Helton v. Singletary, 85 F. Supp. 2d 1323 (S.D. Fla. 2000)
January 20, 2000
Harry SINGLETARY, Respondent.
United States District Court, S.D. Florida.
*1324 Melvin Sidney Black, Miami, FL, for plaintiff.
Michael J. Neimand, Jonathan Ellsworth, Florida Atty. General's Office, Miami, FL, for defendant.
ORDER ON PETITION FOR WRIT OF HABEAS CORPUS
ROETTGER, District Judge.
THIS CAUSE is before the court upon a petition for a writ of habeas corpus pursuant to 28 U.S.C. § 2254. Petitioner is Kris Helton (Petitioner), and Respondent is Harry Singletary, Secretary of the Department of Corrections for the State of Florida (State).
In 1991, Petitioner was living with his fiancée, Marcella Gunderson, and her three children Matthew who was age six, Michael who was age four, and Marshall, who was twenty-two months. They lived in a rental home in Little Torch Key, Florida. Petitioner sold computers. Marcella Gunderson was the director of a kindergarten, and baby-sat children.
On July 31, 1991, at around 6:30 P.M., Marcella Gunderson was trying to feed Marshall supper, chicken and rice, but the child was feeling ill and was reluctant to *1325 eat. She got Marshall to eat a little bit of strawberry yogurt instead. Later, at 8:30 P.M., Marcella Gunderson put Marshall and the four year old to bed.
Petitioner was out of the house. He came home that evening at 9:30 P.M. All of the children were in bed by then. At 11 P.M., Marcella Gunderson and Petitioner lay in their bed watching TV. Marcella Gunderson checked on the children, and then she and Petitioner fell asleep at around 11:15 P.M. Marcella Gunderson awoke at 1:30 A.M., finding Petitioner sitting up in bed smoking, with a worried look on his face. Petitioner told Marcella Gunderson that she had awakened him with her kicking. In the meantime, while she was up, Marcella Gunderson decided to check on the children again. Marshall was not in his crib. Alarmed, Marcella Gunderson got Petitioner to help her look for the child.
In the kitchen the refrigerator door was open, and a jar of Kool-Aid lay spilled on the floor. The front door of the house was also open. In the front yard, at the bottom of the front-door steps, they found Marshall's body lying face down. The child was not breathing. Marcella Gunderson picked him up, and his head and arms flopped back. Marshall was moved into an ambulance and taken to Fisherman's Hospital in Marathon. The ambulance arrived at 2:20 A.M. Resuscitation attempts failed, and the boy was pronounced dead upon arrival.
An autopsy was performed at 3:00 P.M. the following afternoon. The official cause of death was severe trauma to the head, represented by two massive fractures of the skull. The Medical Examiner noted finding pink particulate in the boy's stomach, along with fibrous white matter described as rice. A specific time-of-death assessment was not made on the body at the autopsy. Still, the Medical Examiner would later testify at trial that the death likely occurred one to two hours prior to the child's discovery at 1:30 A.M., though he conceded that this range could conceivably be upwards of four hours. See infra.
Petitioner was charged with first degree murder. The State's case against him was entirely circumstantial. On March 17, 1992, a jury returned a guilty verdict, and Petitioner received life imprisonment as a sentence. On direct appeal, the Florida Third District Court of Appeal unanimously reversed the conviction, finding that the evidence was insufficient to support a circumstantial evidence conviction. But in July of 1994, the court of appeal granted re-hearing upon the State's motion, and withdrew its former reversal. See Helton v. State, 641 So. 2d 146 (Fla.App.1994).
The issue on appeal, whether there was enough evidence to support the conviction, got different treatment from the district court of appeal the second time around. To begin, the court restated the special standard of review applicable to convictions based on circumstantial evidence: a conviction can stand only if the evidence is inconsistent with any reasonable hypothesis of innocence. Id. at 146 (citing State v. Law, 559 So. 2d 187, 188-89 (Fla.1989)). Applying this standard, the court of appeal reviewed all hypotheses which exculpated Petitioner, and measured them against the evidence.
To the theory that Marshall died by accident, the court found that there was no way the child could have opened the dead-bolt-locked front door. To the theory that an intruder committed the murder, the court found there was no sign of intrusion, that there was a tall, strong, well-locked fence around the house, that there were dogs, and a child-proof lock which could only have been opened from the outside with force; accordingly, the court dispensed with the intruder theory. To the theory that Marshall's older brothers could have done it a theory not actually argued by Petitioner the court summarily found no supporting evidence. To the theory that Marcella Gunderson's estranged husband did it, the court found that he did not have keys to the house, and in any event, with the child proof lock, he would have had to force the door open. Considering the theory that Marcella Gunderson committed the murder, the court found that *1326 her testimony was believable, and that the jury was entitled to rely on her testimony as evidence. See Helton, 641 So. 2d at 148-53.
The court of appeal addressed two new theories as well. The first of these, a hybrid of a couple theories, suggested that the brothers opened the front door locks, and either an accident ensued, or an intruder came along. The court rejected this theory on many grounds, but basically because the children could not have reached the child-proof latch six feet off the ground. See id. at 153-54. The other theory postulated that Marcella Gunderson committed the murder prior to Petitioner arriving home at 9:30 P.M. The court found this theory inconsistent with the trial testimony of the Medical Examiner, who, based on the apparent lack of rigor mortis on the body, originally placed the time of death within two hours prior to the discovery of the body at 11:30 P.M. or later.Id. Thus, in finding evidentiary contradiction to all logical hypotheses of Petitioner's innocence, the court of appeal affirmed the conviction. Id. at 154.
