We have considered whether the error in admission of defendant's first confession into evidence was reversible pursuant to the harmless error requirement set forth in Chapman v. California, 386 U.S. 18, 87 S. Viewing the evidence as a whole, applying those standards, we are satisfied that admission of defendant's September 2nd statement was harmless beyond a reasonable doubt and there is no reasonable possibility that the evidence obtained contributed to his conviction. Most of the evidence in the case, with the exception of his actual admission to the homicide, was gathered by law enforcement officers during the course of their investigation of Ms. Guida's disappearance.
Beginning with the first traveler's check, which was cashed in Abingdon, Virginia, defendant left a trail which led unerringly to him, up to and including his arrest, when he was found in possession of the vehicle she had leased on her fateful mission to Tennessee. If we were to find that both of the statements to authorities were inadmissible as evidence against him, he would still have to overcome the unsolicited admission made by him to Paul Carter while both men were confined in the Coffee County Jail. This statement, while being somewhat abbreviated, included all of the essential elements of the two admissions made to the authorities. Defendant's sentencing hearing, including the voir dire of the jury, consumed approximately two weeks, during which the jury members had the opportunity to observe him, his mannerisms, actions and conduct in the courtroom. Our review of the record leaves us with no reasonable doubt that the jury would have reached the same verdict without having heard the admission made by him which we have held should have been excluded from the record. They also had the opportunity to hear the mitigating circumstances proffered by him in evidence at the hearing.
The defendant received a fair trial without any constitutional entrenchment. Even though the jury cannot weigh the invalid aggravating factor against any mitigating factors, the jury can properly consider evidence of the circumstances of the crime and the character of the defendant in making an individualized determination of whether the death sentence is justified. At the second stage of the death penalty procedure, at which the jury is required to consider aggravating and mitigating circumstances in order to achieve particularized sentencing, the Court has not given sufficient guidance in the use of aggravating and mitigating circumstances. The Court has, instead, it seems to me, searched for some basis on which to affirm the sentence of death despite a misuse of aggravating circumstances, most often by a finding of harmless error.
In the present case, the evidence supporting the remaining aggravating factor of prior violent felony convictions is undisputed and overwhelming. In addition to the cold-blooded execution-style murder of Kennedy in Memphis, Tennessee, the defendant committed another similar cold-blooded execution-style murder in Oklahoma within twenty-four hours of Kennedy's murder. Less than thirty days later, he committed an armed robbery in Florida and later engaged in a shoot-out with police officers before his capture, for which he was convicted of attempted murder. A few years earlier, he had been convicted in Wyoming for armed robbery. Moreover, the prosecutor's argument strongly emphasized the defendant's violent felony convictions, with little emphasis on the armed robbery.
Furthermore, no additional evidence, nor any evidence that was not already properly before the jury, was introduced in support of the invalid aggravating circumstance. Finally, the mitigating evidence in this case focused on environmental conditions in childhood and psychological testing which was interpreted to demonstrate brain damage. This, in turn, according to the psychologist, impaired the defendant's judgment and ability to appreciate his conduct was wrong. Extensive medical testing, however, failed to show the claimed brain damage.
There was virtually no mitigating evidence relating to the good character of the defendant, other than the evidence offered to explain his anti-social behavior. First, this review provides a detailed report of published honeybee models that can be used by scientists to decide which published model might be suitable for their specific needs. Many processes within colonies, epidemiology and foraging are well understood and described in the models, and there are multiple feedback loops that regulate colony dynamics and can buffer the colony against changes in the environment.
However, recent colony losses suggest this resilience may not be powerful enough to withstand multiple pressures over time and space; we found no published model that coupled in‐hive colony dynamics with pathology and with foraging dynamics of bees in heterogeneous landscapes. Importantly, such a systems model could also be utilized by policy makers, land managers and beekeepers to forecast the effects of environmental change and implementation of different management options. Three examples are given if such a modelling tool was made readily available to a diversity of stakeholders. The relative benefits of different areas and locations of sown forage mixtures on the survival of large and small colonies could be examined in silico. In turn, this would ensure that advice given for agri‐environment schemes and resources spent on planting such mixtures are more likely to result in benefits to the local colonies.
The issue on harmless error analysis is whether the Court can conclude that beyond a reasonable doubt the invalid aggravating circumstance did not influence the jury in its determination that the sentence would be death. Accordingly, since the defendant did not request that the proof be admitted during the trial, and did not bring the issue to the trial court's attention until the motion for a new trial was filed, we conclude that this issue has been waived by the defendant. Defendant charges the trial court with error during the voir dire of the jury by not excusing a number of potential jurors for cause and in requiring him to exercise his peremptory challenges to exclude a number who were opinionated in favor of the death penalty.
