This is the second appearance of the case before this Court. In the first appeal, Hudson v. State, 309 Ga. App. 580 (711 SE2d 95) (2011) (“Hudson I”), we vacated Claude Wayne Hudson’s sentences for the offenses of aggravated sexual battery (OCGA § 16-6-22.2 (b)) and child molestation (OCGA § 16-6-4 (a) (1)), because the trial court failed to merge the convictions, and we remanded the case for resentencing. Hudson now appeals the new sentence entered upon remand. Hudson contends that the trial court violated his due process rights by increasing his custodial sentence for child molestation because the new sentence was more severe than the original sentence. For the reasons set forth below, we reverse and remand this case again for the trial court to reconsider Hudson’s sentence.
The facts are fully set forth in Hudson I, supra, 309 Ga. App. at 580. Following a jury trial, Hudson was convicted of aggravated sexual battery and child molestation based on evidence that he had molested his niece. See id. The trial court sentenced Hudson to life on the aggravated sexual battery charge, with 25 years to serve and the remainder probated. The trial court sentenced Hudson to thirty years on the child molestation charge, with ten years to serve and the remainder probated. Id. The two sentences were to run concurrently. On appeal from those sentences, this Court held that Hudson’s conviction for aggravated sexual battery should have merged with the conviction for child molestation. Id. at 582 (2). We therefore vacated the sentences and remanded for resentencing. Id. Following remand, the trial court resentenced Hudson to thirty years on the child molestation charge, increasing the custodial term to twenty-five years, rather than ten years, and probating the remainder.
On appeal, Hudson argues that the trial court violated his due process rights under the principle articulated in North Carolina v. *55Pearce, 395 U. S. 711 (89 SC 2072, 23 LE2d 656) (1969), when it increased his custodial sentence on the child molestation conviction by 15 years. In Pearce, supra, 395 U. S. at 724-725 (II) (C), the Supreme Court of the United States created a presumption of vindictiveness that arises whenever a more severe sentence is imposed after a defendant has successfully attacked his sentence on appeal. The presumption of vindictiveness, however, may be overcome by objective information in the record justifying the increased sentence. See id. at 726 (II) (C). The evil sought to be prevented in Pearce is vindictiveness of a sentencing judge, rather than simply enlarged sentences upon resentencing. See Texas v. McCullough, 475 U. S. 134, 138 (II) (106 SC 976, 89 LE2d 104) (1986).
The precedents from the Supreme Court of Georgia are conflicting as to the proper methodology for determining whether a trial court’s resentencing of a defendant resulted in a more severe sentence under Pearce. In 1975, in Anthony v. Hopper, 235 Ga. 336, 337-338 (1) (219 SE2d 413) (1975), the Supreme Court of Georgia, in a unanimous decision, applied a “count-by-count” approach, which requires a court to compare the sentences on each count of an indictment separately.
In 2010, however, in Adams v. State, 287 Ga. 513, 517-520 (2) (696 SE2d 676) (2010), a plurality of the Supreme Court of Georgia criticized the count-by-count approach applied in Anthony and applied the “aggregate” approach instead. Under the aggregate approach, a court must “compare the total original sentence to the total sentence after resentencing. If the new sentence is greater than the original sentence, the new sentence is considered more severe.” (Citations and punctuation omitted.) Id. at 517 (2). The plurality in Adams noted that the aggregate approach has been adopted by “[t]he vast majority of federal and state appellate courts[.]” (Citations and punctuation omitted.) Id.
Here, the severity of Hudson’s new sentence, and whether the Pearce presumption of vindictiveness arises, depends on whether it is judged on a count-by-count basis or in the aggregate on all counts.1 Under the aggregate approach, Hudson’s new sentence was not more severe. That is, Hudson’s initial sentence on both counts would have resulted in a total of 25 years in prison and probation for the *56remainder of his life. After the conviction for aggravated sexual battery was merged into the conviction for child molestation, he was resentenced to a total of twenty-five years in prison and five years on probation.
