Knight v. State

LAWSON, J.

Jonathon Knight appeals from his conviction on a single charge of possession of cannabis (more than 20 grams), arguing that the trial court should have granted his motion for judgment of acquittal. Reviewing the matter de novo, see Pagan v. State, 830 So.2d 792, 803 (Fla.2002), we affirm. Although Knight’s argument finds support in opinions from other districts applying a “special” circumstantial evidence standard,1 we disagree with the way that standard has been used in those cases on similar facts. For this reason, we certify conflict with Evans v. State, 32 So.3d 188 (Fla. 1st DCA 2010); P.M.M. v. State, 884 So.2d 418 (Fla. 2d DCA 2004); N.K.W., Jr. v. State, 788 So.2d 1036 (Fla. 2d DCA 2001); E.H.A. v. State, 760 So.2d 1117 (Fla. 4th DCA 2000); S.B. v. State, 657 So.2d 1252 (Fla. 2d DCA 1995); and Cook v. State, 571 So.2d 530 (Fla. 1st DCA 1990). Additionally, although we certify conflict based upon application of the special circumstantial evidence standard, as we understand it, we also question the continued use of this standard for appellate review and suggest to the Florida Supreme Court that it reconsider the law in this area.

Facts

In the early morning hours of November 21, 2010, Knight was driving a yellow 2010 Camaro owned by a friend, who was riding in the front passenger seat of the vehicle, *452when Orange County Deputy Sheriff Donald Murphy pulled alongside the Camaro, activated his emergency lights, and signaled Knight to pull over. Knight complied, and stopped in an adjacent 7-Eleven convenience store parking lot. Deputy Murphy ordered Knight out of the car, explaining that he had stopped Knight because of excessively loud music emanating from the car, and walked him to the front of his patrol car.

Serendipitously, a K-9 officer pulled into the 7-Eleven parking lot within minutes of the stop, while Murphy was issuing a noise violation citation to Knight. After Murphy released Knight, Knight walked into the 7-Eleven to buy a drink. At approximately the same time, the K-9 officer made an “impromptu” decision to run his dog, Endo, around the Camaro. Endo alerted to the passenger side door, and Murphy re-detained Knight when he walked out of the 7-Eleven. Murphy then searched the vehicle, locating a small bag of suspected cannabis in a “small carry-on style rolling-type suitcase” which contained a luggage tag identifying Knight as the owner of the suitcase. The suitcase had been sitting on the backseat of the car. Murphy seized the substance, which ultimately tested positive as cannabis and weighed 24.4 grams. Deputy Murphy did not locate any drug paraphernalia typically associated with marijuana usage in the vehicle. After completing the vehicle search, Murphy arrested Knight for possession of cannabis. In a search incident to the arrest, Murphy discovered $2,400 cash in Knight’s pockets.2

The State charged Knight with both possession of cannabis with intent to sell or deliver and possession of more than 20 grams of cannabis. At trial, the State presented the testimony summarized above, and rested. Knight moved for a judgment of acquittal, which the trial court denied.

Knight called as his first witness Chaka Miller, the friend who had been in the front passenger seat at the time of the stop which led to Knight’s arrest. Miller testified that he, Knight and another friend (Chad Harris) were visiting Orlando for the weekend to attend the “Florida Classic” football game on the date of Knight’s arrest. He testified that the cannabis found in the car did not belong to him, but that he had not seen Knight with marijuana — or heard him discuss marijuana — at all during the trip. He testified that the group usually paid cash for their hotel rooms when they traveled. Finally, although he did not contradict the State’s evidence that the suitcase belonged to Knight, he did testify that Chad Harris was left in the backseat of the car next to the suitcase after Deputy Murphy removed Knight from the vehicle to issue the citation — implying that Harris could have placed the cannabis in the suitcase at that time.

Knight then took the stand in his own defense, also testifying that the cannabis was not his. He claimed that the money on his person was for his weekend trip expenses, and did not come from selling drugs.3 He flatly denied selling drugs (“I don’t sell drugs.”), and further elaborated that he had that much cash to spend for the weekend because he had received settlements from two separate personal inju*453ry cases, one involving a motorcycle accident and another involving a fight. He digressed to describing his injuries in detail: “I had two brain surgeries, four plates in my head, and I have a 50 percent use of my right shoulder....” Finally, he reiterated that Chad Harris was seated in the back of the car next to the suitcase, and was left there between the time he was removed from the car and the search several minutes later. Knight also did not contradict the State’s evidence that the suitcase belonged to him, and seemed to acknowledge it as his — instead focusing on the fact that Harris had an opportunity to place the marijuana in the suitcase and denying that he had put any marijuana in his luggage.4 On cross-examination the State immediately sought to have Knight reiterate what it obviously viewed as Knight’s admission that the suitcase was his, but Knight then denied owning the suitcase and also denied that his name was on the luggage tag — directly contradicting Deputy Murphy’s testimony.

At the close of the evidence, Knight renewed his motion for judgment of acquittal, which was denied. After deliberations, the jury returned a verdict of not guilty on the charge of possession with intent to sell or deliver and a verdict of guilty on the charge of possession of more than 20 grams of cannabis.

Analysis

A. Possession Cases Generally and Cases Supporting Reversal.

“Proof of possession of a controlled substance may be actual or con-struetive.” Taylor v. State, 13 So.3d 77, 80 (Fla. 1st DCA 2009). The State acknowledges that Knight was not in actual possession of the cannabis at the time in question, and relies on a theory of constructive possession. In order to prove constructive possession, the State must prove that the accused had dominion and control over the contraband and that he had knowledge that the contraband was in his presence. J.J.N. v. State, 877 So.2d 806, 809 (Fla. 5th DCA 2004).5 When the place where the contraband is found is not in the exclusive possession of a defendant, knowledge of the presence of the contraband on the premises and the accused’s ability to maintain control over it will not be inferred, but must be established by independent proof. Id. at 809-10. “Mere proximity to contraband is not sufficient to establish constructive possession.” Id.

In this case, Knight does not challenge the State’s evidence as to “dominion and control.” The State presented evidence that the cannabis was found in a suitcase that belonged to Knight, in the passenger compartment of the car he was driving. This evidence was clearly sufficient to establish Knight’s dominion and control over the cannabis. Instead, Knight focuses on the knowledge element, arguing that because Chad Harris had unsupervised access to the luggage after Deputy Murphy removed Knight from the car, Harris could have slipped the marijuana into the luggage without Knight’s knowledge shortly *454before the vehicle search. This is the same argument that Knight made at trial in support of his motion for judgment of acquittal.

