STATE of Kansas, Appellee, v. Debora J. GREEN, Appellant.
Defendant Debora J. Green appeals the district court's decision denying her motion to withdraw her no contest plea to two counts of capital murder, one count of attempted capital murder, and one count of aggravated arson.
Underlying Facts and Procedural History
In the early hours of October 24, 1995, a fire destroyed the Prairie Village home of defendant and her estranged husband, killing two of their three children. Defendant and one of her daughters escaped the fire; defendant's husband was not at the home at the time.
On November 10, 1995, the Eastern Kansas Multi-County Task Force issued a report determining that the fire “was an intentionally set incendiary fire caused by ignition of a liquid accelerant applied throughout the structure on the main and second floors”; the fire had “multiple points of origin,” including a suspicious area of self-contained fire in the vanity of the master bathroom; “liquid accelerant pour patterns and an unusual magnitude of low burn” were identified; isoparaffins, associated with ignitable liquids, were detected in debris; and “all accidental and natural causes for the fire [such as an intruder, or gas or electrical causes] were eliminated.”
These events, coupled with a recent near-fatal ricin poisoning of defendant's husband, supported charges against defendant of two counts of capital murder, one count of attempted capital murder, aggravated arson, and one count of attempted first-degree premeditated murder.
On February 12, 1996, the district court judge ordered defendant to submit to a competency evaluation. Although Dr. Marilyn Hutchinson later expressed concerns about defendant's state of mind on the night of the fire, Hutchinson declared defendant competent to stand trial.
The Plea Hearing
On April 17, 1996, defendant agreed to plead no contest to all charges. In exchange, the State agreed not to seek the death penalty and recommended that her sentences run concurrent. The State submitted a 17-page proffer of the evidence it would submit at trial, which the district judge had the prosecutor read into the record. In addition to the information from the task force report on the cause of the fire, the following summarizes the evidence included in the proffer:
• Defendant and her husband, both medical doctors, had a troubled relationship. Defendant's husband had expressed his desire to divorce. They had separated, and the husband had moved out of their Missouri home. At some point, defendant and her husband considered reconciling and buying the Prairie Village home; defendant's husband withdrew his offer on the house after expressing uneasiness with resuming a marital relationship. Shortly thereafter, on May 21, 1994, the house they owned in Missouri was significantly damaged by a fire, and they did purchase the Prairie Village home.
• In July 1995, defendant's husband began an extramarital relationship with a woman named Margaret Hacker, and, in late July or early August 1995, he again expressed his intent to obtain a divorce. Defendant's behavior became erratic; she drank heavily; she was unable to supervise the children; she threatened herself and others; and she acted in other inappropriate ways.
• On August 4, 1995, defendant called her husband, saying she had left the residence and was walking the streets of Kansas City, hoping someone would kill her. On August 5, her husband came home to find her in the residence; she later told him that she had been hiding under a bed in the basement in an effort to make him worry.
• Beginning August 11, defendant's husband became ill, and as time progressed, his condition worsened. He was hospitalized on August 18, 1995, and his physician considered his condition life-threatening. After his condition stabilized, he was released to defendant's home on August 25, 1995. He was home for a few hours, when, after eating, he became violently ill again. He was hospitalized again, stabilized, and released on August 30. After going home, he became ill again on September 4, and his illness forced him back into the hospital, where he stayed until September 11. At the time, physicians were unable to pinpoint the cause of his illness, but they had not tested him for poisoning.
• Investigators determined that, before each of the defendant's husband's hospitalizations, he had ingested food served by defendant. After the fire, a sample of defendant's husband's blood was sent, among other places, to the Naval Research Laboratory in Washington, D.C. Testing occurred between November 17 and November 22, 1995; and experts ultimately concluded that his blood had been exposed to ricin, an extremely toxic substance that can be extracted from castor beans. Defendant had an undergraduate degree in chemical engineering.
• Defendant's behavior continued to be erratic after her husband's release in mid-September, and he remained in the home because he was concerned about defendant's ability to care for their three children. He ultimately attempted to have defendant committed to a mental institution, summoning police officers to transport her for a mental evaluation on September 25, 1995. Officers discovered her in bed, intoxicated. She was screened by Dr. Pamela McCoy, the emergency room physician, who stated that when defendant's husband came in, defendant spat at him, called him a “fuck hole” and told him “you will get the children over our dead bodies.” Defendant's husband showed McCoy defendant's purse, which he had found in the residence. It contained several vials of sodium chloride and packets of castor beans. Also in the purse was an Olathe Earl May Garden Center receipt dated August 7, 1995. Subsequent investigation revealed an address book with an entry in defendant's hand for an Earl May Garden Center in Olathe.
