable doubt, would have, given the opportunity, ignored its own findings of fact and the trial court's instructions on the law and found a defendant guilty of only a lesser included offense. In fact, we confess some discomfort with the proposition that members of the judiciary should even engage in such speculation. And we believe our reservations find support in the reasoning employed by the supreme court in Gragg v. State, 429 So.2d 1204 (Fla. 1983).
Gragg was charged with aggravated battery by use of a firearm, aggravated assault by use of a firearm, and possession of a firearm by a convicted felon. The possession of a firearm count was severed, and Gragg proceeded to trial on the other two charges. At the conclusion of the evidence, the jury was properly instructed that it could not return a verdict of guilty as charged unless it found that Gragg had used a firearm to commit the offenses. The jury returned verdicts of guilty of the lesser included offenses of simple battery and simple assault. Claiming that the jury's verdicts collaterally estopped the prosecution from asserting that he had possessed a firearm in conjunction with the remaining charge of possession of a firearm by a convicted felon, Gragg moved for, and was granted, a dismissal of that charge.
The supreme court ultimately concluded that the trial court had properly applied the rule of collateral estoppel (a component of the Fifth Amendment guarantee against double jeopardy) in dismissing the possession of a firearm charge. The supreme court noted that where a previous judgment of acquittal was based upon a general verdict, the rule of collateral estoppel requires a court to "examine the record of a prior proceeding, taking into account the pleadings, evidence, charge, and other relevant matter, and conclude whether a rational jury could have grounded its verdict upon an issue other than that which the defendant seeks to foreclose from consideration." Id. at 1206 (quoting Ashe v. Swenson, 397 U.S. 436, 90 S.Ct. 1189, 25 L.Ed.2d 469 (1970)) (emphasis supplied). Responding to an argument by the state that the verdicts in Gragg's trial on the charges of aggravated battery and aggravated assault might have been the product of a jury pardon, the supreme court held that "courts should not speculate on whether the jury has reached its verdict through compassion or compromise." Id. Instead, a court "should limit its inquiry to whether there was a factual basis, rather than an emotional basis, upon which the jury's verdict could have rested." Id. at 1207. The supreme court therefore held, at least in the context of a collateral estoppel analysis, that courts may not assume that "rational" juries ignore the instructions given them by trial judges and pardon defendants.
Gragg does not, however, address whether courts should, in other contexts, engage in speculation that a jury might act in a manner that is not rational. The question presented by the present case may be expressed as follows:
IN CONDUCTING A PREJUDICE ANALYSIS PURSUANT TO A POSTCONVICTION STRICKLAND V. WASHINGTON INEFFECTIVE ASSISTANCE OF COUNSEL CLAIM, MAY A COURT FIND A "REASONABLE PROBABILITY" THAT, HAD THE JURY BEEN GIVEN THE OPPORTUNITY TO RETURN A VERDICT OF GUILTY OF ONLY A LESSER INCLUDED OFFENSE, THE JURY WOULD HAVE DONE SO, THUS IGNORING ITS OWN FINDINGS OF FACT AND THE TRIAL COURT'S INSTRUCTIONS ON THE LAW?
Because speculation by a trial court in this context seems to us no less questionable than the speculation condemned by the supreme court in Gragg, and because this is a recurring issue in the courts of this state, we certify the foregoing question to the supreme court as a question of great public importance.
The trial court's denial of the claim discussed herein is reversed and this case is remanded for further consideration of the claim. The order is otherwise affirmed.
BOOTH, J., CONCURS; ERVIN, J., CONCURS WITH RESULT.