With respect to this argument, we believe Knight’s case to be indistinguishable from the Second District’s decision in N.K.W. In that case, a juvenile, N.K.W., had been adjudicated delinquent for possession of LSD. Police found the LSD in N.K.W.’s wallet on a shelf during execution of a search warrant at a residence. N.K.W. was inside the residence attending a party when police found the LSD in his wallet. N.K.W., 788 So.2d at 1037. Although the juvenile admitted that the wallet was his, he denied knowledge of the LSD — hypothesizing that because the party was hosted in part to celebrate his upcoming birthday, one of his friends could have placed the LSD in his wallet without his knowledge, as a surprise birthday gift. Citing to Cook, the Second District held that reversal was required because the state failed to present any “direct evidence” that was inconsistent with N.K.W.’s “reasonable hypothesis of innocence” that someone else placed the LSD in his wallet. Id. at 1038.

In Cook, the defendant was arrested and charged with possession of a crack pipe and its residue that had been found in her purse, during the raid of a bar where she worked as a dancer. The defendant testified that she had left her purse on the bar — where others had access to it — during her dance routines, and hypothesized that someone else could have placed the crack pipe there without her knowledge. Because the evidence “suggesting that [the defendant] knew of the presence of the crack pipe within her purse was entirely circumstantial[,]” the First District held that the state was required to produce evidence “inconsistent with the defendant’s reasonable hypothesis of innocence[.]” Cook, 571 So.2d at 531-32. Because the state presented no evidence inconsistent with the hypothesis that someone else could have place the contraband in the defendant’s purse, the court held that the motion for judgment of acquittal should have been granted, and reversed the conviction. Id. at 532.

These two cases are consistent with other cases where the state has attempted to prove the knowledge element(s) in a constructive possession case based upon evidence that the defendant owned the container in which the contraband was located. See Evans, 32 So.3d at 188; 884 So.2d at 418; E.H.A., 760 So.2d at 1117; S.B., 657 So.2d at 1252.6 In each of these similar cases, the courts applied the “special standard of review” applicable to convictions “wholly based on circumstantial evidence,” State v. Law, 559 So.2d 187, 188 (Fla.1989), and reversed convictions based upon a finding that the state presented no evidence inconsistent with the defendant’s reasonable hypothesis of innocence.

Were we to apply this standard here in the same fashion as the above cases, we would also reverse the conviction. As Knight argues, Chad Harris would have had an opportunity to place the cannabis in his suitcase after he had been removed *455from the car, and also had a plausible motive to do so.7 In our view, however, the “special standard” is inappropriate for this case because it is not a “wholly circumstantial case.” Additionally, even if this standard were applied here, we believe that a reversal would improperly invade the province of the jury. Given the evidence presented at trial, whether Knight’s hypothesis of innocence is “reasonable” should be a question for the jury. See, e.g., Beasley v. State, 774 So.2d 649, 659 (Fla.2000) (“ ‘[T]he question of whether the evidence fails to exclude all reasonable hypotheses of innocence is for the jury to determine.’”) (quoting State v. Law, 559 So.2d 187, 188 (Fla.1989)).

We believe it helpful to a complete understanding of the issues in this case to start with a review of the development of Florida’s unique approach to circumstantial evidence cases in the criminal law context. As part of this discussion, we will explain why the Florida Supreme Court may want to reconsider the law in this area altogether. Then, we will address why we believe the special circumstantial evidence standard should not apply in this case under existing precedent. Finally, we will further explain why the reasonableness of Knight’s “hypothesis of innocence” should not be decided as a matter of law under existing precedent.

B. Development of a Unique Standard of Review for Circumstantial Evidence Cases and Why The Standard Should Be Reconsidered.

Courts over the centuries have wrestled with the relative reliability of circumstantial versus direct evidence. See Julie Schmidt Chauvin, “For It Must Seem Their Guilt:” Diluting Reasonable Doubt by Rejecting the Reasonable Hypothesis of Innocence Standard, 53 Loy. L.Rev. 217, 223 (2007). For example, the “seventeenth-century English courts allowed for the use of circumstantial evidence only when necessary, viewing it with great caution,” while “[eighteenth century jurisprudence reflected a reverse in the value attributed to direct and circumstantial evidence,” culminating in a general view that circumstantial evidence “may have greater probative value than direct evidence, for people may lie, but circumstances cannot.” Id. at 223-24 (footnotes omitted). In 1850, taking the more cautious approach to circumstantial evidence, the Massachusetts Supreme Court announced special rules for a jury’s use in evaluating circumstantial evidence, including an instruction that “the circumstances ... should to a moral certainty exclude every other hypothesis” except that of guilt. Id. at 319 (citing Commonwealth v. Webster, 59 Mass. (5 Cush.) 295 (Mass.1850)). *456“Known as the ‘Webster Charge,’ this rule quickly spread throughout mid-nineteenth century United States courts and was the prevailing rule regarding circumstantial evidence until the United States Supreme Court rejected it in the 1954 case of Holland v. United States [, 348 U.S. 121, 75 S.Ct. 127, 99 L.Ed. 150 (1954) ].” Chauvin, supra, at 224 (footnotes omitted).

The appellants in Holland challenged their criminal convictions on tax evasion charges. One of the many issues raised before the United States Supreme Court was whether the trial court had erred in declining to instruct the jury using a Webster-type charge “that where the Government’s evidence is circumstantial it must be such as to exclude every reasonable hypothesis other than that of guilt.” Holland, 348 U.S. at 139, 75 S.Ct. 127. The Court not only rejected the argument, but expressed its view that “where the jury is properly instructed on the standards for reasonable doubt, such an additional instruction on circumstantial evidence is confusing and incorrect[.]” Id. at 139-40, 75 S.Ct. 127 (citations omitted). The Court further explained that:

Circumstantial evidence in this respect is intrinsically no different from testimonial evidence. Admittedly, circumstantial evidence may in some cases point to a wholly incorrect result. Yet this is equally true of testimonial evidence. In both instances, a jury is asked to weigh the chances that the evidence correctly points to guilt against the possibility of inaccuracy or ambiguous inference. In both, the jury must use its experience with people and events in weighing the probabilities. If the jury is convinced beyond a reasonable doubt, we can require no more.

Id. at 140, 75 S.Ct. 127.