Notably, in a dissenting opinion, Judge Joseph Nesbitt stated that he would reverse the conviction and award a new trial. See id. at 154-55 (J. Nesbitt, dissenting). Judge Nesbitt found it significant that the Medical Examiner acknowledged that death could have occurred as many as four hours prior to discovery of the body. Judge Nesbitt further noted that gastric evidence, combining the fact that Marshall still had the small meal in his stomach with what is known about digestive times, suggested that the boy died prior to 9:30 P.M., the time Petitioner arrived home. In view of these circumstances, Judge Nesbitt found that defense counsel's failure to discern and employ this seemingly exculpatory gastric evidence in Petitioner's defense qualified as substantially deficient representation. Judge Nesbitt concluded his dissent: "There can be no doubt that this evidence might have affected the verdict rendered." Id. at 156.
The Florida Supreme Court denied review of Petitioner's direct appeal on January 31, 1995, and the United States Supreme Court did the same on October 2, 1995. Petitioner then pursued collateral relief in state trial court pursuant to Florida Rule of Criminal Procedure 3.850, based in large part on Judge Nesbitt's deficient representation concerns. The trial court denied Petitioner's motion, and the Third District Court of Appeal for Florida denied appeal of Petitioner's collateral motion on December 10, 1997. Petitioner later filed a pro se habeas corpus petition in the state appeals court, and that was denied May 26, 1998. The Supreme Court of Florida denied review of that ruling on August 24, 1998.
Petitioner filed the instant action under 28 U.S.C. § 2254 on December 9, 1998. The petition includes twenty-seven claims of ineffective assistance of counsel. The most prominent claims center around counsel's failure to properly investigate and employ a time-of-death argument based on the gastric evidence. The State *1327 replied to the petition on February 24, 1999. In large part, the response attacks the petition as untimely. This court specifically reserved judgment on the issue of timeliness, and held an evidentiary hearing on the gastric evidence/time-of-death claims on May 11 and 12, 1999, in Key West, Florida.
The State argues that the petition is untimely under the Antiterrorism and Effective Death Penalty Act of 1996 (ADEPA). The ADEPA prescribes a period of one year from the conclusion of a Defendant's direct appeal, minus time spent litigating collateral post-conviction claims, within which to bring a habeas corpus petition. See 28 U.S.C. §§ 2244(d) (1) (A), 2244(d) (2). In the instant case, Petitioner's direct appeal concluded with the United States Supreme Court's denial of certiorari on October 2, 1995. However, since this date was prior to the date the AEDPA became effective, the one year statutory period did not begin running until that date of effectiveness, April 24, 1996. See Wilcox v. Fla. Dept. of Corrections, 158 F.3d 1209 (11th Cir. 1998). As mentioned, this petition was filed December 9, 1998. According to the State, factoring out the time Petitioner spent pursuing post-conviction relief, 412 days elapsed between April 24, 1996 and December 9, 1998. Thus, the State argues, the instant petition is barred under the AEDPA.
Petitioner responds with an equitable tolling defense. Equitable tolling can be used to prevent the harsh application of the ADEPA's statutory deadline when "extraordinary circumstances" have worked to prevent an otherwise diligent petitioner from making a timely filing. Sandvik v. United States, 177 F.3d 1269, 1271 (11th Cir.1999). In the instant case, Petitioner claims equitable tolling is warranted because his collateral counsel misadvised him as to when the one year limitations period began to run. Counsel advised Petitioner that he had one year from the date of denial of his state motion to vacate, December 10, 1997, within which to file his federal petition. Petitioner points out that he did file his petition within this time period it was filed directly on the presumed December 9, 1998 deadline. What Petitioner clearly did not know is that the deadline had been running from April 24, 1996.
The issue is whether these circumstances qualify as extraordinary circumstances. The United States Supreme Court has noted that equitable tolling has been allowed in situations where the claimant has actively pursued his judicial remedies. Irwin v. Dept. of Veterans Affairs, 498 U.S. 89, 111 S. Ct. 453, 457-58, 112 L. Ed. 2d 435 (1990). Conversely, equitable tolling is not available to one who has "failed to exercise due diligence in preserving his legal rights" meaning those who have acted lazy or recalcitrant, or who have ignored notices of limitations deadlines, or who have indefensibly relied on their adversary's statement of the deadline. Id.
In this case, it is clear that Petitioner has diligently pursued his legal rights on appeal. While the one year limitations period began running in April of 1996, it did not expire until about a month prior to the filing of this petition. Petitioner spent roughly a year and a half prosecuting post-conviction appeals. It is significant that *1328 Petitioner filed his petition within the time he reasonably believed he was allowed. This is not a case of laziness or recalcitrance on Petitioner's part. It is, rather, a case of relying on advice of his counsel.
The court is also persuaded by the compelling nature of this petition. The strange history of this case from the dubious weight of evidence, to the circumstantial conviction, to the flip-flopping appellate record contributes to the overall extraordinariness of the circumstances. At bottom, this court is not disposed to apply mechanically the limitations period when review is so glaringly warranted. Such a result would belie the meaning of "equitable tolling."
Accordingly, while misinformation of counsel may not always, of itself, have the seismic thrust needed to qualify as "extraordinary circumstances," the court is nonetheless persuaded that this case fits with the spirit of equitable tolling: a diligent movant who, through no fault of his own, was blocked in filing his petition by the deadline. See Irwin, 111 S. Ct. at 457-58, Sandvik, 177 F.3d at 1271-72. The court therefore finds that while the AEDPA limitations period expired prior to the filing of this petition, equitable tolling is appropriate, and consideration of the merits of this petition is in order.
