Reset A A Font size: Print

Supreme Court of Kentucky.

Robert YELL, Appellant, v. COMMONWEALTH of Kentucky, Appellee.

No. 2006-SC-000327-MR.

Decided: December 20, 2007

Susan Jackson Balliet, Assistant Public Advocate, Department of Public Advocacy, Frankfort, KY, Counsel for Appellant. Gregory D. Stumbo, Attorney General, David W. Barr, Assistant Attorney General, Office of Criminal Appeals, Frankfort, KY, Counsel for Appellee.


Robert Yell appeals his conviction for First-Degree Arson, First-Degree Assault, Second-Degree Manslaughter, and Second-Degree PFO for intentionally causing the house fire which killed his two-year-old son and severely injured his eleven-month-old daughter.   Upon review of Yell's claims of error, we adjudge there was no error that warranted reversal.   Hence, we affirm.

In 2004, Robert Yell was living in a trailer in Russellville with his girlfriend, April Carpenter, their two children, Saralynn Yell, eleven months old, and Cameron Yell, two years old, and April's two children from a prior relationship, Zachary Carpenter and Nicholas Carpenter.   On September 11, 2004, Robert, April, Saralynn, Cameron, Zachary and Nicholas went to a neighbor's house for the afternoon.   The children played outside with the children of the neighbors, Lindsey Bromm and Donald Powell, while Robert and April proceeded to drink a large amount of alcohol and smoke marijuana.   April and Robert became very intoxicated, and at some point, Robert became belligerent with Donald Powell and they began arguing.   Robert tried to get April to leave, but April did not want to go home.   April and Robert then began arguing.   Sometime before 5:00 p.m., Robert, April, and all four children went home to their trailer.

According to April, once they got back to the trailer, she and Robert got into an argument.   April testified that she and Robert fought frequently because she recently had an affair with another man when Robert was in jail.   The fight escalated and Robert began choking her.   April ultimately freed herself and ran back to Lindsey Bromm's house with Zachary and Nicholas following behind her.   April arrived at Bromm's house at approximately 7:00 p.m. After staying there for twenty to forty minutes, April wanted to go back to her trailer to get the other two children.   When April began walking back to the trailer, she noticed smoke and realized the trailer was on fire.   She ran into the burning trailer and retrieved Saralynn from the front room and handed her off to Officer Ron Mills of the Russellville Police Department.   April attempted to go back into the burning trailer to get Cameron, but the smoke and fire were too bad.   The police refused to let April go back into the trailer again.

Ultimately, Cameron was pulled from the blazing trailer by firefighters, but, by that time, he had already died of smoke inhalation.   Saralynn sustained third-degree burns to 50% of her body and extensive respiratory damage as a result of smoke inhalation.   The evidence established that Saralynn suffered permanent, painful injuries as a result of the fire that have necessitated protracted hospitalization and numerous surgeries.

As a result of the events of September 11, 2004, Robert was indicted on October 28, 2004 on one count each of First-Degree Arson, Capital Murder, Attempted Murder, Fourth-Degree Assault, Resisting Arrest, Menacing, Third-Degree Terroristic Threatening, Public Intoxication, Third-Degree Assault, Second-Degree Disorderly Conduct, and for being a First-Degree Persistent Felony Offender (PFO I).   On February 8, 2006, Robert attempted to enter a guilty plea pursuant to a plea bargain with the Commonwealth whereby he agreed to plead guilty to Third-Degree Arson, two counts of First-Degree Assault, one count of Third-Degree Assault, and PFO II in exchange for a recommendation of a twenty-five year sentence.   After determining that the plea was actually an Alford plea, the court refused to accept the plea because it did not usually accept Alford pleas.   Consequently, a jury trial was held from February 10, 2006 through February 22, 2006.   Robert was found guilty of First-Degree Arson, First-Degree Assault, Second-Degree Manslaughter, and PFO II. Robert was also convicted of two misdemeanors, Fourth-Degree Assault and Disorderly Conduct.   Robert was sentenced to:  26 years on the Arson conviction;  20 years (10 years enhanced to 20 years by the PFO II) on the Manslaughter conviction;  26 years (20 years enhanced to 26 years by the PFO II) on the First-Degree Assault;  12 months on the Fourth-Degree Assault;  and 90 days on the Disorderly Conduct.   The sentences for the Arson and First-Degree Assault convictions were to run consecutively, with the remaining sentences to run concurrently, for a total of 52 years imprisonment.   This matter of right appeal followed.


Robert argues that the trial court erred in allowing the Commonwealth to introduce evidence that an accelerant detection dog had alerted to the presence of ignitable liquids at the trailer fire scene.   In a pre-trial suppression motion, the defense objected to the admission of this evidence on grounds that there was no scientific support for its admissibility and no legal authority in Kentucky for its admissibility.   Although the motion did not cite Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), or request a Daubert hearing, the trial court held a self-described Daubert hearing prior to trial in this case.   The hearing, held on January 3, 2006, elicited the following facts relative to the use of the accelerant detection dog, PJ, in investigating the fire scene in this case.