• Subsequent investigation also revealed that a second purchase of castor beans had been made on September 20, 1995, from a North Kansas City Earl May. The assistant general manager of the store confirmed defendant's identity and said she had special-ordered the packets. September was an unusual time to purchase castor beans, because their growing season was over. Defendant had told the manager the beans were for her child's science fair project. Investigators determined that none of the defendant's children were involved in such a project.
• On October 5, 1995, defendant's husband moved out of the home and into an apartment complex nearby. During the next few days, defendant continued to act in a bizarre manner, which included heavy drinking.
• On the evening of October 23, defendant's husband picked up two of the children; took them to a hockey game; and returned them to the Prairie Village home at 8:45 p.m. He then went to Hacker's house and was there until 11:30 p.m. He said he then drove to his apartment and called the defendant's residence. He and defendant argued. He told her he was concerned about her drinking and bizarre behavior, and about the possibility that she had tried to poison him. He told officers later that he told defendant she had better straighten up or he would call authorities; he also told officers that he was very angry during the conversation and that it ended abruptly about midnight.
• At 12:21 a.m. on October 24, the police dispatcher received a hang-up 911 call from defendant's residence; police and fire units were dispatched and arrived at 12:27 p.m. to a house fully involved in fire.
• At the scene, Corporal Steve Hunt was approached by a “panic stricken” young girl, determined to be Kate Farrar, who asked him to save her brother and sister, who were still in the house. Defendant's neighbors immediately suspected defendant was responsible for the fire. Defendant remained casual and nonchalant during interviews with police; she said she fell asleep in her room with the door closed; an alarm awakened her; she opened the bedroom door, saw smoke, closed the bedroom door, and went outside through a sliding glass door in her bedroom. She did not ask whether her two other children were alive or dead for at least an hour after her interview began. She said that, as she was exiting the house, she heard her son, Timothy, calling through the intercom. She told him to “stay in the house and let the professionals rescue you,” even though she knew he had crawled out of his bedroom window to the outside on numerous occasions in the past.
• Physical evidence revealed that defendant's bedroom door was open during the fire, which was inconsistent with her version of events. Defendant's hair was singed, which was also inconsistent with her version and consistent with use of accelerant. Discovered on the bed in defendant's bedroom after the fire was a book entitled “Necessary Lies”; its plot involved several children burning to death in an intentionally set house fire; defendant had gotten the book from the public library. Library records also revealed that defendant had recently checked out several books dealing with intrafamilial homicide.
At defendant's plea hearing, her counsel stated that the defense would certainly challenge some of the State's evidence, but the defendant understood that the material above would be presented to the jury if the case proceeded to trial. Defendant then read a prepared statement, saying:
“I am aware that the State can produce substantial evidence that I set the fire that caused the death of my children. My attorneys are ready willing and able to present evidence that I was not in control of myself when Tim and Kelly died.
“However true that may be, defending myself at trial on these charges would only compound the suffering of my family and my daughter, Kate. I love my family very much. I never meant to harm my children but I accept the fact that I will be punished harshly. I believe that it is best to end this now so that we can begin to heal from our horrible loss.”
Finding a sufficient factual basis for defendant's guilt on each count, the district judge accepted her plea and imposed a controlling hard 40 life sentence for one of the capital murder counts, ordering all of the additional sentences to run concurrent.
The Motion to Withdraw Plea
On March 22, 2004, defendant filed the instant motion to withdraw her plea as to all counts except the attempted first-degree murder of her husband, citing manifest injustice under K.S.A. 22-3210(d). She argued that evidence of new advances in the science of fire investigation, had they been known at the time of her plea, would have rendered the factual basis of the arson charge, and thus the capital murder and attempted capital murder charges, unreliable and insufficient.