He specifically points out juror, Michael Anderson, who he says had formed an expressed opinion, based on what he had read and heard, that the defendant, if guilty, should receive the death penalty. He then directs his complaint to a total of six jurors, one of whom was Michael Anderson. He says these jurors, when voir dired, unequivocally expressed the opinion that, once it was established that the defendant had committed first degree murder, they would definitely impose the death sentence.
Finally, all the relevant evidence in mitigation must be fully considered in applying harmless error analysis to guarantee that the defendant receives an individualized sentencing determination. As in the case of the remaining valid statutory aggravating circumstance, we examine both the quantum and quality of the mitigating factors presented by the proof. (Tenn. 1987), the State committed error by introducing evidence of another murder for which the defendant had not been convicted to establish the aggravating circumstances of mass murder. We concluded that the error was harmless beyond a reasonable doubt, because the record fully supported two other valid aggravators, and contained little evidence of mitigating circumstances.
During the sentencing phase, as in the guilt phase, the defendant sought to introduce testimony from Watson's Oklahoma attorney, and the defendant's Oklahoma attorney, that Watson said she lied during the preliminary hearing in order to escape the death penalty herself. The trial court, however, sustained the State's objection to the testimony on the grounds of relevance and hearsay, finding that the evidence did not go to mitigation, but instead was an attempt to re-litigate the innocence of the defendant. The defendant next asserts that it was error for the trial court to refuse to allow him to impeach Watson's Oklahoma preliminary hearing testimony with evidence that she recanted her testimony. At trial, the defendant sought to introduce proof from Watson's Oklahoma attorney that after the preliminary hearing in Oklahoma, Watson said she had lied and placed the blame on Howell in order to escape the death penalty herself. The State objected and the trial court agreed, ruling the evidence self-serving hearsay and inadmissible, even for impeachment purposes.
Among the jurors who defendant challenged as biased, Lawrence Cooper, was excused for cause, on grounds other than raised in this issue. An examination of the voir dire of these prospective jurors discloses that although they stated the death penalty to be the appropriate punishment for any first degree murder, these were, at best, statements of opinion made without knowledge of the law. Each of these prospective jurors made it clear that he or she would follow the law and consider all of the evidence and the factors relevant to sentencing before reaching a decision. Beyond that, none of these prospective jurors were forced upon defendant at trial and there is no claim that the jury who heard the case was not fair and impartial. We choose these categories because within‐hive dynamics and foraging are essential elements of honeybee dynamics, and varroa mites are generally believed to be an essential stressor.
Nevertheless, we also scanned the literature for models that address further stressors, including pesticides and pathogens. We restricted our search to models of single colonies and omitted population or metapopulation models. But see State v. Campbell, supra, in which improper evidence was admitted to support the invalid aggravating circumstance, yet the error was harmless beyond a reasonable doubt based on the record in that case.
After a careful review of the record, we conclude beyond a reasonable doubt that the sentence would have been the same had the jury given no weight to the invalid felony murder aggravating factor, and we affirm the sentence of death. Later, at the Tennessee trial, Howell's defense counsel tried to impeach the preliminary hearing testimony of Watson. He attempted to introduce testimony from Watson's Oklahoma lawyer and the defendant's Oklahoma lawyer that Watson later stated that she had lied during the preliminary hearing in order to escape the death penalty herself. The Memphis trial judge, however, sustained the State's objection that the evidence was hearsay and self-serving, and ruled the evidence inadmissible. Except for this attempt to impeach Watson's former testimony, the defendant presented no testimony on his behalf at the guilt phase of the trial. The last evidence introduced by the State was the redacted testimony of co-defendant Mona Lisa Watson given in April of 1988 at the defendant's Oklahoma preliminary hearing for the murder of Charlene Calhoun.
Watson agreed to testify against the defendant in return for a life sentence. Although she later decided not to testify at both the Tennessee trial and the Oklahoma trial, Watson was questioned on direct examination by the prosecutors at the Oklahoma preliminary hearing. She testified that she and the defendant had driven Interstate 40 from Memphis to Oklahoma in the Lynn Whitsett truck on November 1 and 2, 1987. Watson said they drank beer and shot cocaine on the trip, and when they got to Dell City, Oklahoma, she said they got off the interstate and stopped at an apartment complex to walk around. After walking around for a few minutes, Watson said Howell shot Charlene Calhoun, the two of them took her car, and they set fire to the Lynn Whitsett truck by igniting lighter fluid she had thrown on the front seat.