PJ was assigned to National Certified Fire Investigator Buster Cannon by the Federal Bureau of Alcohol, Tobacco and Firearms (ATF) in 2002.   Cannon established that he had twenty-three years of experience as a policeman and firefighter.   At the time Cannon received PJ, she had already been trained and imprinted in accelerant detection.   Early in 2002, Cannon and PJ successfully completed a five-week course in canine accelerant-detection training with the ATF. Both Cannon and PJ have completed annual forty-hour re-certification programs for 2003, 2004, and 2005 and are subject to annual testing by the ATF. Since 2002, Cannon and PJ have worked approximately 200 fire scenes together.   PJ is trained by Cannon twice a day and reports of her training are submitted monthly to the ATF. PJ detects ignitable liquids and trains with a food/reward system whereby Cannon gives PJ kibbles when she correctly alerts (sits at the site) where ignitable liquids are detected.   Cannon trains PJ with an array of flammable liquids at varying levels.   During her training sessions, PJ may be exposed to as little as one micro liter (roughly one-half of an eye dropper) to as much as 15 micro liters of a given accelerant.   Cannon testified that samples that a dog has alerted to will sometimes test negative in a laboratory for the presence of accelerants because the sample may contain a level of accelerant too small for a laboratory to detect.   Cannon explained that the accelerant may be consumed in the fire or may evaporate before collection and testing.   Cannon testified that PJ's false alerts may be as high as 5%. Through his experience working with PJ, Cannon has learned that if it is a true alert, PJ will get very excited and he will not be able to pull her off of the site of the alert.   If Cannon suspects it is a false alert (by her nonchalant behavior), PJ will surrender the spot without much resistance.

Before going into a fire scene, PJ is calibrated by detecting a drop of accelerant that Cannon places away from the fire scene.   When PJ properly alerts to the accelerant, she is rewarded and then taken into the fire scene.   At the fire scene, controls are used.   Samples are collected from the areas where PJ alerts and the samples are placed in individual containers.   A control sample containing no accelerant is also placed in a container.   All the containers are then taken away from the fire scene, and PJ is taken past them.   In the instant case, PJ alerted to all the samples taken from the scene and did not alert on the control sample.

On January 9, 2006, the trial court issued its order allowing the evidence regarding PJ's alerts at the fire scene in the present case. The trial court found that Cannon's testimony established that using properly trained dogs was a reliable method for determining the presence of accelerants at a fire scene, and that PJ was adequately trained to detect accelerants.   The court went on to say that the evidence that PJ had alerted to the presence of ignitable liquids at the scene of the fire in this case would be helpful to the jury, and that the jury was capable of giving appropriate weight to such evidence.   The trial court also found that due to his experience with PJ, Cannon could generally tell if PJ's alert was a false alert or a true alert.

At trial, Cannon testified that after arriving at the fire scene on the morning after the fire and calibrating PJ, PJ went into the trailer and alerted to six locations for the presence of ignitable liquids.   Contrary to his earlier ruling, the trial court allowed Cannon at trial to identify the locations of the six alerts in the trailer.   As stated above, PJ again alerted to the presence of accelerants in the six separate samples taken from the scene, but did not alert on the control sample containing no accelerants.   However, the testimony of Kenneth Rider from the Kentucky State Police forensic lab established that all six samples tested negative in the laboratory for the presence of ignitable liquids.

On appeal, Robert asserts that there is no legal authority in Kentucky for allowing evidence of accelerant detection by a dog pursuant to Daubert.   Robert argues that allowing the evidence was error in this case because PJ's alerts were unconfirmed by the laboratory testing of the samples, and there was no scientific validation of PJ's ability to detect accelerants at levels lower than detectable in a laboratory analysis of the samples.   The Commonwealth argues that Daubert does not apply in canine accelerant detention cases because the dog's ability to accurately detect accelerants through its sense of smell is not dependent on scientific explanation.

 A trial court's determination on the admission of expert scientific testimony pursuant to Daubert is reviewed under an abuse of discretion standard.  Miller v. Eldridge, 146 S.W.3d 909, 914 (Ky.2004).   The trial court's decision will be overturned only upon a showing that the ruling was “arbitrary, unreasonable, unfair, or unsupported by sound legal principles.”   Goodyear Tire & Rubber Co. v. Thompson, 11 S.W.3d 575, 581 (Ky.2000).