Specifically, she argued that investigators conduct investigations in accord with the National Fire Protection Association's code, NFPA 921: Guide for Fire and Explosion Investigations, which is the de facto standard of care in the fire investigation industry. In 1995, the task force conducted investigations under the 1992 manual; that manual was updated in 2001, and again in 2004. She asserted that the new version of NFPA 921 suggested that several of the factors the task force had relied upon to conclude the fire was the result of arson were inaccurate and might have been attributable to other phenomena.
According to defendant, the fire's “multiple points of origin”-which is “almost always” indicative of arson-were more plausibly explained as “falldown” fire; the “pour patterns,” low burn patterns, and the speed and intensity of the fire were better explained by the now more fully understood phenomenon of “flashover” rather than application of an accelerant; and isoparaffins, although associated in 1995 only with a short list of ignitable fluids, were now associated with a long list of commonplace household products, innocently explaining their presence in the debris of the home.
Defendant acknowledged that the State's theory for the fire was “based on the best information available at the time,” and that her no contest plea was based on this same information. However, she argued, fire science and investigation had advanced so much over the past decade that the theory put forward by the State “cannot possibly be true.”
In a supplemental motion to set aside her plea, filed August 2, 2004, defendant reiterated her argument and added that the factual basis for her plea should be reviewed under a heightened reliability standard because she had faced the possibility of the death penalty.
The district court held a preliminary evidentiary hearing on October 12, 2004, to discuss discovery; the district court limited the plea withdrawal proceeding to “evidence that this could not have been an arson fire.”
In a second supplemental motion, filed January 3, 2005, defendant argued that the court should declare the plea bargain void or permit her to withdraw her plea because the death penalty had been declared unconstitutional in State v. Marsh, 278 Kan. 520, 102 P.3d 445 (2004).
The hearing on defendant's motion was held on January 10, 2005, before the same judge who had presided at the plea hearing. As a threshold matter, the prosecutor argued that the hearing should not go forward because the defendant's expert was not able to say that the fire was not arson. “Their expert is going to say this fire is an undetermined origin and that isn't enough ․ to reopen a case nine years after a defendant pled because you've got a new expert that's going to say well, I disagree with this part or that part but I can't tell you what caused this fire․”
Defendant argued that the central question before the court was whether reasonable doubt now infected the factual basis for her plea. In her view, the defense need not disprove arson, but need only prove the State could not have proved it beyond a reasonable doubt.
Defendant presented the expert testimony of Dr. Gerald Hurst. Hurst had reviewed the file associated with defendant's case, which included several hundred photographs, the police incident log, the task force and fire investigation reports, reports of interviews with firemen and first-responding officers, and complete sets of analyst drafting runs and laboratory results. He opined that since 1995 there had been advancements in the field of fire investigation, particularly in relation to a phenomenon called “flashover” burning. He said it is now known that flashover can cause several of the physical indicators upon which the task force had based its arson conclusion. In particular, “pour patterns” are no longer considered a reliable indicator of use of an accelerant if a fire has progressed to flashover.
“Flashover,” as that term is now used in fire investigation, occurs over a short period of time when the heat output of a burning object or a group of objects in a room generates a smoke layer that banks down from the ceiling and reaches a critical temperature-in the range of 500 to 600 degrees centigrade. At that point, radiation becomes so intense that it ignites every exposed combustible surface in the room. Before flashover, there is a hot smoke layer in the top of the room that is clearly divided from clear cool air below; at flashover, the division between hot and cold disappears; everything exposed, often in just a few seconds, springs into flame; and intense “post-flashover” burning occurs. The speed and intensity associated with flashover can result in the charring of baseboards and large burn holes in floors that resemble irregular, pool-shaped “pour patterns.” Under today's investigatory standards, Hurst testified, such patterns do not lead to a conclusion that accelerant has been applied. Rather, such evidence is more likely to support a determination that flashover occurred.
Hurst also testified that a more plausible explanation for what the task force had called a suspicious self-contained fire in a vanity drawer of the master bedroom-despite the lack of damage to the master bedroom or bath-came from “falldown,” i.e., embers from a burning beam that had fallen in the vicinity and ignited combustible fluids in the open drawer. Also supporting this conclusion, he said, were firefighter reports. The reports indicated firefighters entered the bathroom window and would have walked past the vanity into the bedroom. They showed only light smoke; the fire was in the hallway and starting to come in the bedroom door. Hurst believed this supported his theory that the vanity drawer fire began later.