“By 1982, all federal courts had adopted Holland’s rejection of the reasonable hypothesis of innocence standard for use in criminal cases based entirely on circumstantial evidence, as a matter of federal law.” Chauvin, supra, at 227 (footnote omitted). As a result, the federal courts no longer instruct juries using a Webster-type instruction, and “[a]ll eleven circuits ... hav[e] rejected the reasonable hypothesis of innocence as a standard of appellate review.” Id. at 248 (footnote omitted). Following Holland, most state courts also rejected both the Webster instruction for use with juries and as a “special standard” for evaluating verdicts based upon circumstantial evidence on appeal. Id. at 254-55; see also 1 Wharton’s Criminal Evidence § 4:5 (15th ed.2011); Irene Merker Rosenberg & Yale L. Rosenberg, “Perhaps What Ye Say Is Based Only On Conjecture”-— Circumstantial Evidence, Then and Now, 31 Hous. L.Rev. 1371, 1400-01 (1995); Ca-roll J. Miller, Annotation, Modem Status of Rule Regarding Necessity of Instmction on Circumstantial Evidence in Criminal Trial — State Cases, 36 A.L.R.4th 1046 (2012).8

Florida eliminated its Webster-type instruction in 1981. In the Matter of Use by Trial Courts of Standard Jury Instructions in Criminal Cases, 431 So.2d 594 (Fla.1981). Then, however, our supreme court retained (without any analysis) a Webster-type appellate standard of review for evaluating sufficiency of the evidence challenges in cases “where a conviction is wholly based on circumstantial evidence.” Jaramillo v. State, 417 So.2d 257 (Fla.1982).9 Florida is one of only three states *457to have taken the “somewhat discordant” position that a Webster instruction should not be used to instruct the jury but should be used to judge the jury’s verdict. Rosenberg, supra, at 1419.10 Expressly rejecting this approach, the Texas Court of Criminal Appeals explained:

As we have emphasized on numerous occasions, the sufficiency of the evidence must be measured against the jury charge. Given the fact that a jury is to be guided by the charge in reaching their verdict, and given the fact that juries are no longer instructed on the law of circumstantial evidence, it no longer makes sense for appellate courts to use the circumstantial evidence “construct” to review the jury’s verdict and to determine, thereby, whether the jurors acted “rationally.” To do so evaluates the jurors’ rationality by a different standard than that by which they were instructed to reach their verdict.

Geesa v. State, 820 S.W.2d 154, 161 (Tex.Crim.App.1991) (internal citations omitted), overruled in part on other grounds by Paulson v. State, 28 S.W.3d 570 (Tex.Crim.App.2000).

This criticism makes sense to us. If, on the one hand, “ ‘[t]he question of whether the evidence fails to exclude all reasonable hypotheses of innocence is for the jury to determine,’” Beasley, 774 So.2d at 659 (quoting Law, 559 So.2d 187 at 188), why not tell the jury that this is the decision it must make in reaching its verdict (in a circumstantial evidence case)? Conversely, if this framework for analysis really is “confusing and incorrect,” In the Matter of Use by Trial Courts of Standard Jury Instructions in Criminal Cases, 431 So.2d at 595 (quoting Holland, 348 U.S. at 140, 75 S.Ct. 127), why would we use it in appellate review — especially when the jury is not first told that this is the construct it should use in evaluating the evidence? In our view, Florida should join the federal courts and the vast majority of states that have abandoned use of a special circumstantial evidence standard of review, for several reasons.

First, the special standard is misleading. The “special standard of review of the sufficiency of the evidence” which “applies where a conviction is wholly based on circumstantial evidence” is most often articulated by Florida’s appellate courts as follows: “Where the only proof of guilt is circumstantial, no matter how strongly the evidence may suggest guilt, a conviction cannot be sustained unless the evidence is inconsistent with any reasonable hypothesis of innocence.” Law, 559 So.2d at 188 (citations omitted). The standard is misleading because it completely ignores the proper role of the jury and directs the appellate judge to determine, ab initio, from argument or a cold record, whether a hypothesis of innocence is reasonable. As explained by one federal appellate court:

It is ... [often said] that unless the evidence excludes the hypothesis of innocence, the judge must direct a verdict [or] ... that if the evidence is such that *458a reasonable mind might fairly conclude either innocence or guilt, a verdict of guilt must be reversed on appeal. But obviously neither of those translations is the law. Logically, the ultimate premise of that thesis is that if a reasonable mind might have a reasonable doubt, there is, therefore, a reasonable doubt. That is not true.[I]f a reasonable mind might fairly have a reasonable doubt or might fairly not have one, the case is for the jury, and the decision is for the jurors to make.... If [the judge] concludes that either of the two results, a reasonable doubt or no reasonable doubt, is fairly possible, he must let the jury decide the matter.

Curley v. United States, 160 F.2d 229, 232-33 (D.C.Cir.), cert. denied, 331 U.S. 837, 67 S.Ct. 1511 (1947). This portion of the Curley opinion would later be cited with approval by the United States Supreme Court in its landmark decision setting the standard of review that we still follow today in criminal cases. Jackson v. Virginia, 443 U.S. 307, 319, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979) (holding that “the relevant question [in a sufficiency of the evidence review for criminal convictions] is whether, after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt”).

Not surprisingly, by recognizing that “whether the evidence fails to exclude all reasonable hypotheses of innocence is for the jury to determine,” Law, 559 So.2d at 188, while at the same time adopting an appellate standard of review that suggests otherwise, Florida’s appellate courts have spawned two lines of subtly conflicting cases. This dichotomy was recognized by our former colleague, Judge Cowart, in a well-researched concurring opinion listing dozens of conflicting opinions from Florida’s appellate courts. See Dunn v. State, 454 So.2d 641 (Fla. 5th DCA 1984) (Co-wart, J., concurring specially). As Judge Cowart explained:

Defendant cites many eases in which appellate courts have encroached upon the province of the jury by determining as a matter of law either (a) that the jury could not believe or could not disbelieve the testimony of some particular witness, or (b) that some suggested possibility (hypothesis) of innocence sounded so reasonable to the appellate court that, as a matter of law, the jury, after weighing all the evidence and all possible inferences, was not allowed to disregard what the jury considered to be a slight or weak or remote possibility of innocence (which nearly always rests upon the credibility of the defendant’s testimony) and to accept what the jury considered to be a strong or satisfying or overwhelming probability of guilt. The viability of the cases cited by defendant ... is questionable in view of better reasoned cases which recognize the jury question inherently involved.

Id. at 644 (footnotes omitted).

The special standard of review is also misleading (and confusing) in that the “no matter how strongly the evidence may suggest guilt” language ignores the correlation between the “strength” of circumstantial evidence and the “reasonableness” of various hypotheses of innocence. In other words, the stronger the circumstantial evidence, the more likely that a rational jury will be justified in rejecting explanations other than the guilt of the accused as unreasonable. For example, consider a murder case in which neighbors witness a couple return to their two-story home while in a heated discussion at 6:00 p.m. one evening. The husband calls police to report that he has found his wife strangled to death in their upstairs bedroom at 1:00 *459a.m. He claims to have neither seen nor heard anything to explain the crime, and posits that some unknown intruder must have come and gone undetected and committed the murder while he was in the downstairs living room watching television. With testimony from a medical examiner that the mortal injury could not have been self-inflicted, and that death occurred between 6:00 p.m. and 9:00 p.m., the most reasonable inference to be drawn from this entirely circumstantial evidence is that the husband — as the only other known (and likely) occupant of the house at the time of the crime — committed the murder.