After submission of the briefs in the instant case, this Court on August 23, 2007, rendered its opinion in Debruler v. Commonwealth, 231 S.W.3d 752 (Ky., 2007) (final as of September 13, 2007), wherein we addressed the issue of whether canine scent tracking was subject to reliability analysis under Daubert.   In Debruler, a child was abducted and taken to a vacant house from which she was fortunately able to escape.   In investigating the crime, the police brought two K9 unit dogs to the vacant house, along with two articles of the suspect's clothing.   Both dogs tracked the scent of the suspect's clothing through the house, and evidence of this scent tracking was ultimately admitted in the suspect's kidnapping trial.   This Court held that Daubert does not apply to such evidence because canine scent tracking is not grounded in any scientific technique, theory or methodology.  Id. at 756. Rather, it is based on the dog handler's personal observations of the dog's actions relative to his experience with and training of the dog.  Id. This Court also rejected the appellant's claim that canine scent tracking required a Daubert reliability analysis because it involved technical or specialized knowledge.   Id.;  see Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 141, 119 S.Ct. 1167, 1171, 143 L.Ed.2d 238 (1999).

Here, we are not persuaded that the Daubert analysis would accurately assess the reliability of canine scent tracking testimony. Contrary to Appellant's assertions, the factors identified in Daubert provide little guidance as to the reliability of non-scientific, experience-based knowledge such as canine scent tracking.   More specifically, canine scent tracking is not a technique amenable to peer review or scientific standards and testing.   Rather, it concerns the behaviors of the dog and the meanings of those behaviors, a knowledge acquired through experience and training.   For this reason, foundational evidence of the canine's scent tracking record;  the qualifications of its handler, its training and history provide far more insight into the general reliability of the testimony than a Daubert analysis.


 In the hearing in the present case, we believe the Commonwealth satisfied the foundational requirements required in Debruler.   Cannon testified to his qualifications for being a dog handler and to his ATF certification, testing and training record in accelerant detection with PJ.

 As to Robert's claim that the evidence of PJ's alerts to the presence of accelerants at the fire scene was unreliable because it was contradicted by the negative results of the lab testing of the samples, we believe that there was a sufficient showing of reliability to allow the evidence despite the negative laboratory results.   As noted above, Cannon testified that through his training with PJ, he has established that PJ is able to accurately detect the presence of as little as one micro liter (one-half of an eyedropper) of an accelerant in a control sample.   Kenneth Rider of the KSP forensic laboratory, who tested the samples, testified that in order for a lab to detect the presence of an accelerant, there needs to be 15-20 parts per million of accelerant in the sample.   Both Rider and Cannon testified that simply because a sample tests negative for ignitable liquids in a laboratory does not mean that an accelerant was not used in the fire.   Rider and Cannon explained that ignitable liquid can be consumed in the fire or evaporate, leaving behind too small of an amount of accelerant to be detected in laboratory testing.   Cannon testified to two particular fires where samples from areas where PJ had alerted tested negative for accelerants in the lab, but it was later confirmed from confessions of the defendants that accelerants had been used in the particular locations where PJ had alerted.   The controls employed by the handler-the calibration of the dog before going into the scene and re-introducing the dog to the samples along with a non-accelerant control sample away from the fire scene-also help to insure the reliability of the evidence.   As recognized by the trial court, the jury heard the evidence of the negative lab results of the samples and was free to weigh this evidence against the evidence of PJ's alerts to accelerants in the trailer.   This was but one piece of evidence that could assist the jury in determining whether this fire was intentionally set.   As shall be discussed below, PJ's alerts to the presence of accelerants in the trailer was not the only evidence that this fire was the result of Arson.


Robert argues that there was insufficient circumstantial evidence that he intentionally started the fire in this case (KRS 513.020), which was the basis for the Manslaughter and First-Degree Assault convictions for causing the death of Cameron and serious physical injury to Saralynn in the fire.   Thus, he maintains that the trial court erred in denying his motion for directed verdict on those three charges.  KRS 513.020(1) provides:

A person is guilty of arson in the first degree when, with the intent to destroy or damage a building, he starts a fire or causes an explosion, and;

(a) The building is inhabited or occupied or the person has reason to believe the building may be inhabited or occupied;  or

(b) Any other person sustains serious physical injury as a result of the fire or explosion or the firefighting as a result thereof.

 Our standard of review on a motion for directed verdict is set forth in Commonwealth v. Benham, 816 S.W.2d 186, 187 (Ky.1991) as follows:

On motion for directed verdict, the trial court must draw all fair and reasonable inferences from the evidence in favor of the Commonwealth.   If the evidence is sufficient to induce a reasonable juror to believe beyond a reasonable doubt that the defendant is guilty, a directed verdict should not be given․

On appellate review, the test of a directed verdict is, if under the evidence as a whole, it would be clearly unreasonable for a jury to find guilt, only then the defendant is entitled to a directed verdict of acquittal.

“The test is the same when the only evidence of guilt is circumstantial.”  Ratliff v. Commonwealth, 194 S.W.3d 258, 267 (Ky.2006) (citing Bussell v. Commonwealth, 882 S.W.2d 111, 114 (Ky.1994) and Nugent v. Commonwealth, 639 S.W.2d 761, 763-64 (Ky.1982)).