INEFFECTIVE ASSISTANCE OF COUNSEL
The United States Supreme Court enunciated the principles governing ineffective assistance of counsel claims in Strickland v. Washington, 466 U.S. 668, 104 S. Ct. 2052, 80 L. Ed. 2d 674 (1984). A petitioner must satisfy a two prong test: first, counsel must have made errors so profound that the performance dropped below a constitutional minimum standard; second, there must be a reasonable probability that but for counsel's errors, the result would have been different. See id. at 2064-68. This test inheres in the Sixth Amendment's guarantees to the right to assistance of counsel, and the right to a fair trial; it is designed ultimately to determine whether the proceeding was fair enough to instill confidence in the outcome. See id. at 2063.
The right to counsel has been interpreted as the right to reasonably effective counsel. Id. at 2064. Counsel's representation simply need not fall below an objective standard of reasonableness for attorneys in criminal cases. See id. at 2064. See also Spaziano v. Singletary, 36 F.3d 1028, 1041 (11th Cir.1994) (judging reasonableness by whether "the approach taken by defense counsel would not have been used by professionally competent counsel"). In meeting this standard, attorneys have certain obligations to their clients, including the duty to investigate as reasonably necessary. See Strickland, 104 S. Ct. at 2066. There are, though, no fixed set of rules or guidelines for a criminal defense attorney to follow. As Strickland makes clear, there is a wide range of assistance that is reasonable and acceptable in any given case. Id. at 2065.
Because of this, a heavy measure of deference is accorded counsel. Id. Judging reasonableness involves looking at the conduct from counsel's perspective at the time, in light of all the circumstances, and does not involve vigorous second-guessing, scrutinizing hindsight, or exaggerated appraisal of every mistake. Id. Indeed, most errors will not support an ineffective assistance of counsel claim, even if the errors are unreasonable, for most errors will not prejudice the defense to the point where the result can be questioned. Id. at 2067.
*1329 B. Gastric Evidence
The evidentiary hearing on gastric evidence, held May 11 and 12, 1999, featured the competing testimony of two pathology experts, Dr. John Feegel, Petitioner's expert, and Dr. Michael Bell, for the State. Neither expert was present at the original autopsy of Marshall Gunderson, but based their testimony on their review of the autopsy report. The original Medical Examiner did not testify.
The autopsy revealed "pink particulate" or "pink goo" in Marshall Gunderson's stomach. Dr. Bell testified that he did not find rice in the stomach, but he was only looking at photographs. The fact that the Medical Examiner reported rice-like particles in the stomach trumps Dr. Bell's skepticism. The State also noted that no chicken was found in the stomach, suggesting that some digestion did occur. However, this argument not only conflicts with Dr. Feegel's explanation of how digestion works (there is no priority of digestion based on the type of food), but it is also inconsistent with the evidence. Marcella Gunderson testified that she tried to feed Marshall some chicken, but that Marshall did not want any. The fact that none was found only verifies that the child did not eat any.
The question becomes, what is the significance of Marshall still having in his stomach this small meal he ate at 6:30 P.M.? In this regard, there is some convergence of opinion between Dr. Feegel and Dr. Bell. Dr. Feegel concluded with much certainty that a small meal will usually digest in one to one-and-a-half hours. Dr. Bell testified that he had read the same. But Dr. Bell asserted that this range has been known to vary, according to different texts, by two or more hours; the range, Dr. Bell testified, is too susceptible to outside factors to be reliable. Notwithstanding these "other factors," the court can reasonably conclude based on the weight of authority that under normal conditions Marshall should have digested his small meal within one to two hours after eating.
Assuming arguendo that the murder occurred after 11 P.M., the fact that Marshall had not digested the meal prior to his murder could be accounted for by intervening factors. The State proffered Marshall's ear-ache, his taking Benadryl, and his ingestion of starchy foods as evidence that digestion in this case had been suspended or retarded. However, a simple ear-ache is not an illness severe enough to retard digestion. The starchy foods argument is more compelling. Although the amount of rice Marshall ate was not significant, yogurt too may be considered a starchy food, and this was most prominent among the stomach contents. The amount of Benadryl found in the boy's blood and urine was very small, but Dr. Bell pointed out that this amount may be misleading, since some of the Benadryl may have matriculated into the system.
On the other hand, these retardant factors are buffeted by less speculative facts. Foremost is the pink coloration of the yogurt found in Marshall's stomach. For Dr. Feegel, this fact was decisive in his conclusion that digestion had hardly begun when it was interrupted by death. Early in the course of digestion, Dr. Feegel explained, food coloring is one of the first things washed away, leaving what he described as a slushy gray liquid. This would be the case even if emptying time, or the total time it takes for the stomach to empty, got delayed. The integrity of the pink color undercuts the State's theory of suspended digestion.
Dr. Bell disputed that the pink color of the "goo," as he described it, held any significance. But Dr. Bell's testimony regarding coloration, and on digestion as a whole, seemed based more on what he had read and less on his experience. With Dr. *1330 Feegel, the opposite seemed true. As such, the court is more compelled by Dr. Feegel's conclusions on the significance of the color pink. These conclusions are buttressed by the fact that the meal Marshall consumed at 6:30 P.M. was small. Even Dr. Bell testified that the size of the meal affects emptying time. Given this, even in the presence of the retardant factors, it seems unlikely that disposal of the meal would be suspended for upwards of six or even seven hours.