Hurst suggested that a separate fire near a basement bar, which the task force had characterized as a separate point of origin attributable to an accelerant pour, was more likely to have been caused by one of the firefighters tracking embers from falldown a few feet away.
Hurst also suggested that positive testing for the presence of accelerants was likely to have come from common household products not used to set the fire. Accelerants discovered in a small closed container in the kitchen pantry were likely paint or mineral spirits. Hurst also suggested that, in the garage, which was not severely damaged, a labeled jug of Gulflite charcoal lighter fluid had been knocked over; it could have been tracked into the house by firefighters and created false positives.
Hurst's ultimate conclusion was that a lack of evidence of accelerant indicated the fire could have been accidental and that it progressed to flashover and full room involvement. Holes in the floor were not the results of liquid accelerant pour patterns but of post-flashover burn. Flashover, falldown, and human error provided a better explanation for the fire than arson. In his view, given the evidence, it was less likely that defendant had poured accelerant throughout the house and started separate fires in different areas, as the task force concluded. According to Hurst, a fire investigator cannot declare a fire to be the act of arson unless and until he or she has specifically eliminated all reasonably possible natural and accidental causes. If an investigator cannot do so, then the cause of the fire must be declared to be undetermined.
On cross-examination, Hurst was unable to state affirmatively that the fire was accidental, yet he maintained that accident could not be ruled out. Ultimately Hurst stated that he did not know what caused the fire. He acknowledged that the phenomenon of flashover was known at least by 1992 to the “leading experts in the country,” but said it was not widely known.
One of the “leading experts” referred to by Hurst, Dr. John David DeHaan, was called to testify at the motion hearing by the State. DeHaan had also reviewed the extensive file in the case. He discussed the areas of damage, noting that the most extensive damage centered around the middle of the house, primarily the rooms around the stairway, with much less damage to the master bedroom. Like Hurst, DeHaan testified that the evidence suggested flashover in much of the middle of the house.
DeHaan then discussed the interaction between flashover and pour patterns. He agreed with Hurst in certain respects. For example, if a floor covering is burned through in a localized area by a pour, then radiant heat produced as the room approaches flashover can accentuate that damage. But even if there was no accelerant poured, the intense combustion at floor level triggered during flashover can cause irregular puddles that mimic the presence of flammable liquid pour patterns. In essence, the effects of flashover can obscure the presence of pour patterns, or can create damage that looks like pour patterns, even if there were none.
DeHaan suggested that the only way to tell if accelerant was used is to detect it in some area protected from radiant heat and/or flashover combustion or, more reliably, in a room that has not flashed over, where vestiges of accelerant will probably be detectable if present at all. The evidence from defendant's house yielded two positive samples of accelerant in areas that had not flashed over. One was on carpet just inside the master bedroom door. Although a portion of the hallway outside the door had flashed over, the bedroom was undamaged. The sample contained an isoparaffinic petroleum product, consistent with charcoal starter fluid. DeHaan testified that, even if the charcoal fluid had been tracked in from the garage by firefighters, it would not have resulted in a positive lab test on the other side of the house; published experiments indicate such traces test negative after two or three steps.
DeHaan also testified that the isolated fire in the vanity drawer was not a likely result of falldown because the vanity itself was completely intact; there was no “communication” or “route” to that isolated fire. The only explanation was that a separate fire was started in the drawer.
Regarding the basement, DeHaan agreed with Hurst that the fire in the basement was small, that the room did not flashover, and that no accelerant was detected. There was no damage to the ceiling and no evidence of falldown. DeHaan identified particles of cloth and paper, suggesting that these may have been ignited to cause the fire in the basement.
DeHaan also testified about the fire dynamics of the defendant's house. The house was designed along a long hallway, with one large open area in the front and center where the stairway went up, perpendicular to the main floor hallway, to the second floor. Because fire naturally vents up and out, DeHaan testified, this open stairway would have acted as a chimney, and an accidental fire started at any point on the main floor would have flowed predictably, extending into the central entryway and up to the second floor. This fire, in contrast, was unnatural. There was damage at both ends of the main floor hallway. Given the home's size, layout, and the fact that it was occupied, DeHaan concluded the fire “had traveled to more places by the time [the occupants detected it and] public safety people arrived ․ than I could explain by any single source of fire traveling by normal means.”