 As to the evidence that the fire was intentionally set, besides the dog's alerts to the presence of accelerants, David West (a fire investigator for KSP), Al Gregory (Assistant Fire Marshall), Samual Flowers (Kentucky Fire Marshall), and Buster Cannon all testified that the burn pattern of the fire pointed to Arson.   They testified that the low, deep (all the way through the floor), v-shaped burn patterns and the fact that these areas of deep burn were not connected were indicative of a fire that was intentionally set.   Flowers explained that fires usually burn upward, but when an accelerant is used, a fire burns low and deep.   Flowers and West both testified that their investigation of the fire scene revealed that there were three separate fire origins in the trailer that were not connected at floor level.   This coupled with the fact that there was no evidence of any electrical fire, fire from the stove, or act of God led them to the conclusion that the fire was intentionally set with a combustible liquid.   Cannon also testified that there was a shiny coating in the areas of deep burn, which is consistent with an accelerant being used.   Finally, West testified that charcoal lighter fluid was found in the trailer, which could have been used to start the fire, and a lighter was found in the yard behind the trailer the day after the fire.

 As to the evidence that Robert was the one who started the fire, there was evidence that Robert was angry at April on the day in question and was at the trailer with Cameron and Saralynn at the time of the fire.   April testified that she and Robert had gotten into a fight that afternoon about her affair that culminated in Robert choking her.   When she left with Zachary and Nicholas, Robert was at the trailer with Cameron and Saralynn.

Kay Lyons, who had taken Robert to the liquor store before the fire, also testified that April and Robert were fighting on the day of the fire.   Lyons testified that after she brought Robert back to his house, Robert asked her to take Zachary and Nicholas with her.   Lyons replied that she would take all four children.   Robert indicated that the other two children were taken care of and were in the house asleep.   When she refused to only take Zachary and Nicholas, Robert got angry and took out a knife and said, “you don't know what I can do to you.”   At that point, Lyons went to the police station to report Robert's threat.

Robert's own testimony was consistent with the evidence that he was at the trailer at the time of the fire and had the means to start the fire.   Robert testified that after Lyons dropped him off, he went and got the children from the house of Charlotte Bellor, a neighbor who was friends with April.   The last thing he remembered was coming back to the trailer with more alcohol and sitting in his chair in the trailer.   After that, Robert maintains that he blacked out and remembers nothing until he was at the sally port at the jail later that night.   Robert also testified that he usually carried a lighter on his person and that he likely had one on the night of the fire.

The first police report regarding the events of that evening was made by Charlotte Bellor.   Bellor called 911 to report that Robert was drunk and trying to kick her door in.   In that call, Bellor also reported that she could see smoke and fire coming out of Robert's trailer.

The next time Robert was seen was at the home of Tina Maskin, another neighbor of Robert's.   Maskin testified that Robert came to her home in a very intoxicated state.   Maskin allowed him to come in the house and, while seated on her couch, she observed that he kept flicking a lighter and playing with a pocket knife.   Maskin told her children to lock themselves in the bathroom.   Maskin then noticed that the police were at Robert's home and advised Robert of this.   Robert went into Maskin's bedroom and tried to hide under her bed.   When Maskin told him that his trailer appeared to be on fire, Robert responded that April had better have gotten Cameron and Saralynn out.

When Officer Ronald Mills of the Russellville Police Department responded to the fire, Robert was at the scene.   According to Mills, Robert kept repeating the statement, “The bitch better have gotten Cameron and Saralynn out.”   And, when Robert was taken into custody, he made the statement multiple times to Officer Chad Eggleston and Officer Ed Higgins, “If that bitch had done to you what she did to me, you would have done the same fucking thing.”

From our review of the evidence in the light most favorable to the Commonwealth, there was more than sufficient evidence to induce a reasonable juror to believe beyond a reasonable doubt that Robert intentionally set the fire in the trailer.   The Commonwealth presented evidence that the fire was intentionally set, that Robert was at the trailer at the time of the fire, had a motive to start the fire (anger at April), had the means to start the fire (lighter and lighter fluid), and that the trailer was occupied by other persons who sustained serious physical injury in the fire.   Additionally, there was the evidence of the statement made by Robert to police after the fire that could be viewed as a confession-“If that bitch had done to you what she did to me, you would have done the same fucking thing.”   And contrary to Robert's assertion that finding him guilty of arson required a pyramiding of inferences in violation of Pengleton v. Commonwealth, 294 Ky. 484, 172 S.W.2d 52 (1943), the inference that the fire was intentionally set and the inference that Robert set the fire were not dependent on one another.   As can be seen from our analysis above, there was a completely separate body of evidence supporting the reasonable inferences that the fire was the result of arson and that Robert was the one who set the fire.