Nevertheless, the State argued that other more reliable factors conflict with the conclusions suggested by the gastric evidence. Foremost is the apparent lack of rigor mortis in Marshall when he was found. Rigor mortis, or rigidity, is the stiffening of the body after death. The process of stiffening occurs in the range of two to seven hours, with full rigor being established near the end of the range. In this case, Marcella Gunderson testified at the trial that when she picked the child up off the ground, Marshall's head and arms "flopped" back. The State reasons from this that rigor mortis had not begun to form, and deduces therefore that the child had been recently killed. Dr. Feegel, however, found that the disproportionate mass of a child's head, as compared to the rest of the body, could have broken the rigidity in the child's neck; the same may have accounted for the arms falling back. Dr. Bell expressed no real opinion as to this explanation. Much of the testimony seemed in response to Marcella Gunderson's characterization of the arms and head as having "flopped" back how, the suggestion was, could these parts "flop" if rigor mortis had been setting in for four or more hours?
But analysis of Marcella Gunderson's verbs is hardly solid ground for determining the actual state of the body. The better approach seems to question what signs of rigor mortis Marcella Gunderson could have appreciated as a lay person in an emergency situation. Dr. Feegel stressed that because rigor mortis begins to set in on all the muscles at the same time, the smaller muscles naturally stiffen faster than the larger muscles; rigor first becomes apparent in the eyelids, the jaw, and the extremities. Because of this, even one with some experience may have trouble detecting rigor mortis after one or two hours. That person will not know where to look. According to Dr. Feegel, it is unlikely someone with no experience could assess the onset of rigidity.
Dr. Bell, perhaps inadvertently, bolstered Dr. Feegel's theory in an aside. Dr. Bell recalled how bodies sometimes are brought to him with tracheotomy tubes still in the necks. This, Dr. Bell explained, is the result of paramedics not being able to open the jaw to get the tubes in through a patient's mouth. The reason the paramedics cannot open the jaw, of course, is that rigor mortis has set in. Trained paramedics are so driven toward saving lives that even they do not notice the onset of rigor mortis. It would not be surprising, then, assuming again that Marshall died shortly before 9:30 P.M., if Marcella Gunderson was not struck with Marshall's rigidity four hours later four hours being in the middle of the development range toward full rigidity. It is doubtful she had the expertise to detect it.
A phenomenon similar to rigor mortis in its effect on a dead body is livor mortis, or lividity. Livor mortis results from the blood settling in the vessels from gravity. It appears as red and dark-blue or purple discoloration. Livor mortis will normally appear within one or two hours after death, and is apparent on the side on which the body is lying. In this case, there was no observation of lividity on Marshall's body at the time of discovery or shortly thereafter. Petitioner introduced photographs of Marshall's body taken in the Emergency Room of Fisherman's Hospital after 2:00 A.M. Dr. Bell testified that he could not detect any signs of lividity from the photographs. The flash of the camera and the angle of shadows may *1331 have affected the verisimilitude of the photographs. Notwithstanding, Dr. Bell argued that this lack of lividity supports what was also suggested by the apparent lack of rigor mortis that the child had been killed within two hours of 1:30 A.M.
Dr. Feegel testified that lividity becomes affected when a body is moved frequently. The direction of gravity's pull is changed, and so the pattern of livor mortis will change as well. In this case, Marshall's body was moved from the yard to the ambulance to the hospital; the body may have been moved more than this if Petitioner's theory is to be believed. According to Dr. Feegel, this frequent moving could account for the lack of apparent lividity on the child after 2:00 A.M., even if he had been killed four or five hours earlier.
Dr. Bell did not directly dispute Dr. Feegel's testimony, and in fact acknowledged that moving the body will affect the appearance of lividity. Dr. Bell also indicated that loss of blood can affect lividity. However, Dr. Bell stressed, the bruising, loss of blood, and internal bleeding in Marshall's case was likely not enough to have an appreciable effect on the formation of livor mortis. There was testimony from the Medical Examiner, however, that Marshall lost as much of a third of his blood through internal bleeding. Notwithstanding, Dr. Bell testified that while a change of a body's position may cause a change in the pattern of lividity, that change will not be immediate, but will be a very gradual change, taking place over the course of a couple hours as the blood trickles toward the new direction of gravity's pull.
All of the three competing time-of-death factors, then, are subject to variables. A meal may be held up in the stomach, rigor mortis may be difficult to detect, the patterns of livor mortis may change over time. No one factor appears more compelling or less compelling than the other two, and that is the difficulty. Indeed, a small meal will digest just as certainly as a dead body will stiffen and livor mortis will form. This kind of conflicting evidence does not allow one to resort to comforting platitudes, such as "the most obvious explanation is likely the best explanation." Rather, something must be reconciled with its opposite. Time-of-death is a vastly inexact science when it is reliant on outside factors. As Dr. Feegel explained, when one is talking about time-of-death, one is talking about a range of hours. Importantly, though, a defendant does not have the burden to prove time-of-death, or anything else. Given this, it is enough for the court to find that Petitioner's theory is at the very least tenable.
C. Trial Counsel
Petitioner's trial counsel was Gerod Hooper, who testified at the evidentiary hearing as Petitioner's second witness. Mr. Hooper became a member of the New Jersey bar in 1981, and a member of the New York bar in 1982. He worked as an attorney for the New York City Police Department until 1988, prosecuting police officers in administrative proceedings. He did not practice criminal defense until 1988, when he retired from the police department and started a private practice in Central Valley, New York. Criminal defense accounted for a small portion of Mr. Hooper's practice in Central Valley. After one year, Mr. Hooper moved to Florida. He became a member of the Florida Bar in January, 1990, and began work with the Monroe County Public Defender's office in March of that year. Mr. Hooper was appointed to Petitioner's case in the latter part of 1991, shortly after Petitioner's arraignment. It was Mr. Hooper's third first degree murder case.