Then, Watson said she and the defendant drove back to Memphis in Calhoun's car, and thereafter to Florida on November 3, 1987. (Tenn. 1993), in which the aggravating circumstances were , previous felony convictions, and , felony murder. The defendant in Smith was not, in my view, death-eligible, because the record did not show that the killing was deliberate or intentional or accompanied by a conscious purpose of producing death or a conscious realization that death likely would occur. However, even with the felony murder aggravating circumstance eliminated, the Tennessee sentencing statute still includes in the class of death-eligible defendants accidental and unintentional murderers whose culpability is minimal. It still allows convictions for first degree felony murder of persons who killed accidentally or unintentionally and those who did not kill, did not intend to kill and did not intend that any person suffer any physical harm.
The statute still does not effectively limit the class of death-eligible defendants to those most deserving of death as punishment and, therefore, it violates the Tennessee constitutional prohibition against cruel and unusual punishment. The last argument raised by the defendant attacking the constitutionality of the death penalty is that Tennessee's statute fails to sufficiently narrow the class of death eligible defendants under the Eighth Amendment to the U.S. Constitution, and Article I, § 16 of the Tennessee Constitution, because all persons convicted of felony murder are eligible for the death penalty on the sole basis that they committed the underlying felony.
Our review of the record convinces us that the "fundamental error" rule of State v. Reece, supra, is not applicable to this case. In this case, the testimony about the defendant's other crimes was relevant to his guilt, and no hearsay problems were involved. The next issue we address is whether the trial court erred in redacting that part of Watson's preliminary hearing testimony which indicated that she and the defendant had been drinking and shooting cocaine before the Kennedy killing. The defendant contends that it was error for the trial court to exclude this testimony because it would have been relevant to the mental state of the defendant and his intent at the time of the murder. The redacted testimony about which the defendant complains was given by Watson at a later preliminary hearing held a week after the first preliminary hearing.
Defendant submitted all of these circumstances to the jury which rejected these theories of defense. Under the proof in this record, and the correct and proper instructions conveyed to the jury by the trial judge, the evidence supports the jury's finding of no mitigating circumstances sufficiently substantial to outweigh the aggravating circumstances found. Defendant objected to the testimony of Officer Reginal Hendricks on the basis that it was irrelevant to sentencing. Hendricks testified how defendant, who smelled of alcohol when stopped while driving the victim's car, led the police on a reckless, high-speed chase through the streets of Baltimore until he crashed the car into a light pole. Defendant continued to flee on foot through a housing project populated largely by blacks and, when apprehended, behaved violently and made racial remarks to the crowd observing his arrest.
He gave the officers his brother's name instead of his own when he was booked. Most of the foregoing testimony was admissible background evidence as well as proof establishing larceny of the automobile as an aggravating circumstance. Some details of his flight and capture, such as almost provoking a race riot, were less relevant and had the potential for prejudice.
However, in light of the strength of the State's case against him this was not reversible error. None of the foraging models were linked to a representation of real landscapes, characterized by mosaic of patches like arable fields or orchards providing nectar and pollen for a certain amount of time. This makes it impossible to explore the potential contribution of changes in land use and agricultural practice to colony losses. Some of these aspects are implemented in a model by Lonsdorf et al. . Although we did not include it into our review as it is a general pollinator model with the focus on pollination services, it would be of use for developing the landscape module of an integrated model.
The design of our own upcoming integrated model BEEHAVE will be based on the framework described above (Fig.2). It can be used to examine the effects of more stressors than other models . As a conceptual example of its use to study interactions between stressors, we can simulate the effect of different populations of varroa mites on colony survival in landscapes with differing floral availability. We identified eight colony models, 11 varroa models and 12 foraging models as being relevant for our review .
13 relevant publications were found in our collections that were not detected by the data base search (referred to as 'additional' in Table S1). We did not find references to further relevant models in the publications we reviewed. In Table S1 , we list and briefly describe further honeybee models found in other publications that were not included in our analyses, and we give reasons for their exclusion. We review existing honeybee models that are relevant to examining the effects of different stressors on colony growth and survival. Most of these models describe honeybee colony dynamics, foraging behaviour or honeybee – varroa mite – virus interactions.
We, therefore, affirm the conviction of first-degree felony murder and the sentence of death. The sentence will be carried out as provided by law on the 10th day of February, 1994, unless otherwise ordered by this Court or by other proper authority. Costs of this appeal are assessed against the defendant, Michael Wayne Howell.