 Witness Lindsey Bromm was subpoenaed to testify on behalf of the Commonwealth pursuant to a summons served on her in Missouri, where she resided at the time of trial.   It is undisputed that the procedural requirements of KRS 421.250 to procure her attendance were not followed.   During the trial, prior to her testimony, the Commonwealth informed the court during a bench conference that there was an outstanding felony warrant for her arrest on a charge of Theft By Unlawful Taking from when Bromm previously resided in Kentucky.   The court asked the attorneys how they wanted the matter to be handled.   Defense counsel did not express any preference.   The prosecutor asked that Bromm be arrested after her testimony, but left the issue of her bond to the court's discretion.   Defense counsel responded, “that's fine.”   The court then called Lindsey Bromm up to the bench (out of the hearing of the jury) and informed her of the outstanding bench warrant for her arrest in Kentucky and of the charge against her.   The judge told her that he was releasing her on her own recognizance on the condition that she show up on Monday to testify at Robert's trial and show up to answer to the charges against her.   Bromm agreed.   The court thereupon asked the bailiff to take Bromm for her bond interview.   Subsequently, on February 13, 2006, Bromm testified for the Commonwealth in this case.

On appeal, Robert claims that informing Bromm that she would be arrested and then having her arrested violated KRS 421.260(1) which provides:

If a person comes into this state in obedience to a summons directing him to attend and testify in this state he shall not while in this state pursuant to such summons be subject to arrest or the service of process, civil or criminal, in connection with matters which arose before his entrance into this state under the summons.

 Robert contends that such violation constituted structural error that undermined the fundamental framework of the trial, because Bromm was essentially threatened with prosecution and thus there was a high risk that her testimony was prejudiced in favor of the Commonwealth.  “ ‘[A] structural error’ affects the entire framework of the trial and therefore defies harmless error analysis.”  Quarels v. Commonwealth, 142 S.W.3d 73, 81 (Ky.2004) (quoting Arizona v. Fulminante, 499 U.S. 279, 309, 111 S.Ct. 1246, 1265, 113 L.Ed.2d 302 (1991)).

We agree that under KRS 421.260(1), it was error to have Bromm arrested while she was present in Kentucky to testify as a witness in this trial.   However, there was no evidence that Bromm was threatened with prosecution by the Commonwealth relative to her testimony at Robert's trial or that her testimony was in any way coerced by the Commonwealth.   Accordingly, we adjudge this was not a structural error that affected the entire framework of the trial.   Rather, it was a “ ‘trial error’ which may be ‘quantitatively assessed in the context of other evidence presented ․’ ” Quarels, 142 S.W.3d at 81 (quoting Fulminante, 499 U.S. at 307-08, 111 S.Ct. at 1264).

 Neither party raised the issue of KRS 421.260 before the trial court.  RCr 9.22.   Hence, the error must rise to the level of palpable error to be reversible.  RCr 10.26.   Under RCr 10.26, “an error is reversible only if a manifest injustice has resulted from the error.”   Martin v. Commonwealth, 207 S.W.3d 1, 3 (Ky.2006) (quoting Graves v. Commonwealth, 17 S.W.3d 858, 864 (Ky.2000)).  “To discover manifest injustice, a reviewing court must plumb the depths of the proceeding ․ to determine whether the defect in the proceeding was shocking or jurisprudentially intolerable.”  Id. at 4. Here we cannot say that informing Bromm of the arrest warrant and executing the arrest warrant (although she was never jailed) before her testimony constituted palpable error.   Bromm voluntarily appeared and testified at Robert's trial, even though she was not properly subpoenaed as an out of state witness pursuant to KRS 421.250.   There was no indication that Bromm testified more favorably for the Commonwealth because of the arrest warrant, and there has been no allegation by Robert that Bromm did not testify truthfully.   Although Bromm's testimony corroborated April's version of the events of September 11, 2004, and supported April's alibi, she provided no testimony about the fire or about Robert's possible involvement in the fire.


 On February 8, 2006, Robert attempted to enter a guilty plea pursuant to a plea bargain with the Commonwealth whereby he agreed to plead guilty to a combination of reduced/dismissed charges in exchange for a recommendation of a twenty-five year sentence.   At the plea hearing, the court stated to Robert, “Tell me what you did.” Robert responded, “I left them in the trailer by theirself.   I recklessly left them in the house while I was drunk and recklessly started the fire.”   Robert continued, “I left the kids in the house and I went to look for April.   Then the house caught on fire and they got injured.”   The trial judge told Robert that the facts as given to him by Robert did not constitute First-Degree Assault.   The court then asked, “Did you do anything to start the fire?   Tell me the truth.”   Robert answered, “Recklessly, I did your honor.”

A bench conference then ensued wherein the judge instructed defense counsel that he was looking for the truth and was not going to allow anyone to plead guilty simply by reciting a legal term.   Defense counsel then explained that they had prepared for Robert to plead guilty to a reckless state of mind relative to the reduced Third-Degree Arson charge.   Again the judge asked Robert to tell him the truth.   After a break, Robert again attempted to plead guilty, stating “I believe I recklessly left a cigarette out when me and April were fighting and arguing.   A cigarette got knocked out of my hand and fell on the couch, and I believe that's how the fire got started.”   The court concluded that the plea was actually an Alford plea and refused to accept it because it had a policy of rarely accepting Alford pleas.   The court told Robert that he would accept the Alford plea only as an open plea, allowing the Commonwealth to put on its case in chief for jury sentencing.