After a series of preliminary meetings with Petitioner, Mr. Hooper considered employing several theories for Petitioner's defense. One, which was suggested by Petitioner, was that Marcella Gunderson killed Marshall prior to 9:30 P.M., before Petitioner arrived home. Mr. Hooper was aware of the Medical Examiner's report that there appeared to be yogurt and rice in the child's stomach, and knew that Marshall had last been fed at 6:30 P.M. Mr. *1332 Hooper also knew that Marcella Gunderson had stated she had last seen the child alive at 8:30 P.M. Further, Mr. Hooper had rudimentary knowledge of time-of-death evidence, and knew that stomach contents could be used for this purpose. But Mr. Hooper did not investigate the stomach contents evidence, and decided against advancing the theory that the child died prior to 9:30 P.M.
Mr. Hooper had plausibility concerns. As he saw it, the earlier death theory presupposes a string of unlikely events: Marcella Gunderson had to have killed Marshall while her other two kids were awake in the house; she then would have had to conceal Marshall's body until the others were put to bed; after Petitioner had come home and gone to bed, she then would have had to retrieve the body, take it outside, arrange for the murder to look like an accident, and return before Petitioner woke her up at 1:30 A.M. Beyond this, though, and at the root of his concerns, Mr. Hooper was hesitant to accuse Marcella Gunderson outright, for fear of turning the jury against him.
Mr. Hooper was also concerned with making inconsistent arguments. Mr. Hooper initially settled on four theories. The first was that Marshall had climbed out of his crib, fell, and died by accident; another was that an intruder committed the murder; another was that Marcella Gunderson's estranged husband killed Marshall; and the fourth suggested that Marcella Gunderson was the culprit. To argue that the murder occurred prior to 9:30 P.M. would, according to Mr. Hooper, preclude all theories but the last.
Yet at trial, while Mr. Hooper intimated the possibility of an accident, or an intruder, or the estranged husband, the main thrust of his argument went toward showing that the only two possible killers were Petitioner or Marcella Gunderson. Mr. Hooper estimated that in the absence of any physical evidence, the case against Petitioner was no stronger than the case against Marcella Gunderson. Where the level of certainty as to one or the other would appear not to exceed even the soft "more likely than not" standard, a guilty verdict "beyond a reasonable doubt" would be all but impossible, or so Mr. Hooper had assumed.
D. Conclusions of Law
The first prong of the Strickland test inquires into whether Mr. Hooper's strategy, his failure to investigate or pursue the gastric evidence, and his overall defense scheme, were reasonable under the circumstances. Foremost, the court must consider Mr. Hooper's experience. It is relevant to consideration of the reasonableness of his strategic decisions, and of his level of investigation. Spaziano v. Singletary, 36 F.3d 1028, 1040 (11th Cir. 1994). As indicated, Mr. Hooper was by no means an experienced defense attorney. While his experience in New York and New Jersey cannot be discounted, it is his inexperience in criminal defense which is notable. His strategic decision not to put forth the earlier time of death (or "alibi") defense, then, does not get the almost unassailable presumption of reasonableness which is accorded to experienced counsel. Compare Williams v. Head, 185 F.3d 1223, 1228-29 (11th Cir.1999) (finding the heavy presumption of reasonableness is stronger still when counsel is experienced).
Also important to the court's inquiry is the strength of the defense. What is most striking about the defense of alibi is its absoluteness in a case such as this, proof that the accused was not at the scene at the time of the murder is inviolable proof of innocence. Of course, a defense attorney is not obliged to pursue such a theory if it is far-fetched, if it is reliant on ultra-alternative scientific theories, if it is distasteful, or for a host of other reasons. In this case, though, there is sound scientific authority supporting the time-of-death evidence. See supra. As the court has analyzed thoroughly, Petitioner's gastric evidence argument, while certainly open to vagaries, remains defensible. A jury could be convinced, if not outright, at least to the *1333 extent the theory interposes a reasonable doubt.
Still, the compelling nature of the defense is only one consideration. The other circumstances prompting Mr. Hooper's hesitance toward the time-of-death challenge also require the court's consideration under the objective reasonableness standard. Mr. Hooper's biggest worry was that purporting a time-of-death theory would cancel three of his four other theories. However, it is the nature of alternative theories that they are mutually exclusive. To argue that an intruder murdered the child does not really derogate from the theory that the estranged husband did it. Naturally, to believe one theory is not to believe the other, but that does not mean that they cancel each other. By the same token, one can argue that the child was killed prior to 9:30 P.M. as one theory, and in the alternative argue that even if the murder occurred after 11:00 P.M., the culprit was someone other than Petitioner. While the theories are separate and carry their own implications, a time-of-death argument does not contradict the others, so much as provide an alternative (and by all means defensible) basis of innocence. Mr. Hooper should have discerned this distinction, as any reasonably competent defense attorney would have.
The implications of alleging the murder occurred prior to 9:30 P.M. also discouraged Mr. Hooper. But again, a defendant need not prove how he is innocent. Counsel need only propose a theory of innocence plausible enough to raise a reasonable doubt. It need not be a fully formed theory, and defense counsel need not connect all the dots. In this case, making the time-of-death argument has never required accusing Marcella Gunderson, or accusing anyone for that matter. It is a scientific argument based on facts. If the jury wished to flesh out the implications of the gastric evidence, it could have. But Mr. Hooper did not have to prove or even suggest what he decided was a string of improbabilities. Concluding that the jury would have been strained to believe a theory must still have a basis in reason. Mr. Hooper's concerns were not justified.