To these basic facts, we can add entirely circumstantial evidence that, ultimately, would convince the most ardent proponent of Florida’s special standard that the jury would be justified in returning a verdict of guilty. First, we could add evidence of a strong financial motive for the husband to commit the murder, plus perhaps testimony about extreme discord in the marriage that had been escalating. Next, outdoor video surveillance recordings covering most (but not all) of the home’s possible entrances and exits show the housekeeper leaving at 5:00 p.m. on the evening of the murder, and the housekeeper confirms this information, adding that she neither saw nor heard anyone else in the home at any time on the day of the murder. The recordings show the couple returning home at 6:00 p.m., and no other person around the house that night. Later, friends of the husband admit to police that the husband has repeatedly told them that he would “strangle his wife” if she ever tried to leave him. Police also discover that the wife had met with a divorce attorney late in the afternoon on the day of the murder. Given the strength of this hypothetical circumstantial evidence, we doubt that anyone would suggest that a jury could not reject the husband’s hypothesis that an unknown assailant committed the crime. However, this conclusion seems inconsistent with the “no matter how strongly the evidence may suggest guilt” language in the standard, because the standard reads as if the “strength” of the evidence cannot be considered at all in the sufficiency review of a circumstantial evidence case. Clearly, this is incorrect.

The special standard of review is also misleading because it incorrectly suggests that an allowance for the jury’s credibility determinations should never factor into a sufficiency of the evidence review in circumstantial evidence cases. Taking this case as an example, the jury was in the unique position to evaluate the credibility of the witness who appears to have been called to help lay the foundation for Knight’s argued hypothesis of innocence by testifying that Chad Harris was in the backseat with the suitcase after Knight was removed from the vehicle. The jury may have noticed — and factored into its overall evaluation of the evidence — that this witness was also in a position to know whether or not Harris had placed the drugs into Knight’s suitcase, but avoided that topic altogether. Although no one would question the jury’s right to judge the credibility of this witness, or to take this determination into account when reaching its ultimate verdict, the special standard would treat this and other credibility determinations as irrelevant for purposes of review in a circumstantial evidence case.

The special standard should also be rejected because it is unhelpful as an analytical tool for the reasons addressed above. The standard is just not worded in a way that matches the necessary analysis. We strongly suspect that what appellate judges really do in circumstantial sufficiency of the evidence cases (as in all sufficiency of the evidence cases) is look at the totality of the evidence presented and, *460“giving full play to the right of the jury to determine credibility, weigh the evidence, and draw justifiable inferences of fact,” ask whether “a reasonable mind might fairly conclude guilt beyond a reasonable doubt.” Curley, 160 F.2d at 232; see also Pagan v. State, 830 So.2d 792, 803 (Fla.2002) (“If after viewing the evidence in the light most favorable to the State, a rational trier of fact could find the existence of the elements of the crime beyond a reasonable doubt, sufficient evidence exists to sustain a conviction.”). If, after a careful review and evaluation of the evidence using this standard, the judge determines that a rational fact-finder could not fairly conclude guilt beyond a reasonable doubt, the judge then shifts back to the language of the mandated standard to announce that the state did not present evidence inconsistent with a reasonable hypothesis of innocence. Conversely, if the judge determines that a rational fact-finder could fairly conclude guilt beyond a reasonable doubt considering all of the evidence, the judge is forced to return to the mandated standard to announce that the defendant’s hypothesis of innocence is either not reasonable or a question for the jury. See, e.g., Huck v. State, 881 So.2d 1137, 1147 (Fla. 5th DCA 2004) (rejecting the argued hypotheses of innocence variously as “not particularly reasonable[,]” “hardly plausible[,]” or “speculative and unreasonable”); Henderson v. State, 679 So.2d 805 (Fla. 3d DCA 1996) (emphasizing “that the State was not required to exclude any unreasonable hypothesis” and concluding that the argued hypothesis “in light of the evidence, created a legitimate question for the jury to determine”) (emphasis in original). Either way, the special standard does not aid the appellate analysis. Although we use the words, as necessary, it is not the way we do (or should) think about the issue. We suspect this is why “even a majority of the minority of jurisdictions that still require a cautionary instruction with respect to circumstantial evidence do not use an analogous sufficiency of the evidence test on appellate review.” Rosenberg, supra, at 1418 (footnote omitted).

Finally, we believe that the special standard should be rejected because it is unnecessary. Proponents of the special standard argue that the standard is needed to ensure “that the inferences made by the jury when evaluating circumstantial evidence were not based on mere speculation but were in fact reasonable enough to find guilt beyond a reasonable doubt.” Chau-vin, supra, at 245. We disagree. If there is no direct evidence relating to an element of a crime and the inferences that can reasonably be drawn from circumstantial evidence are insufficient for a trier of fact to find that element beyond a reasonable doubt, a conviction clearly cannot be sustained under the general standard of review because no “rational trier of fact could find the existence of the elements of the crime beyond a reasonable doubt” under those circumstances. Pagan, 830 So.2d at 803. A reviewing court considering all of the evidence relating to each element of the crime necessarily has to carefully analyze what “justifiable inferences of fact” a rational fact-finder could draw from the evidence under the standard of review that applies to all cases. Curley, 160 F.2d at 232. For this reason, the “special standard” adds nothing — except confusion — to the analysis.

This is also true because the circumstantial evidence standard is not so much a unique articulation drawn to address the special nature of circumstantial evidence (the other justification for use of the special standard put forth by its propo*461nents),11 as it is an alternative way of arguing the issue that could be applied equally to a direct evidence case. For example, consider a case in which a man named Abe is one of three neighbors who had grown to hate one another after years of well-known conflicts. Abe is found dead in his front yard, having been shot by a single bullet fired from some unknown distance. No murder weapon is discovered, but the second neighbor, Ben, ultimately comes forward to testify that the third neighbor, Charlie, confided in him that he was going to kill Abe, and that he (Ben) was near a window in his home when he heard the shot and quickly looked out to see Charlie lowering the gun he had just fired and heading to his car to leave. He also testifies that Charlie admitted to intentionally killing Abe in a conversation following the murder. Although Ben is impeached as to bias, motive to lie, opportunity to have committed the crime himself, and his ability to see what he claims to have seen on a dark night through a distant window, he sticks to his story. Charlie testifies on his own behalf that on the night of the murder he left his home (and the area) at least one hour before the state’s evidence indicates that Abe was killed. He denies owning a gun or killing Abe, and denies talking to Ben about anything for months. He also testifies that he has seen Ben with guns in the past, and that Ben has bragged to him about his marksmanship skills. He posits that it is just as reasonable to believe that Ben committed the crime and is now lying to both spare himself and frame his other sworn enemy.

The jury would be told that they should return a verdict of guilty if they are convinced beyond a reasonable doubt that Charlie unlawfully killed Abe from a premeditated design to effect Abe’s death. Additionally, however, in this direct evidence case, the jury could also be told that they should not return a verdict of guilty unless they are convinced that the evidence is inconsistent with any reasonable hypothesis of innocence. If this additional instruction were given, the argument might shift from whether Charlie’s testimony along with the impeachment topics covered in Ben’s cross should create a reasonable doubt in the minds of jurors to whether that same evidence is sufficient to convince the jury of the reasonableness of Charlie’s hypothesis of innocence. In either case, the jury must weigh the strength of the evidence suggesting guilt (Ben’s testimony) against the strength of the contrary evidence (and argument). Again, in our view, shifting to a discussion of the reasonableness of a defendant’s hypothesis of innocence adds confusion, for the reasons already explained (and recognized by all federal courts and the vast majority of state courts in this country), and nothing else. It is simply a different way to view and articulate the fact-finder’s task in a criminal case — whether direct or circumstantial.