A court may refuse to accept a guilty plea.  RCr 8.08.   In Hoskins v. Maricle, 150 S.W.3d 1, 21 (Ky.2004) (citing Cobb v. Commonwealth, 821 S.W.2d 817, 818 (Ky.App.1992)), this Court made clear that “in Kentucky, a defendant has no right to plead guilty to a lesser included offense for the purpose of precluding conviction of the indicted offense.”   A “charge bargain”, which dismisses or amends one or more charges in exchange for a guilty plea on the reduced charges, or a “hybrid bargain”, which is a charge bargain with an additional agreement as to sentencing, can be approved or rejected at the discretion of the trial court, so long as the court independently reviews the agreement and sets forth the prosecutor's reasons for the bargain and the court's reasons for rejecting the bargain, Id. at 22-24.

The written plea agreement is not in the record before us, but it is clear from the videotape of the plea proceedings that the plea bargain in the present case was a hybrid bargain.   Although the court did not explicitly state the prosecutor's reasons for forming the plea bargain, Robert does not complain about this deficiency.   The court clearly articulated that it was rejecting the plea because it wanted Robert to truthfully admit the facts behind the charges he was pleading guilty to, and that the plea was essentially an Alford plea.   Robert argues that it was an abuse of discretion to reject the plea for this reason.   We disagree.   There was no abuse of discretion in refusing to accept the guilty plea in this case.


 In anticipation of its upcoming child witness, Zachary Carpenter, who was seven years old at the time of trial, the Commonwealth moved to have Zachary testify in chambers via closed circuit television because of anxiety the child was having about testifying with Robert in the room.   The court held a hearing on the motion.   At the hearing, Melinda Gill, Zachary's school-based therapist, testified that Zachary was highly susceptible to influence and she was concerned that he would skew his testimony to say what the most powerful person in the room wanted to hear.   Gill stated that Zachary would most assuredly view Robert as the most powerful person in the room.   Gill also testified that Zachary was having some anxiety about testifying at Robert's trial, although she stated that when she told Zachary about the trial, Zachary smiled and asked if he was going to get to talk to Robert.   Zachary's foster mother testified that Zachary had been complaining of a stomachache since being told that he would be testifying at the trial.   The Commonwealth further expressed concern for Zachary because during April's testimony, Robert had gotten up from his seat at the defense table and moved, and the prosecution believed this might have been in an effort to intimidate April.   The prosecutor claimed that April's demeanor changed after this move by Robert.

The court decided to allow Zachary to testify in chambers, although it expressed that it was risky for the Commonwealth.   Before Zachary was to be called, the Commonwealth changed its mind and informed the court that it now felt that Zachary should testify in open court in the presence of Robert.   However, the Commonwealth still sought to minimize the chance that Robert could intimidate Zachary.   The trial judge suggested that Robert move his chair at the defense table six inches toward the wall and not be permitted to move during Zachary's testimony.   Robert's chair was moved (the jury was not in the courtroom at this time) and the judge then sat in the witness chair himself and noted that the child should only be able to see the top of Robert's head when testifying.   Defense counsel expressed her continuing objection that Robert would be unconstitutionally denied his right to confront his accuser face to face.

Zachary was then called as a witness and sat in the witness chair to the right of the judge that all previous witnesses had sat in.   The configuration of the courtroom appeared to be that in looking out into the courtroom from the vantage point of the witness box, the defense table was far to the left.   It was not possible in viewing the videotape of the trial to ascertain whether or how much Zachary and Robert could see each other during Zachary's testimony.

Zachary's testimony at trial corroborated April's and Bromm's testimony that Zachary, Nicholas and their mom were at Bromm's house before the fire.   Zachary testified that when his mom and he walked back to the trailer, they could see that it was on fire.

On appeal, Robert argues that by requiring him to move his chair to obstruct Zachary's view of him during Zachary's testimony, he was denied his right to confront his accuser face to face in violation of the Sixth Amendment of the United States Constitution.  “The Confrontation Clause of the Sixth Amendment ․ provides:  ‘In all criminal prosecutions, the accused shall enjoy the right ․ to be confronted with the witnesses against him.’ ”  Maryland v. Craig, 497 U.S. 836, 844, 110 S.Ct. 3157, 3162, 111 L.Ed.2d 666 (1990).   Section 11 of the Kentucky Constitution reads, in part, “In all criminal prosecutions, the accused has the right to ․ meet the witnesses face to face.”  Greene v. Commonwealth, 197 S.W.3d 76, 83 (Ky.2006). cert, denied, 549 U.S. 1184, 127 S.Ct. 1157, 166 L.Ed.2d 1001 (2007), made it clear that KRS 421.350, which allows witnesses under the age of twelve to testify out of court under certain circumstances if they are a victim or witness to one of the specifically enumerated crimes (mostly sexual crimes), is limited to trials for those particular offenses only.   In Greene, it was held to be error to allow a child witness to testify via closed circuit television in a Murder trial because Murder and Manslaughter were not in the class of specifically enumerated offenses in the statute, Id. However, the Court found the error to be harmless where the child's testimony was cumulative of other evidence presented and not critical to the Commonwealth's case.  Id. at 84-85.