Supreme Court first applied the Chapman, supra, harmless error analysis to federal constitutional errors occurring in capital sentencing proceedings. In that case, the constitutional error — admission of evidence in violation of the defendant's Sixth Amendment right to consult with counsel before submitting to a psychiatric examination designed to determine future dangerousness — was not held to be harmless. Finally, the defendant contends that the prosecutor's closing argument at the sentencing hearing requesting that the jury not feel sorry for the defendant because of his head injuries and telling them to overcome their sympathy for the defendant violated the Eighth Amendment. Examination of this argument in context reveals that the prosecutor was urging the jury not to return a life sentence based on the mitigating circumstances presented by the defendant. Based upon our review of the record, we conclude that the probative value of the evidence presented regarding the Oklahoma killing, and the Florida shoot-out, outweighed its prejudicial effect. As a result, we hold that the trial court properly admitted this evidence with respect to the defendant's identity and flight.
See Zagorski, 701 S.W.2d at 813; and Taylor, 669 S.W.2d at 698. While the defendant raised this issue as error in his motion for a new trial, our review of the record shows that Howell did not request the trial court to include this portion of the record of the Oklahoma proceeding in the proof. Our review of the record also shows that during the trial, the defendant included in the proof other parts of Watson's testimony regarding the defendant's drug use on the ride from Memphis to Oklahoma. To model honeybee population dynamics, all colonies in a sufficiently large area would need to be represented and swarming, robbing, drifting and the availability of suitable natural nest sites would need to be incorporated. Of the reviewed models, only HoPoMo represents colony division, but no landscape‐level population dynamics were simulated. Neighbouring colonies as a source of continuous mite invasion are taken into account by Calis, Fries & Ryrie and Vetharaniam & Barlow .
Current population models either keep the within‐hive colony dynamics extremely simple (e.g. Al‐Khafaji et al. 2009) or neglect it completely (e.g. Matis & Kiffe 2002; Mistro, Rodrigues & Ferreira 2005). To extrapolate to population models, including the interaction among different colonies, the strategy used in modelling metapopulations will be useful. We did not find published mechanistic models predicting the impact of bacterial or microsporidian (Nosema spp.) infection on a honeybee colony. Likewise, there were no models calculating the risks of limited genetic diversity on colony growth or the impact of many beekeeping practices, apart from some varroa treatments. Some of these potential stressors might be easily added to the suggested integrated model, changes in parameter values (e.g. mortality rates, foraging probabilities etc.). However, if they require substantial structural adaptions and interactions of different parts of the model, then it would be necessary to develop a specific new module.
Our review shows that robust and tested conceptual designs of processes in existing models are suitable to be used as building blocks in an integrated model. A cohort‐based colony model using daily time steps seems to be the best basic design for a new integrated colony model. The model should include nectar and pollen consumption and stores so that linking it to an explicit foraging model makes sense.
The model of Martin shows that it is important to describe the effect of varroa mites on honeybees in detail. It is not varroa infestation per se which drives a colony to extinction, but the viruses transmitted by varroa. A further insight from the varroa models is that timing of infestation and interaction with the honeybee cohort dynamics has a strong influence on varroa dynamics and chances of success of control measures. Hence, since honeybee cohort dynamics depends on weather, the current state of the colony and the incoming forage, it seems necessary to link varroa models to dynamic colony models rather than just imposing a static colony structure. (court concluded that the jury's consideration of two invalid aggravating circumstances was harmless beyond a reasonable doubt because there were three other valid aggravators and no mitigating evidence). After evaluating the relevant factors, we conclude that any doubt cast on Watson's reliability as a witness and her former testimony by impeachment evidence that she later made an unsworn, general statement that the testimony was false would have been minimal.
Accordingly, we hold that the trial court's failure to admit evidence that Watson later recanted her preliminary hearing testimony was harmless beyond a reasonable doubt. There is no doubt that in this case, Watson's former testimony was important to the prosecution's case as the only direct evidence that the defendant killed Kennedy. However, the corroborative proof of her Oklahoma preliminary hearing testimony was overwhelming and the prosecution's overall case was otherwise strong. Moreover, the jury was informed that Watson's testimony at the Oklahoma preliminary hearing was given in exchange for the charges against her being lowered to second-degree murder and a plea recommendation of life imprisonment. As a result of the foregoing, we conclude that defendant's Oklahoma counsel had a similar motive to defendant's Tennessee counsel to develop Watson's testimony about Alvin Kennedy's murder at the preliminary hearing in Oklahoma. Accordingly, we conclude that the admission of Watson's Oklahoma preliminary hearing testimony with respect to the murder of Alvin Kennedy in Memphis did not violate the hearsay rule nor the defendant's confrontation rights under the Sixth and Fourteenth Amendments to the U.S.