Meanwhile, the argument Mr. Hooper actually advanced, while avoiding implicating Marcella Gunderson at all (counsel in fact stated in closing argument that he "was not accusing Ms. Gunderson of killing the child"), and glossing over the more improbable intruder/estranged-husband/accident theories, was contrary to the law of Florida. As the Third District Court of Appeal explained, when two people are equally likely to have committed murder, and the evidence is limited to the testimony of one as against the other, a conviction can still stand. See Helton v. State, 641 So. 2d 146, 151-52 (Fla.App. 1994). The jury is free to accept in whole the testimony of one witness, and likewise reject the testimony of another. See Helton, 641 So. 2d at 151. That is what the jury did in this case, and counsel made a big mistake in assuming Marcella Gunderson's testimony would not be enough. Cf. Cave v. Singletary, 971 F.2d 1513, 1517-18 (11th Cir.1992) (holding that counsel's misunderstanding of the law of felony murder helped qualify as ineffective assistance).
This is particularly true in light of the marked absence of inculpatory physical evidence against Petitioner. Convincing exculpatory physical evidence for Petitioner would have carried much weight in undermining Marcella Gunderson's testimony, and in unhinging the State's case. Clearly, the error of Mr. Hooper's decision to do substantially nothing to combat the State's circumstantial case when a viable defense was available is that much more apparent.
For these reasons, then, the court finds that Mr. Hooper's decision to bypass the *1334 time-of-death argument, without any investigation into the theory's merits, was in error. A reasonably competent defense attorney in Mr. Hooper's position would not have pursued that tactic. Accordingly, the court finds that Mr. Hooper's representation fell below an objective standard of reasonableness.
Petitioner must further show that Mr. Hooper's error prejudiced the defense to where the outcome cannot reasonably be relied upon as a just result. See Strickland, 104 S. Ct. at 2064. This is the easier prong for Petitioner to satisfy, and the court's analysis need not go much further than it already has. As stated, the time-of-death argument does have firm enough scientific and evidentiary support to form a plausible, if not compelling, defense. As in all criminal cases, whereas the State bore the heavy burden of proving guilt beyond a reasonable doubt, counsel need have only submitted enough evidence to create a reasonable doubt. In light of the State's wholly circumstantial case against Petitioner, the gastric evidence could have provided that doubt. Mr. Hooper's decision not to investigate or pursue the time-of-death argument all but doomed Petitioner's chances. The court finds, then, that but for Mr. Hooper's decision, there is a reasonable probability the result of the trial would have been different.
In making the foregoing ruling, the court is mindful of the Eleventh Circuit's recognition that successful ineffective assistance of counsel claims based on counsel's strategic decisions are "few and far between." See, e.g., Williams v. Head, 185 F.3d 1223, 1227 (11th Cir.1999), Spaziano v. Singletary, 36 F.3d 1028, 1039 (11th Cir.1994). Still, it is an objective standard of reasonableness by which the acts of counsel are finally judged, and not insurmountable presumptions, or cursory truisms about the fate of most claims. Indeed, the mere incantation of the word "strategy" does not insulate counsel's decisions from review under the reasonableness standard. See Cave v. Singletary, 971 F.2d 1513, 1517 (11th cir.1992). This case, then, must be one of the rare examples.
For the reasons stated herein, upon consideration of the petition, the evidentiary hearing and the record, it is
ORDERED AND ADJUDGED that the petition for habeas corpus pursuant to 28 U.S.C. § 2254 is GRANTED. Petitioner's conviction is reversed, and a new trial is ordered.
ORDER ON STATE'S MOTION TO STAY
THIS CAUSE is before the Court on the state's motion to stay pending appeal the court's December 23, 1999 order granting. Petitioner's writ of habeas corpus. In determining this motion the court must consider four factors, originally outlined in Hilton v. Braunskill, 481 U.S. 770, 107 S. Ct. 2113, 95 L. Ed. 2d 724 (1987): whether the moving party has made a strong showing that he is likely to succeed on the merits; whether the movant will be irreparably injured absent a stay; whether the issuance of the stay will substantially injure the other parties; and where the public interest lies. Id. at 776.
Focusing on the initial factor, whether the movant has made a strong showing of probable success on appeal, the state posits three grounds on which it will appeal. The first challenges the court's determination that equitable tolling should apply to excuse Petitioner's late filing. The second argues that the court failed to give the proper deference as required under § 2254(d) to the state court's denial of Petitioner's ineffective assistance of counsel claims. The State's third argument on appeal attacks this court's conclusion that Petitioner's counsel was constitutionally ineffective as unsupported by the facts and the law.
This court's December 23rd order canvassed the first and third of the State's arguments. But the order did not address the State's second argument, regarding whether this court ran afoul of § 2254(d) *1335 by failing to give deference to the state court decisions. The argument presents problematic and intriguing issues. Under § 2254(d):
An application for a writ of habeas corpus on behalf of a person in custody pursuant to the judgment of a State court shall not be granted with respect to any claim that was adjudicated on the merits in State court proceedings unless the adjudication of the claim
(1) resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States; or
(2) resulted in a decision that was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.