In the present case the prosecution clearly could not avail itself of KRS 421.350 because none of the offenses with which Robert was charged were included in the class of offenses in the statute.   However, Zachary testified in open court, and although the record does not confirm whether Robert and Zachary could see each other's faces during Zachary's testimony, it was established that each was nevertheless aware of the other's presence at trial.   During the bench conference before Zachary's testimony, the attorneys noted that Zachary had seen Robert in the hallway and knew he would be present at trial.   In our view, even if there was error in obstructing Zachary's view of the defendant, any error was harmless.   As in Greene, Zachary's testimony was cumulative of other witnesses-April and Bromm-and was not critical to the Commonwealth's case.

For the above stated reasons, the judgment of the Logan Circuit Court is affirmed.

Although I concur with the majority on the other issues, I must respectfully dissent in regards to the admissibility of evidence regarding arson dog alerts to the presence of accelerants where scientific tests do not confirm the presence of accelerants in the samples taken from the sites alerted to.

At trial, it was acknowledged that the arson dog, PJ's, arson detection skills had not been recertified for at least four months prior to the fire and she had never been trained on lighter fluid.   She was having a bad day and was more interested in playing, than she was in investigating the fire.

At the trailer, she “wasn't concerned,” and PJ's handler had to “make PJ get her nose down” and reward her to get her to perform her sitting behavior.   Only then did PJ sit at six locations in the trailer and receive six more food treats.   No one, including PJ's handler, Buster Cannon, retained any record of where, or how, or with what specific encouragement PJ sat down six times in the trailer.   Samples were then collected for testing from the six spots where PJ sat down.   Every one of the KSP lab results of these samples were negative for accelerants.

The KSP forensic lab has the ability to accurately identify accelerants in vapor given off by samples from the scene at concentrations as low as 15-20 parts per million.   The ability to confirm the presence of these liquids at 15-20 parts per million is state-of-the-art science.

At the Daubert hearing, PJ's handler admitted that, even regarding concentrations above 15-20 parts per million, he had no documentation of PJ's accuracy rate, because he kept no log on PJ's field work.   He had rarely, if ever, received any follow-up information that would inform him whether PJ was right or wrong in any cases PJ had worked.   In fact, he admitted that, for all he knew, PJ could be “all wrong.”   Yet, in closing argument, the Commonwealth drove home PJ the dog's opinion at least five times, even though the KSP forensic lab results indicated that there were no ignitable liquids on a single scrap from the trailer.

Having grown up with, trained, and hunted scenting dogs most of my life, I believe there is a high probability that the specter of good ol' PJ, mans' best friend, who even the judge said would not lie,1 overshadowed all the other evidence in this case and played a highly prejudicial role in convincing the jury to convict Appellant of intentional arson.   The belief of humanity in the extraordinary abilities of scent-tracking dogs is so trustful even, that we took judicial notice of their abilities over 100 years ago.   Pedigo v. Commonwealth, 103 Ky. 41, 44 S.W. 143(1898).

This case, however, does not involve tracking dogs, which have proved their mettle over the centuries by locating prey, or the escapee, at the end of the trail.   Nor does it involve questions of “probable cause” common to the use of narcotic alert dogs, who again alert to the existence of contraband, the existence of which is soon evident.   See Morton v. Commonwealth, 232 S.W.3d 566, 569 (Ky.App.2007) (“when the drug dog detected the odor of drugs inside Morton's vehicle ․ Hord was provided with probable cause”).   It deals with dogs trained under differing standards to alert to the presence of accelerants, something we can never see, but one which is fraught with the presence of the trust developed between us over thousands of years.

“[W]e [should] recognize that the real danger posed by admitting [such] evidence lies not simply in its fallibility, but in its potential to prejudice.”  People v. Cruz, 162 Ill.2d 314, 205 Ill.Dec. 345, 643 N.E.2d 636, 662 (1994).

It is well known that the exercise of a mysterious power, not possessed by human beings, begets in the minds of many people a superstitious awe ․ that they see in such an exhibition a direct interposition of Divine Providence in aid of human justice.   The very name by which the animal is called has a direct tendency to enhance the impressiveness of the performance, and it would be dangerous in the extreme to permit the introduction of such testimony in a criminal case under conditions which did not fully justify its consideration․

Pedigo v. Commonwealth, 44 S.W. at 145-46.