Finally, we note that Florida provides an extra measure of protection against erroneous convictions in both direct and circumstantial evidence cases by allowing the trial judge to order a new trial upon finding that the “verdict is contrary to law or the weight of the evidence.” Fla. R.Crim. P. 3.600(a)(2) (emphasis added). Thus, not only is a defendant entitled to a detached and neutral review of the sufficiency of the *462evidence, both by the trial court and on appeal, in Florida we authorize a trial judge to “weigh the evidence and determine credibility just as a juror is required to do.” Geibel v. State, 817 So.2d 1042, 1044 (Fla. 2d DCA 2002). By this standard, Florida has provided a real and meaningful additional protection against erroneous convictions by allowing the trial judge to set aside a verdict that is contrary to the weight of the evidence, even where it meets the constitutionally-mandated sufficiency of the evidence test.

C. Why the Special Circumstantial Evidence Standard Should Not Apply in this Case Even if it is Retained as an Appellate Standard of Review in Florida.

As we have already discussed, the Florida Supreme Court mandates that Florida’s appellate courts use a “special standard of review of the sufficiency of the evidence ... where a conviction is wholly based on circumstantial evidence.” Jaramillo, 417 So.2d at 257 (emphasis added). Under this standard “ ‘[wjhere the only proof of guilt is circumstantial, no matter how strongly the evidence may suggest guilt a conviction cannot be sustained unless the evidence is inconsistent with any reasonable hypothesis of innocence.’ ” Id. (quoting McArthur v. State, 351 So.2d 972, 976 n. 12 (Fla.1977) (emphasis added). However, where “the State presents both direct and circumstantial evidence, courts do not apply the special standard of review applicable to circumstantial evidence cases.” Mosley v. State, 46 So.3d 510, 526 (Fla.2009) (citing Pagan v. State, 830 So.2d 792, 803 (Fla.2002)) (emphasis added). As we (and other Florida district court judges) understand this binding precedent, the “special standard” is intended to apply only when “the evidence for each element of each offense” is wholly circumstantial. Helms v. State, 38 So.3d 182, 185 n. 1 (Fla. 1st DCA 2010) (citations omitted); see also Bussell v. State, 66 So.3d 1059-1062 (Fla. 1st DCA 2011) (rejecting use of special standard when reviewing sufficiency of the evidence to support conviction for possession of child pornography because “[tjhe only element established through circumstantial evidence was possession”); Grant v. State, 13 So.3d 163, 167 (Fla. 1st DCA 2009) (explaining that special standard only applies to cases in which all of the evidence is circumstantial and expressly rejecting argument that the standard applies when the evidence relating to one or more elements is wholly circumstantial); Jenkins v. State, 1 So.3d 317, 320 (Fla. 3d DCA 2009) (“[Wjhere one or more of the elements of the crime are proven by direct evidence, this heightened standard of review is not applicable.”); State v. Burrows, 940 So.2d 1259, 1262 (Fla. 1st DCA 2006) (“When the State presents both direct and circumstantial evidence, the special standard of review applicable to circumstantial evidence cases does not apply.”); but see, e.g., Alleyne v. State, 42 So.3d 948, 950 (Fla. 4th DCA 2010) (“Where proof of an element of a crime is based wholly on circumstantial evidence, [the] special standard of review applies.”) (emphasis added).

The problem with applying the special standard of review in this and similar possession cases is that the convictions are not wholly based upon circumstantial evidence. In N.K.W., for example, the state proved the defendant’s dominion and control over the drug LSD with direct evidence that the substance was found near him and in his wallet, and that the substance was in fact LSD. Similarly, in Cook the state established the defendant’s dominion and control over the crack pipe with direct evidence that it was found near her person, in her purse, and that the residue in the pipe was cocaine. In these two cases the defendant’s own admissions *463provided direct evidence sufficient to prove dominion and control. Finally, in our case, Deputy Murphy’s testimony provided direct evidence — based upon his personal observations — establishing that the cannabis was within Knight’s dominion and control.

In all of these cases, the state is admittedly relying on circumstantial evidence to prove the knowledge element. But, of course, state of mind elements such as knowledge, intent or premeditation are usually established through circumstantial evidence. See, e.g., Anderson v. State, 48 So.3d 1015, 1018 (Fla. 5th DCA 2010) (noting that “[i]n a criminal prosecution, knowledge is rarely proven by direct evidence”) (citations omitted); Jones v. State, 192 So.2d 285, 286 (Fla. 3d DCA 1966) (“Intent, being a state of mind, is often not subject to direct proof and can only be inferred from circumstances.”). In our view, it would be a grave mistake to expand use of the “special standard” to include the many cases where the state presents direct evidence to support a finding of guilt as to most elements, but is left to rely on circumstantial evidence to prove the defendant’s state of mind.

For example, consider a case in which a supposed battery victim identifies the defendant in court as the jealous ex-boyfriend of his current girlfriend and convincingly testifies that on the date alleged in the information he and the defendant “locked eyes” as they were walking on the same public sidewalk; that the defendant immediately veered toward him and — looking directly at him with an expression of malice on his face — swung his left elbow out in an unnatural manner and struck him with such an unusually strong force that he was propelled backwards onto the ground, causing injury to the back of his head. A neutral third-party witness who stopped to render assistance after witnessing the incident also testifies convincingly to the same basic facts. Testifying in his own defense, the accused acknowledges a past relationship with the alleged victim’s girlfriend but claims no animosity toward either and testifies that on the day in question he was deep in thought, grimacing in pain from a recent back strain as he walked, which also caused a strange gait in which he occasionally had to labor with his upper body to swing his left leg forward. He testified that he did not notice the victim until after they had accidentally bumped into one another as he was laboring to swing his left leg forward.