28 U.S.C. § 2254(d). The Eleventh Circuit outlined a methodology for applying this subsection in Neelley v. Nagle, 138 F.3d 917 (11th Cir.1998). First, the district court must survey the "legal landscape" to determine whether the federal law applicable to a petitioner's claim has been clearly established by the United States Supreme Court. Next, the court must determine whether the state court's adjudication of the claim was contrary to the established Supreme Court law, either because the state court failed to apply the precedent, or because the state court reached a conclusion different than that of the Supreme Court on substantially the same facts. If the state court's decision is not contrary to federal law, the district court must determine whether the state court unreasonably applied the law. See id. at 924. Under this third inquiry, the state court decision must stand unless it is not debatable among reasonable jurists that the state court's decision was wrong. Id. at 924-25.
Applying these principles to the instant case starts out easily enough: the law applicable to Petitioner's claims of ineffective assistance of counsel has been clearly established in Strickland v. Washington, 466 U.S. 668, 104 S. Ct. 2052, 80 L. Ed. 2d 674 (1984). However, the analysis quickly becomes problematic. The reason for this inheres in the state court's treatment of Petitioner's claims. Petitioner made his claims of ineffective assistance of counsel at the state level both in a January 31, 1997 motion for post-conviction relief under Rule 3.850 of the Florida Rules of Criminal Procedure, and again in a petition for a writ of habeas corpus filed pro se on April 23, 1998 with the Third District Court of Appeal of Florida. The trial court issued essentially a one word order denying his Rule 3.850 motion on June 5, 1997, and the court of appeals similarly issued a one word, per curium affirmance of the ruling on December 10, 1997. Petitioner's state habeas corpus claims were also flatly denied without comment on May 27, 1998. In the record before this court, then, there exists no reasoning, no analysis, no findings of fact, and no legal bases for the state's denial of the ineffective assistance of counsel claims.
As a threshold issue, most courts have agreed that even cursory or perfunctory decisions, including one-word decisions, do qualify as adjudications "on the merits" for purposes of § 2254(d). See, e.g., Aycox v. Lytle, 196 F.3d 1174, 1177 (10th Cir.1999), Cardwell v. Greene, 152 F.3d 331, 339 (4th Cir.1998); but see Mercadel v. Cain, 179 F.3d 271, 274 (5th Cir.1999) (finding, in the absence of any reasoning by the state court, that evidence of a procedural defect in the petition can make the state court determination "procedural," rather than "on the merits"). The court will agree with the weight of authority on this count.
However, there are practical difficulties in applying § 2254(d)'s scheme of deference to such cursory decisions. As the second step under the Neelley scheme, the *1336 court must determine whether the state court decision was "contrary to" established Supreme Court law, "because the state court failed to apply the precedent." See Neelley, 138 F.3d at 924. In this case, there is no indication at all whether or not the various state courts applied the principles of Strickland and reached their decisions accordingly. This court could treat the issue, then, as a nullity, assume that the state courts did apply the Strickland analysis, and proceed to the third step in the Neelley scheme of determining whether the state courts' application of Strickland was "unreasonable." Or this court could make the inverse assumption that the state courts' perfunctory dispensations indicate the Supreme Court's decision in Strickland was ignored. Clearly, one word decisions were not what Congress had in mind when it drafted § 2254(d). It is doubtful that Congress intended for district courts to engage in speculation as to what went into the state court's decision. Indeed, when state courts simply leave nothing of substance for the federal courts to defer to, there appear no obvious or compelling reasons for making assumptions against the petitioner and for the state, merely for the sake of deference. Rather, where a terse state court decision conflicts with the conclusions reached by a district court after exhaustive review, a finding that the federal law went ignored at the state level is justified, and consistent with the "contrary to" language of § 2254(d) (1). Such a decision would certainly help to encourage state courts to shore up their judgments with at least a modicum of reasoning. As the Seventh Circuit has noted, "of course the better the job the state court does in explaining the grounds for its rulings, the more likely those rulings are to withstand further judicial review. That is just realism." Hennon v. Cooper, 109 F.3d 330, 335 (7th Cir.1997).
In the instant case, this court found, after a three day evidentiary hearing and painstaking consideration of the whole record, that Petitioner's claim of ineffective assistance of counsel was meritorious under Strickland. In the absence of any evidence that the correct legal principles as defined by the Supreme Court were applied at the state court level, this court cannot reasonably find that the state court applied federal precedent. The court finds, therefore, pursuant to the principles outlined in Neelley, that the state court rulings against Petitioner were contrary to federal law. Under § 2254(d) (1), the court need not be bound by those state court judgments.
In the alternative, even if the court were to pass over the § 2254(d) (1)'s "contrary to" analysis, and consider the case under the mixed-question-of-law-and-fact prong the "unreasonable" application of law analysis the court still finds that its conclusions withstand § 2254(d)'s requirements. Under Neelley, assuming that the state court did apply the proper law, the court must consider whether the application of that law was "unreasonable." The court measures unreasonableness by whether the result in state court is wrong to the point where its incorrectness is not debatable among reasonable jurists. See Neelley, 138 F.3d at 924.
Again, though, the lack of any analysis at the state court level hinders the court's consideration of unreasonableness. Having to extrapolate from the single word "denied" a structure of reasoning for a complicated series of legal issues seems a fruitless, pointless exercise. Faced with a similar dilemma, the Fourth Circuit observed, "because the state court decision *1337 fails to articulate any rationale for its adverse determination of [the petitioner's] claim, we cannot review that court's `application of clearly established federal law,' but must independently ascertain whether the record reveals a violation of [his] right to the effective assistance of counsel." Cardwell, 152 F.3d at 339. Nevertheless, this court is also aware that Congress did not intend for the district courts to engage in de novo review. See Neelley, 138 F.3d at 924. The independent review required by practicalities, then, must be merged with the principles of deference at work in § 2254(d). Accordingly, the court is left to consider independently and in the broadest possible terms, whether, given the facts and record as this court understands them, reasonable jurists would debate that Petitioner's ineffective assistance of counsel claim is meritorious.