In fact,

[The canine accelerant alert] program is built around the lab analysis of samples taken by investigators after the dogs key in on the location of residual accelerants on a fire scene.   Despite the dogs' inherent ability to locate accelerants, lab analysis of samples is the key to the program's success.  “There are kind of two camps on the dogs' use,” says Sgt. Jeff Howard, arson section manager with OSP [Oregon State Police], “The Connecticut State Police and the ATF [The Federal Bureau of Alcohol and Tobacco Firearms] have stressed the party line from the beginning-the dog is just a tool.   Period.  He who uses the dog as an expert without lab confirmation is a fool.”

Kelly Andersson, Arson Dogs, Wildland Firefighter Magazine (1997), http:// ​www.​workingdogs.​com/​doc​0130.​htm. These are not my words, but the words of the Arson Section Manager of the Oregon State Police.   Even the guideline “published by the National Fire Protection Association suggests that evidence of a canine alert that is not confirmed by laboratory testing should not be considered valid.”  Commonwealth v. Crouse, 447 Mass. 558, 855 N.E.2d 391, 402 (2006).

“While the reliability of tracking dogs has been widely recognized and generally accepted by the courts, with thirty-two states approving of admissions of trailing by [dogs], courts have been much less receptive to accelerant-detection [canine alerts] not confirmed by laboratory analysis.”   State v. Sharp, 395 N.J.Super. 175, 928 A.2d 165, 169 (Law.Div.2006).   In fact, “[t]here is substantial scientific agreement that unconfirmed canine alerts are not reliable in the absence of laboratory confirmation, largely due to the fact the canines cannot meaningfully discriminate between background pyrolysis products [those created by the combustion] and accelerants.”  Id. at 170, 928 A.2d 165;  see also John D. DeHaan, Kirk's Fire Investigation 543 (6th Ed.2006) (“without a specific verifiable identification of just what is present [by a lab], the investigator cannot decide the significance of such positive alerts”).

Most courts that have considered the issue of uncorroborated canine accelerant alerts have found them to be novel scientific evidence, not generally accepted in the scientific community of arson investigators.   See Farm Bureau Mut. Ins. Co. v. Foote, 341 Ark. 105, 14 S.W.3d 512 (2000) (holding that testimony that accelerant detection canine is more reliable than laboratory equipment is without scientific validity);  Carr v. State, 267 Ga. 701, 482 S.E.2d 314 (1997) (holding that reliability of uncorroborated canine alerts is questionable);  Sharp, 928 A.2d at 171;  State v. Webber, 716 A.2d 738, 741-742 (R.I.1998) (reversing conviction under balancing test of R.I. Rule of Evidence 403 where lab tests of alert areas were negative for accelerants);  People v. Acri, 277 Ill.App.3d 1030, 214 Ill.Dec. 761, 662 N.E.2d 115 (1996) (stating that uncorroborated alerts are not generally accepted);  State v. Schultz, 58 P.3d 879, 885 (Utah App.2002) (stating that uncorroborated alert is “novel scientific evidence”):  see also Fitts v. State, 982 S.W.2d 175, 183 (Tex.App.1998) (“[canine's] reliability was ․ evidenced by the laboratory test showing the presence of gasoline at the scene.”):  but see State v. Buller, 517 N.W.2d 711 (la.1994) (requiring dog accuracy foundation be laid in making determinations of admissibility subject to Iowa Rule of Evidence 403 balancing test);  Commonwealth v. Crouse, 447 Mass. 558, 855 N.E.2d 391 (2006) (permitting testimony of accelerant detection despite an absence of lab results where gas station surveillance video tape showed the defendant pumping gasoline into the right rear cargo area of the truck in the exact area of the dog's alert along with statements of the defendant that he had purchased the gasoline);  Commonwealth v. Gwynn, 555 Pa. 86, 723 A.2d 143, 152 (1999) (admitting evidence of unconfirmed canine accelerant alert in trial where Appellant's confession indicated he had poured the gasoline and ignited the fire accidentally);  Fones v. State, 765 So.2d 849, 850 (Fla.App.2000) (permitting testimony of accelerant dog alert despite absence of lab results).

In Debruler v. Commonwealth, 231 S.W.3d 752, 757 (Ky.2007), we held that testimony relating to a scent-tracking dog's action “concerned the results of an investigative technique, not a scientific procedure.”   Yet here, by approving testimony of canine accelerant alerts without lab verification, we are supplanting scientific procedure with an investigative technique.   I can not be a party to this until testimony regarding such unconfirmed alerts is generally accepted in the scientific community of arson investigators.   See Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999);  Goodyear Tire & Rubber Co. v. Thompson, 11 S.W.3d 575 (Ky.2000).   For these reasons, I record my dissent.


1.   “[D]ogs are such honest animals.   They don't lie unless they want food.”

LAMBERT, C.J.;  ABRAMSON, CUNNINGHAM, MINTON, and SCHRODER, JJ., concur. SCOTT J., files a separate opinion concurring in part and dissenting in part in which NOBLE, J. joins.

Copied to clipboard