DeHaan also concluded that protection marks, extending from the door jamb, clearly demonstrated the door to the master bedroom was open during the fire, which was, again, inconsistent with defendant's version of events.
Although DeHaan also agreed with Hurst that fire investigators must reject all alternative hypotheses before forming an opinion and that, if this cannot be done, a fire's cause must be categorized as “undetermined,” he ruled out accident and concluded that “this fire was deliberately ignited, multiple locations in the structure, based on all the evidence that was presented, and probably included the use of an ignitable liquid.”
On cross-examination, DeHaan acknowledged that the use of ignitable fluid was a probable conclusion. He could not estimate how extensively it was used and agreed it was not used throughout the house. However, he opined that some amount of accelerant was certainly used, and that direct ignition accounted for other points of origin.
DeHaan also acknowledged on cross that the reasons for his arson conclusion differed from those of the task force. Under today's fire investigation standards, which include improved understanding of flashover, he admitted that the task force was incorrect to the extent it relied exclusively on the existence of pour patterns to conclude that an accelerant was used.
The State also called David H. Campbell, a firefighter and veteran cause-and-origin investigator, who disagreed with Hurst that flashover inevitably negates preexisting pour patterns. Campbell also disagreed with both experts in that he believed one could rely on a pattern to determine whether it was caused by an accelerant pour. Campbell testified he could determine, by observing a pour pattern and the scene as a whole, whether the pattern was caused by flashover, falldown, or accelerant; “[a] pour pattern has a total different look to it than a flashover pattern does.”
Campbell also said that, even with his knowledge of flashover and falldown, he agreed with the task force observation that there were “ghost” patterns consistent with pouring of accelerant in the house entryway and up the stairs leading to the second floor. In addition, he testified that the patterns in the master bedroom revealed liquid burn, consistent with pouring of ignitable liquid that burned and ran back out to the hallway. While he could not rule out flashover to explain damage to the hallway, he testified that the evidence was more consistent with accelerant pour because there was not enough fuel in the form of furnishings or other combustible material or objects in the hallway to generate the heat required for flashover. Ultimately, Campbell opined that “the fire was intentionally set with the aid of combustible liquid and multiple fire sets.”
The District Court's Decision
In its memorandum decision denying the instant motion, the district judge set out the standard codified in K.S.A. 22-3210(d), which gives a district court discretion to permit withdrawal of a plea after sentencing and set aside a conviction in order “[t]o correct manifest injustice.” The judge agreed with defendant that a plea must have a “sufficient factual basis”; and that, in deciding that issue, the court “must be satisfied that all elements of the crime charged are present.” The district court determined that, as all parties had earlier agreed, there was “substantial evidence” supporting the charges at the time of defendant's plea.
The district court adopted defendant's suggestion that new evidence may form the basis for permitting withdrawal of a plea. However, he found defendant's “new evidence” insufficient. “[D]efendant contends that because of advances in the science of fire investigation the State's evidence could no longer support a conclusion that the fire was the result of arson. After three days of mostly expert testimony, it is clear that defendant's evidence falls considerably short of proving that contention.”
The judge further stated:
“There is substantial and compelling evidence to believe [defendant started the fire at her residence.] This evidence is not changed by a deeper understanding of the details or behavior of the fire once it was started. Defendant fails to show that the evidence as a whole would not support a finding of guilt if the case were to now go to trial or that the factual basis presented by the State now fails to support the plea entered in 1996. The factual basis ․ was, and remains, sound.”
The district judge also concluded that defendant's plea was knowingly, voluntarily, and intelligently made. He had taken care to ensure she understood the implications of her decision to plead no contest, that she understood the substantial evidence against her, and that she considered the harsh punishment she could face.
Finally, the district judge also addressed defendant's argument that Marsh, 278 Kan. 520, 102 P.3d 445, required the court to allow withdrawal of her plea. He ruled that our later Marsh decision striking down the death penalty did not affect defendant's case. “Had this defendant not entered into a plea agreement, a penalty of death could have been sought by the State, determined by a jury, and imposed by the court.” Thus defendant had received the benefit of her bargain: life in prison and concurrent sentences. Under all of these circumstances, there was no manifest injustice justifying withdrawal of defendant's plea.