At the close of the evidence, and after argument, the jury will be instructed that it should return a verdict of guilty to the charge of battery if the state has proven beyond a reasonable doubt that the defendant intentionally touched or struck the victim against his will. If the jury finds the victim and third-party witness to be wholly credible — and judges the defendant to be a shifty, deceptive, evasive, bald-faced liar — which is unquestionably a judgment squarely within its proper function in this case which relies almost exclusively on direct evidence12 — the jury clearly should be able to return a guilty verdict. But, juxtapose these facts against the “special” standard that arguably applies since the evidence of intent is circumstantial: “Where the only proof of guilt is circumstantial, no matter how strongly the evidence may suggest guilt, a conviction cannot be sustained unless the evidence is inconsistent with any reasonable hypothesis of innocence.” Law, 559 So.2d at 188 *464(citations omitted). The unique problem in this type of case is that a person can always claim to have had a different state of mind than his or her actions suggest. And, when you remove the role of the fact-finder (in the courtroom) and view the explanation solely as a logic exercise on a cold appellate record, the explanation will almost always appear plausible. This is true because we know that thoughts are not always clear-cut or rational. As such, it would be a reasonable hypothesis that almost anyone at almost any time could have a state of mind different than the state of mind which a rational fact-finder would attribute to the person based upon observed facts.13 For this reason, a standard that does not allow a conviction based upon circumstantial evidence of a defendant’s state of mind, “no matter how strongly the evidence may suggest” the required state of mind, is unworkable and is obviously contrary to the law. See, e.g., State v. Tovar, — So.3d - (Fla. 2d DCA 2012). As explained by the Second District in Tovar:

“[A] trial court should rarely, if ever, grant a motion for judgment of acquittal based on the state’s failure to prove mental intent.” Hardwick v. State, 630 So.2d 1212, 1214 (Fla. 5th DCA 1994) (quoting Brewer v. State, 413 So.2d 1217, 1220 (Fla. 5th DCA 1982)). “Whether one had intent is generally a question given to a jury, for reasonable men may differ in determining intent when taking into consideration the surrounding circumstances.” State v. Herron, 70 So.3d 705, 706 (Fla. 4th DCA 2011).

Id.; see also Salter v. State, 77 So.3d 760, 763 (Fla. 4th DCA 2011) (explaining that intent “ ‘may be proven by a combination of surrounding circumstances from which a jury can reasonably infer a defendant’s guilt,’ ” and reiterating that “intent to participate in a crime is a question for the jury and a trial court properly denies a motion for judgment of acquittal where an issue remains for the jury to decide”) (quoting Parker v. State, 795 So.2d 1096, 1099 (Fla. 4th DCA 2001) (additional citations omitted)); Guinan v. State, 65 So.3d 589, 593 (Fla. 4th DCA 2011) (explaining that “direct evidence of intent is rare and must be proven through surrounding circumstances”; that to support a conviction the state’s circumstantial evidence of intent need only be “inconsistent with innocence” and “need not conclusively rebut every possible variation of events which can be inferred”; and that “[significantly, the absence of direct proof on the question of the defendant’s mental intent should rarely, if ever, result in a judgment of acquittal.”) (quotations and citations omitted); State v. Clyatt, 976 So.2d 1182, 1184 (Fla. 5th DCA 2008) (holding that lack of a victim’s consent to a battery can be proven by circumstantial evidence and that “[wjhether the State’s circumstantial evidence will be sufficiently convincing to establish [the victim’s] lack of consent beyond a reasonable doubt is a question for the jury”); Grover v. State, 581 So.2d 1379, 1380 (Fla. 4th DCA 1991) (“It is black-letter of course that intent, being a state of mind, is rarely if ever susceptible of direct proof. Almost inevitably, as here, it must be shown solely by circumstantial evidence.”) (citation omitted); Firestone v. State, 407 So.2d 1070, 1071 (Fla. 4th DCA 1981) (holding that circumstantial evidence *465was sufficient to support a jury finding that defendant entered a home without the victim’s consent, thereby upholding a burglary conviction). In recognizing that a defendant’s or victim’s state of mind is both capable of proof through circumstantial evidence and still a question for the jury in most cases, these and the many other decisions addressing the issue have not applied the “special” standard reserved for a “wholly circumstantial” case. See, e.g., Evans v. State, 997 So.2d 1281 (Fla. 4th DCA 2009) (finding that circumstances were insufficient for jury to find “intent to tamper or conceal” evidence beyond a reasonable doubt without using “special” standard of review); Arnold, v. State, 892 So.2d 1172 (Fla. 5th DCA 2005) (finding circumstances sufficient for jury to find premeditation without reference to the “special” standard of review).

These district court cases are completely consistent with cases from the Florida Supreme Court upholding murder convictions without reference to the “special standard,” even where only circumstantial evidence supported a jury’s finding of intent or some other state of mind element. See, e.g., Bright v. State, 90 So.3d 249 (Fla.2012) (applying normal standard of review to uphold conviction for first degree premeditated murder charge in death penalty case where intent and premeditation were proven by circumstantial evidence); Withers v. State, 104 So.2d 725 (Fla.1958) (applying normal standard of review to uphold murder conviction where premeditation was proven by circumstantial evidence). In many cases, however, the supreme court uses a hybrid analysis to address circumstantial evidence relating to premeditation.14 This approach starts with an explanation that “ ‘[circumstantial evidence of premeditation can include the nature of the weapon used, the presence or absence of adequate provocation, previous difficulties between the parties, the manner in which the homicide was committed, and the nature and manner of the wounds inflicted[ ],’ ” McMillian v. State, 94 So.3d 572, 580 (Fla.2012) (quoting Pearce v. State, 880 So.2d 561, 572 (Fla.2004) (other citations omitted)), along with a declaration that “ ‘Premeditation is a factual issue for the jury[.]’ ” Id. (quoting Asay v. State, 580 So.2d 610, 612 (Fla.1991)). Then, the court explains that: “[I]n a case where the evidence of premeditation is entirely circumstantial, not only must the evidence be sufficient to support the finding of premeditation, but the evidence, when viewed in the light most favorable to the State, must also be inconsistent with any other reasonable inference.” Id. (quoting Twilegar v. State, 42 So.3d 177, 190 (Fla.2010), cert. denied, — U.S. -, 131 S.Ct. 1476, 179 L.Ed.2d 315 (2011)). This standard, as applied by the Supreme Court, appropriately recognizes that circumstantial evidence pointing to premeditation, standing alone, is often “strong enough” that a reasonable fact-finder could reject other inferences that logically could be drawn from the evidence and thereby find premeditation beyond a reasonable doubt. This, of course, is very different from the articulated standard that “no *466matter how strongly [circumstantial] evidence may suggest guilt, a conviction cannot be sustained unless the evidence is inconsistent with any reasonable hypothesis of innocence.” Law, 559 So.2d at 188 (citations omitted).

When it comes to convictions involving a state of mind element, we also find State v. Bamum, 921 So.2d 513 (Fla.2005), cert, denied, Barnum v. Florida, 549 U.S. 993, 127 S.Ct. 493, 166 L.Ed.2d 365 (2006), to be particularly instructive. Barnum was convicted of armed robbery, attempted first-degree murder of a law enforcement officer, and several lesser charges. Focusing on the attempted first degree murder charge, Justice Lewis, writing for the majority of the court, noted only that “knowledge that the victim was a law enforcement officer” was an element of the crime and “a disputed issue of fact at trial.” Id. at 515.15 In her concurring opinion in Bamum, Justice Pariente (joined by Justices Anstead and Quince) further elaborated that the defendant had taken the witness stand and testified that he did not believe the victim to be a law enforcement officer at the time of the offense. This was the only direct evidence relating to this state of mind element of the crime. In addition, the officer did not display his badge as he approached the defendant. Yet, as the opinion explains, the officer’s testimony that he announced himself to the defendant as a law enforcement officer “was sufficient [circumstantial evidence] for the trier of fact to have concluded that this element was met [beyond a reasonable doubt[.]” The rationale for this conclusion, as explained by Justice Pariente, is that:

Because intent is a mental state seldom subject to direct proof, the determination whether the defendant knew that the victim was a law enforcement officer is an issue that in most circumstances remains for the jury. See Washington v. State, 737 So.2d 1208, 1215-16 (Fla. 1st DCA 1999) (“The law is clear that a trial court should rarely, if ever, grant a motion for judgment of acquittal on the issue of intent. This is because proof of intent usually consists of the surrounding circumstances of the case.”) (citation omitted).

Id. at 535.

In contrast with these cases are a few more recent cases from the district courts which, relying on Walker (the premeditation case from the Florida Supreme Court discussed in footnote 14, supra), apply the special standard anytime that the defendant’s intent is proven solely through circumstantial evidence. See, e.g., Galavis v. State, 28 So.3d 176 (Fla. 4th DCA 2010); Siplin v. State, 972 So.2d 982 (Fla. 5th DCA 2007).

In summary, there is a conflict and much confusion about whether the “special standard” applies only when “the evidence for each element of each offense” is wholly circumstantial, Helms, 38 So.3d at 185 n. 1, or anytime that any “element of a crime is based wholly on circumstantial evidence.” Alleyne, 42 So.3d at 950. This confusion is particularly evident in the inconsistent manner that our courts have addressed state of mind elements, which are almost always proven by circumstantial evidence, and for which the special standard is, in our view, inappropriate. This confusion was also frankly acknowledged recently in *467Green v. State, 90 So.3d 835 (Fla. 2d DCA 2012):

I am not entirely convinced that the outcome of this case is controlled by the special standard generally applicable to cases “where a conviction is wholly based on circumstantial evidence.” State v. Law, 559 So.2d 187, 188 (Fla.1989); see also Walker v. State, 957 So.2d 560, 577 (Fla.2007); Galavis v. State, 28 So.3d 176, 178 (Fla. 4th DCA 2010). I am the first to admit that the line that distinguishes direct evidence from circumstantial evidence is sometimes not intuitive to me. Moreover, it is not always easy for me to decide whether a conviction or an element of an offense is based wholly on circumstantial evidence, or based on both direct and circumstantial evidence thereby rendering it unnecessary to apply the special standard of review applicable to circumstantial evidence cases. See Wilson v. State, 493 So.2d 1019, 1022 (Fla.1986), receded from on other grounds by Evans v. State, 838 So.2d 1090 (Fla.2002).

Id. at 838 (Altenbernd, J., concurring). Thus, even if the supreme court does not join all federal courts and the vast majority of states in abandoning the “special” standard of review altogether, it still needs to clarify when the standard applies. But, consistent with our understanding of the law in this area, we hold that the special standard does not apply in this case because this is not a wholly circumstantial evidence case.

D. Why the Defendant’s Knowledge of the Presence of the Substance Should Be a Jury Question (Under Either Standard).

Although the special standard reads as if — and is sometimes applied as if — the reviewing court must either accept any inference that could logically be drawn from circumstantial evidence as a reasonable hypothesis of innocence or, alternatively, determine the reasonableness of the proffered hypothesis of innocence on its own, our supreme court has also repeatedly declared that the “question of whether the evidence fails to exclude all reasonable hypotheses of innocence is for the jury to determine_” Caylor v. State, 78 So.3d 482 (Fla.2011) (quotations and citations omitted). Obviously, the question is not always one for the jury. But, we are aware of no cases addressing the special standard which adequately articulate what test an appellate court should use to determine when, under the special standard, the issue is one that should be made by the jury as opposed to decided as a matter of law.16 In our view, if we are to retain the *468special standard, the supreme court should also address this question. And, in our view, if we are to retain the special standard, the answer must be whether, viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could reject the proffered hypothesis of innocence as unreasonable. If, taking into account the proper role of jurors along with all of the evidence and all reasonable inferences, the appellate court determines that no reasonable juror could reject the hypothesis of innocence as unreasonable, the conviction should be set aside on a sufficiency of the evidence review. Alternatively, if the appellate court determines that a reasonable juror could reject the hypothesis as unreasonable after weighing the credibility of witnesses and in light of all of the evidence and all reasonable inferences, the matter is to be decided by the jury and not on a motion for judgment of acquittal.

Turning to this case, the two inferences that could logically be drawn from the circumstantial evidence are that: (1) Knight knew that the marijuana was in his suitcase (and is guilty); or, (2) Knight did not know that the marijuana was in his suitcase because Chad Harris placed it there without his knowledge after Deputy Murphy removed Knight from the car (Knight’s hypothesis of innocence). This is admittedly a close case. However, we believe that a reasonable fact-finder could reject Knight’s hypothesis of innocence as unreasonable based upon: (1) the very short window of opportunity Harris would have had to move the marijuana from another hiding place to the suitcase (the K-9 deputy arrived within minutes of the stop); (2) the fact that an inference could reasonably be drawn from Chakra Miller’s testimony that Harris did not place the marijuana in Knight’s suitcase (Miller was in a position to detect any attempt by Harris to hide 24.4 grams of marijuana in Knight’s suitcase, but did not testify to any facts indicating that Harris actually did so); and (3) the jury’s unique ability to assess Knight’s demeanor on the witness stand during the whole of his testimony.

Finally, we note that we would be grappling with this same “close call” under what we view as the appropriate standard of review: “If after viewing the evidence in the light most favorable to the State, a rational trier of fact could find the existence of the elements of the crime beyond a reasonable doubt, sufficient evidence exists to sustain a conviction.” Pagan, 830 So.2d at 803. Applying this general standard in no way changes the fact that there are two possible, inferences that could be drawn from the evidence relating to the knowledge element in this case, or that a conviction can only be sustained if a trier of fact could rationally find guilt by rejecting the second inference as unreasonable, speculative, imaginary or forced based upon the evidence. Accordingly, even if another appellate court disagrees with the ultimate conclusion we reach in affirming the conviction in this case, it is important to emphasize that that disagreement would have nothing to do with the broader issue of whether the “special standard” should *469be rejected as confusing, unnecessary, and incorrect.

AFFIRMED; CONFLICT CERTIFIED.

ORFINGER, C.J., concurs. TORPY, J., concurs and concurs specially with opinion.

. "Circumstantial evidence is proof of certain facts and circumstances from which the trier of fact may infer that the ultimate facts in dispute existed or did not exist.” Baugh v. State, 961 So.2d 198, 203 n. 5 (Fla.2007) (quoting Davis v. State, 90 So.2d 629, 631 (Fla.1956)). By contrast, "[djirect evidence is that to which the witness testifies of his own knowledge as to the facts at issue." Id.

. The trial court would eventually grant a defense motion in limine which prohibited the State from eliciting testimony from Deputy Murphy that the cash was bundled using rubber bands in a manner typical of the method that drug dealers use to carry their cash.

. Contrary to Miller’s testimony that the group paid for their expenses in cash, Knight testified that someone in the group used a credit card and that he would give that person cash to cover his portion of the hotel bill.

. There was one exchange between Knight and his counsel during which the jury could have understood Knight to be agreeing that the suitcase was his. However, the question was poorly worded, and could have been intended by Knight as simply an acknowledgement that Deputy Murphy had identified the suitcase as belonging to him.

. As explained in the legislature eliminated the element of knowledge of the illicit nature of the drugs for offenses occurring after May 13, 2002. Id. at 809 n. 2. Thus, although some of the cases with which we conflict involved two knowledge elements, in this case the State was only required to prove knowledge of the presence of the contraband. Butera v. State, 58 So.3d 940, 942 (Fla. 2d DCA 2011).

. By contrast, in Jackson v. State, 995 So.2d 535 (Fla. 2d DCA 2008), the Second District upheld a conviction for possession of cocaine and MDMA based upon evidence that the container — a purse — in which the drugs appeared to have been located belonged to the defendant, even though others had access to the container. We believe that the panel in Jackson analyzed the issue correctly, and view the Jackson opinion as consistent with our analysis and holding in this case.

. If Harris had the cannabis on or near his person, he might have decided that the illicit substance would be less readily found if placed in the luggage, as a precaution in the event that the stop resulted in a search of the car or his person — or, that if the cannabis was later found, its location in someone else's luggage could absolve him from any criminal responsibility for the drugs. Thus, Knight had a credible argument for why the jury should have accepted his hypothesis of innocence as reasonable. Unlike this case and N.K.W., in some of the similar cases there does not appear to have been any identifiable motive for others with access to the container to have placed any contraband in it. Even if we were to declare Knight’s hypothesis of innocence to be reasonable as a matter of law in this case, we still would not understand why a jury could not reject the reasonableness of a defendant’s hypothesis of innocence that someone else put drugs in his bag, wallet, backpack or other container, in cases like E.H.A. (from the Fourth District) or S.B. (from the Second District) where no one else with access to the container had any discernible motive to place drugs there.

. Although this annotation was originally published in 1985, the online version to which we have access is updated "by the weekly addition of relevant new cases” according to the publisher, and includes a 2012 copyright date.

. The supreme court simply cited to the pre-1981 standard of review cases as the law in *457Florida, without examining whether it should continue using the standard after eliminating the Webster instruction. Id..; compare, e.g., People v. Pintos, 133 II1.2d 286, 139 Ill.Dec. 832, 549 N.B.2d 344, 348 (1989) (rejecting the reasonable hypothesis of innocence standard of review for wholly circumstantial evidence cases after analyzing the issue and concluding that "[n]othing can be gained on appeal by redefining the applicable standard in the obscure and misleading language we have rejected for jury use.”).

. In fact, “even a majority of the minority of jurisdictions that still require a cautionary instruction with respect to circumstantial evidence do not use an analogous sufficiency of the evidence test on appellate review.” Rosenberg, supra, at 1418 (footnote omitted).

. See, e.g., Rosenberg, supra, at 1420 (arguing that the reasonable doubt sufficiency standard is insufficient to provide “concrete substance for the presumption of innocence” and that "the reasonable hypothesis [of innocence] standard seeks to accomplish this end in a more refined and discriminating manner by providing a framework for assessment of circumstantial evidence.”) (footnotes omitted).

. Even though the state's evidence is based entirely on the in-court testimony of the eyewitness victim and another eye-witness (a classic direct evidence case that hinges entirely on the jury’s evaluation of the relative credibility of the witnesses), the state is still relying on circumstantial evidence to prove intent.

. For example, even in a murder case with ample evidence of premeditation, it would still be possible for a defendant who planned a murder to change his or her mind — firmly deciding not to commit the murder — but then kill the person in the heat of passion anyway. How, on a cold record, could a reviewing court logically dismiss the defendant's "reasonable hypothesis” that his or her state of mind at the time of the murder was contrary to the ample evidence of premeditation?

. This analysis is confused somewhat by Walker v. State, 957 So.2d 560 (Fla.2007). In Walker, a defendant convicted of first-degree premeditated murder argued that there was insufficient evidence to support his conviction under the special standard of review from circumstantial evidence cases. Noting that there was direct evidence that the defendant had killed the victim, the Court rejected this argument, noting that "This is not a purely circumstantial evidence case.” Id. at 577. However, the Court then explained that because "the State's case as to the intent element for first-degree premeditated murder ... [was] based on circumstantial evidence” it would "apply the special standard of review only to the State’s evidence establishing the element of premeditation.” Id.

. The primary issues in Bamum centered on the retroactive application of an earlier Florida Supreme Court case holding knowledge that the victim was a law enforcement officer to be an essential element of the crime to be proved at trial (as opposed to a sentencing enhancement factor). Because this and other important aspects of Bamum are irrelevant to the issue we address, we focus narrowly on the one aspect of that case that is relevant.

. As discussed in Judge Torpy’s concurring opinion, this issue is further confused in that our courts sometimes use similar but materially different articulations of the “special" circumstantial standard. The articulation that the state need only present evidence that is inconsistent with the defendant's hypothesis of innocence is sometimes also stated as the test to use for determining when the issue is one for the jury. See, e.g., Law, 559 So.2d at 189 ("where there is an inconsistency between the defendant’s theory of innocence and the evidence, when viewed in a light most favorable to the State, the question is one for the finder of fact to resolve”). This works in the rare (and easy) case in which a piece of circumstantial evidence directly contradicts some aspect of the defendant's story. See, e.g., Kocaker v. State, — So.3d - (Fla. 2013). But, it adds nothing in a case like this one. Here, for example, the circumstantial evidence is inconsistent with Knight’s general claim that he had no knowledge of the presence of the drugs, but not with his more specific theory that Harris could have placed the drugs in his suitcase without his knowledge. The same would be true if there had been no passengers in the car and Knight had theorized that a friend or roommate who did not make the trip could have slipped the drugs into his suitcase shortly before he loaded the car, unknown to him, thereby using him as an unwitting courier to get the drugs *468to someone who Knight was meeting in Orlando. Similarly, there would be no evidence to directly contradict a hypothesis that the K-9 officer planted the drugs during the search, without Knight’s knowledge, or, for that matter, that aliens were somehow involved. Given these four increasingly unlikely scenarios, all with no piece of evidence offered at trial to directly contradict the hypothesis of innocence, the "standard” articulated in Law provides no guidance as to when the appellate court must accept the hypothesis of innocence as reasonable (as a matter of law), can reject it as unreasonable, or should conclude that the issue is one for the jury.