Under the count-by-count approach, however, Hudson’s new sentence was more severe. Specifically, Hudson’s initial sentence on the child molestation count would have resulted in ten years in prison and twenty years on probation. Hudson was resentenced on the same count for a total of twenty-five years in prison and five years on probation. Accordingly, although the probationary period of Hudson’s sentence was less, his prison sentence was increased by 15 years.
Hudson contends that the aggregate approach in Adams is not controlling because the decision had the full concurrence of only three Justices. As such, he argues that Georgia still follows the count-by-count approach applied in Anthony. In view of the divergent rulings from the Supreme Court of Georgia, this Court, the trial court, and the State Bar are somewhat handicapped in determining which approach should be applied. Nevertheless, in the wake of only a plurality opinion in Adams, and in an abundance of caution, we conclude that Georgia still follows the count-by-count approach as applied in Anthony. Until and unless Anthony is overruled by a decision of the Supreme Court of Georgia, which is concurred in by a majority of the Justices, it remains controlling.2 Accordingly, when we apply the approach set forth in Anthony, Hudson’s new sentence was more severe than his original sentence on the child molestation count, giving rise to the Pearce presumption of vindictiveness.
Since the Pearce presumption of vindictiveness was triggered, the trial court was required to overcome it with affirmative reasons or facts to substantiate the new sentence. See Pearce, supra, 395 U. S. at 726 (II) (C).
Those reasons must be based upon objective information concerning identifiable conduct on the part of the defendant *57occurring after the time of the original sentencing proceeding. And the factual data upon which the increased sentence is based must be made part of the record, so that the constitutional legitimacy of the increased sentence may be fully reviewed on appeal.
(Emphasis supplied.) Id. In its written resentencing order, the trial court attempted to dispel the presumption of vindictiveness by stating that the same term of confinement as the original sentence was intended not only to deter Hudson from engaging in the same type of sexual misconduct upon release, but also primarily to punish Hudson for his offense. The trial court also cited the victim’s age, Hudson’s familial relationship with the victim, and Hudson’s prior child molestation offense as aggravating circumstances justifying the new sentence. However, Pearce requires the reasons for the new sentence to be based upon identifiable conduct occurring after the time of the original sentencing proceeding. See id. And here, the circumstances cited by the trial court are the same as those that previously existed at the first sentencing proceeding, and thus provided no basis for an increased sentence on the child molestation charge upon remand. Therefore, the case is remanded for reconsideration of Hudson’s sentence so that it conforms with the law established by Pearce and its progeny.
Judgment reversed and case remanded.
Ray and Branch, JJ., concur fully and specially.The State incorrectly cites Johnson v. State, 307 Ga. App. 499, 500-501 (705 SE2d 303) (2010), as having applied the aggregate approach rather than the count-by-count approach. Contrary to the State’s assertion, this Court in Johnson, supra, 307 Ga. App. at 500, was not required to make a decision as between the two approaches, because the defendant’s new sentence was not harsher either by count or in the aggregate. Therefore, regardless of the approach utilized, the Pearce presumption did not apply in Johnson.
See, e.g., Blake v. State, 272 Ga. App. 402, 403-404 (1), 406 (612 SE2d 589) (2005) (concluding that the binding precedent of Anthony controlled the methodology and outcome of the case and applying count-by-count approach) (physical precedent only); see also Thompson v. Thompson, 288 Ga. 4, 5 (1), n. 2 (700 SE2d 569) (2010) (“[Where Supreme Court of Georgia case] purported] to overrule the long-standing line of cases, it d[id] not achieve that goal because only a plurality, not a majority, of th[e] [Supreme] Court ha[d] determined that those cases must ‘give way.’ Thus, the bench and bar [were] apprised that [a prior majority opinion of the Supreme Court] and all of the other cases which applied] [that same legal analysis] . . . remained] controlling authority for the present and that the holdings in those cases [were to] be followed as accurate statements of the applicable law.”) (citation and punctuation omitted).