On this appeal, defendant withdrew her claim based on Marsh, 278 Kan. 520, 102 P.3d 445, in light of the United States Supreme Court decision in Kansas v. Marsh, 548 U.S. 163, 126 S.Ct. 2516, 165 L.Ed.2d 429 (2006), which reversed our decision. Defendant also withdrew a cumulative error argument.
As a preliminary matter, defendant cites, inter alia, Boykin v. Alabama, 395 U.S. 238, 89 S.Ct. 1709, 23 L.Ed.2d 274 (1969), for the proposition that we should review the factual basis supporting her plea under a heightened reliability standard because she was facing the death penalty at the time of her plea. Defendant is correct that, in the context of a capital sentence, this court has required a heightened degree of reliability. See State v. Kleypas, 272 Kan. 894, 1036, 40 P.3d 139 (2001), cert. denied 537 U.S. 834, 123 S.Ct. 144, 154 L.Ed.2d 53 (2002); see also State v. Bethel, 275 Kan. at 457-58, 66 P.3d 840 (2003). However, where no sentence of death was imposed, this court is not required to review a case under this more stringent standard. See Bethel, 275 Kan. at 457-58, 66 P.3d 840 (“[b]ecause of the State's agreement not to pursue the death penalty, [defendant] is not directly affected by it and cannot raise issues” concerning its constitutionality). On this review, defendant is entitled to no protections beyond those set forth in K.S.A. 22-3210 and interpreting case law, and those required by due process.
As this court has often stated, motions to withdraw plea are governed by K.S.A. 22-3210(d), which reads:
“A plea of guilty or nolo contendere, for good cause shown and within the discretion of the court, may be withdrawn at any time before sentence is adjudged. To correct manifest injustice the court after sentence may set aside the judgment of conviction and permit the defendant to withdraw the plea.”
Under this section, the decision to deny a motion to withdraw a plea, even after sentencing, lies within the discretion of the district court. That decision will not be disturbed on appeal absent a showing of abuse of discretion, and the defendant bears the burden of establishing it. Judicial discretion will vary depending upon the character of the question presented for determination. Generally a district judge's decision is protected if reasonable persons could differ about the propriety of the decision, as long as the decision was made within and takes into account any applicable legal standards. An abuse of discretion may be found if a district judge's decision goes outside the framework of or fails to properly consider statutory limitations or legal standards. State v. Shopteese, 283 Kan. ----, Syl. ¶ 2, 153 P.3d 1208, 2007 WL 776811 (2007); State v. Edgar, 281 Kan. 30, 38, 127 P.3d 986 (2006).
In evaluating a post-sentencing motion to withdraw a plea, the district court should consider: (1) whether the defendant was represented by competent counsel; (2) whether the defendant was misled, coerced, mistreated, or unfairly taken advantage of; and (3) whether the plea was fairly and understandingly made. See State v. Bey, 270 Kan. 544, 545, 17 P.3d 322 (2001).
Defendant's legal position relies on two arguments that merit discussion. First, she argues that new evidence can require post-sentencing withdrawal of a plea. Second, she argues that she need only demonstrate the current existence of reasonable doubt in order to undermine the past factual basis for her plea.
With regard to her first argument that new evidence can require post-sentencing withdrawal of a plea, we note first that new evidence is not among the explicit considerations in K.S.A. 22-3210(d), even as supplemented by case law and due process requirements. However, we have discussed the withdrawal of a guilty plea based on newly discovered evidence.
In State v. Walton, 256 Kan. 484, 489, 885 P.2d 1255 (1994), Robert S. Walton alleged that evidence discovered after the entry of his plea exonerated him, thus providing “good cause” under K.S.A. 22-3210(d) for granting his presentencing motion to withdraw. The district court conducted a full hearing, considered all the evidence, and concluded that Walton's plea was informed and voluntary. Furthermore, because the new evidence did not exonerate Walton, there was no basis for withdrawing the plea.