As noted, the court exhaustively analyzed the issues involved in Petitioner's ineffective counsel claim in its December 23rd order. The conclusions of that order were by no means reached lightly. An independent review of the record has revealed that there was a violation of Petitioner's Sixth Amendment right to effective counsel, as defined under Strickland. This court cannot say, given counsel's failure to investigate the only viable avenue of defense based on physical evidence available to Petitioner at the time of trial, that reasonable jurists would disagree that counsel's performance was ineffective. Accordingly, even were this court to assume that the state courts did apply the Strickland analysis, the court finds that the state courts' application of Strickland was unreasonable under § 2254(d) (1). Therefore, again, this court is not bound by the state denials of Petitioner's ineffective assistance of counsel claims.
To the extent that these findings pertaining to § 2254(d) are absent from this court's December 23rd order, the court hereby amends that order to include them.
For these reasons, this court finds that the state has not made a strong showing that it is likely to succeed on the merits on appeal, as required under the Supreme Court's Braunskill decision. See 481 U.S. 770, 107 S. Ct. 2113, 95 L. Ed. 2d 724 (1987). This court also finds that the state cannot satisfy the other three Braunskill requirements either. The court cannot foresee any real injury that could affect any party in this case. Likewise, this court finds that the public interest lies more in having its citizens get fair trials than it does in staying orders for the state's appeal. Accordingly, it is
ORDERED AND ADJUDGED that the state's motion to stay this court's December 23, 1999 order granting Petitioner's writ of habeas corpus, as amended by this order, is hereby DENIED.
DONE AND ORDERED.NOTES
 As previously noted, though, the Medical Examiner would later hedge his assessment and acknowledge that the range of death could have occurred as many as four hours prior to discovery. See id. at 154 (J. Nesbitt, dissenting).
 The other claims can be summarized as follows:
Failure to object to the testimony of allegedly unqualified paramedic regarding the child's rigor mortis; failure to explore evidence that Marcella Gunderson committed the murder; failure to explore Marcella Gunderson's close relationship with the sheriff's office; failure to move for change of venue after a county employee urged jurors to "burn" the Petitioner; failure to excuse jurors; failure to object to introduction of coral rock as evidence of the murder weapon; failure to object to closing argument in which State urged the jurors to determine who was telling the truth; failure to proffer evidence that Petitioner got along with victim; failure to object to "rebuttal" evidence that although keys to the house had been stolen, the locks were then changed; failure to object to witnesses being in the courtroom during testimony; failure to object to evidence of threats by Marcella Gunderson's estranged husband; failure to confront evidence on the condition of the crib; failure to rebut evidence that Petitioner wanted to chop up the crib after the incident; failure to rebut evidence that Petitioner asked Marcella Gunderson to call his business after the incident; failure to object to misrepresentative photographs of the interior of the house; failure to rebut evidence that petitioner was swilling beer on the night of the murder; failure to investigate the actual cause of child's death; failure to object to felony murder instruction.
 Every circuit which has addressed the issue has recognized an equitable tolling defense to the AEDPA's limitations period. E.g. Davis v. Johnson, 158 F.3d 806, 810-11 (5th Cir.1998). The Eleventh Circuit has recently adopted the reasoning of these cases. See Sandvik, 177 F.3d at 1271.
 The State makes the point that since there is no Constitutional right to collateral counsel, Petitioner's reliance on counsel's advice cannot justify equitable tolling; as Petitioner phrases it, there is no such thing as an ineffective assistance of collateral counsel. However, Petitioner is not making an ineffective assistance of collateral counsel claim. He is making a plea for equitable relief based on poor advice of his counsel. It is irrelevant whether Petitioner had a right to collateral counsel.
 Dr. Bell did testify that in his experience it was not unusual to find contents in the stomach which had been ingested some six hours prior to death. It is somewhat difficult to reconcile Dr. Bell's observation with his earlier testimony that digestion will transpire in four hours for large meals, though "he has read" where emptying time may be prolonged for as long as six hours. Again, while undoubtedly Dr. Bell is qualified as an expert in pathology, his actual expertise regarding digestion is somewhat suspect.
 The court points out the jury's rejection of Petitioner's testimony merely to highlight that this was certainly a possibility under the law, and that a reasonably competent defense attorney would have recognized long before trial that such a rejection would likely prove fatal to the defense. The court is not engaging in the forbidden hindsight logic of, "the strategy was unreasonable because the jury rejected it."
 The Supreme Court has recently undertaken the issue of § 2254(d)'s import. See Williams v. Taylor, ___ U.S. ___, 119 S. Ct. 1355, 143 L. Ed. 2d 516 (1999) (granting certiorari in part to consider the Fourth Circuit's interpretation of § 2254(d)). The extent to which the Neelley decision will remain viable after the Supreme Court addresses the issue is unclear.
 Congress did not create its scheme of deference to bind federal courts inexorably to every state court decision, no matter how ill-conceived or half-realized. Congress carved out its exceptions to the general rule of deference carefully, to allow for review in cases where the state court erroneously applied or ignored federal law. See § 2254(1), (2). The rule of deference likewise should not extend into the field of blind assumptions on the state's behalf.
 This is merely a broad summary of the basis for granting the petition; the full analysis is found in the court's December 23rd order, and is hereby incorporated in its entirety for purposes of this issue.