More recently, in Bey, 270 Kan. at 558-59, 17 P.3d 322, the defendant sought to withdraw his plea, arguing new evidence allowed him to do so. He pointed to statements allegedly made by the jail mate of a codefendant, which implicated the codefendant rather than the defendant in the victim's shooting death. The district court considered the statements and determined that they did little to exonerate the defendant. On appeal, the defendant urged this court to draw an analogy between his motion and one for new trial. We did not clearly adopt his invitation on the way to affirming, Bey, 270 Kan. at 557-59, 17 P.3d 322, but we acknowledged that there must be a factual basis for a plea and that a district court determining whether such a factual basis exists must establish that all elements of the charged crime are present. Bey, 270 Kan. at 546, 17 P.3d 322 (citing State v. Shaw, 259 Kan. 3, Syl. ¶ 1, 910 P.2d 809  ).
It is obvious that, if new evidence disproves an element of a crime, then the factual basis for a guilty or nolo contendere plea to the charge of committing that crime is undermined. It is a defendant's burden to prove that the factual basis of a plea is so undercut by new evidence that the prosecution could not have proved its case beyond a reasonable doubt. In such a situation, the court may permit withdrawal of the plea and may set aside the resulting conviction, because doing so corrects manifest injustice under K.S.A. 22-3210(d) and comports with due process.
Defendant's further argument that she need only demonstrate the current existence of reasonable doubt in order to undermine the past factual basis for her plea is simply without merit. Even if we assume for the moment that the evidence adduced at the hearing on her motion to withdraw plea was sufficient to establish the current existence of reasonable doubt, at the time the district judge accepted defendant's plea, his charge was to determine the existence of a factual basis for it at that time and in that place. He was not required to foretell the future and anticipate its effect on the fire science underlying the task force conclusion.
Defendant's argument reflects a basic misunderstanding of the plea process. Entry of a plea of guilty or nolo contendere necessarily implies acknowledgment by all concerned-the defendant, the State, and the court-that a jury could go either way and that a risk-benefit analysis has taken place on both sides. The prosecution and the defense have something to gain and something to lose in any plea bargain. They make their peace with the trade-off in exchange for reducing the uncertainty of their situation. Instead of leaving it up to a jury to acquit or convict, to recognize or fail to see any reasonable doubt that may exist, they cut a deal. The requirement that a judge confirm the factual basis of a plea is not a mandate that he or she sit in place of a jury and evaluate reasonable doubt as a matter of law. It merely requires the judge to ensure that the State has enough proof to support each element of the crime. It is not necessary that the proof be particularly persuasive, even at that time and in that place, much less that it will stand the test of time and advancing human understanding. It need only be sufficient for a reasonable factfinder to arrive at a guilty verdict.
On the facts of this case, we conclude that the district court did not abuse its discretion in denying defendant's motion to withdraw her plea. The district court appropriately considered whether defendant was represented by competent counsel; whether she was misled, coerced, mistreated, or unfairly taken advantage of; and whether her plea was fairly and understandingly made. See Bey, 270 Kan. at 545, 17 P.3d 322. Relevant to these standards, we observe in particular that defendant was represented by three very experienced attorneys; she was not misled about the quality of the State's evidence against her; and the mere existence of the death penalty as a potential penalty for the offenses with which she was charged did not amount to coercion. Further, her articulate colloquy at the plea hearing demonstrated that her plea was fairly and understandingly made.
Most important, however, in view of defendant's specific argument, we hold that defendant has not brought forward new evidence. At best, she has developed a competing interpretation of old evidence. Hurst's testimony about advances in fire science and their potential effect on the reasoning and conclusion of the task force did not disprove an element of any of the crimes on which defendant entered her plea. DeHaan's and Campbell's testimony demonstrated that arson was still very much alive as an explanation for the fire. It was not ruled out; it was merely challenged.
In addition, we are compelled to note that defendant's focus on advances in fire science ignores the inescapable. The task force conclusion was far from the only evidence supporting the aggravated arson and other charges. The State's proffer included an abundance of other evidence pointing to the defendant's guilt. Its largely circumstantial nature would not have detracted from its value in support of any ultimate conviction. See State v. Dixon, 279 Kan. 563, 621, 112 P.3d 883 (2005); State v. Holmes, 278 Kan. 603, 632, 102 P.3d 406 (2004).
Given all of the above, we hold that defendant did not meet her burden to demonstrate that the factual basis for her plea is so undercut by new evidence that the prosecution could not have proved its case beyond a reasonable doubt. There was no manifest injustice to correct, and the district judge did not abuse his discretion in so ruling.
The opinion of the court was delivered by BEIER, J.: