Commonwealth v. Treiber, S., Aplt

[J-49-2013] [M.O. – Eakin, J.]
IN THE SUPREME COURT OF PENNSYLVANIA
EASTERN DISTRICT
COMMONWEALTH OF PENNSYLVANIA, : No. 656 CAP
:
Appellee : Appeal from the Order entered on
: 3/27/12 in the Court of Common Pleas,
: Criminal Division of Erie County at No.
: CP-25-CR-0000842-2001
v. :
:
:
STEPHEN TREIBER, :
:
Appellant : SUBMITTED: May 30, 2013
DISSENTING OPINION
MR. CHIEF JUSTICE SAYLOR DECIDED: August 17, 2015
According to representations made by the Commonwealth, the prosecution’s use
of canine DNA evidence at Appellant’s trial was “crucial to the Commonwealth to prove
identity, intent and lack of accident.” Commonwealth’s Reply to Defendant’s Post-
Sentence/New Trial Motion, dated December 3, 2002, at 2. Indeed, in his sworn
testimony during the post-conviction proceedings, the trial prosecutor reaffirmed that
“this type of DNA match [is] the type of thing that a jury wraps itself around,” and that
the evidence “destroyed any kind of defense that [Appellant] tried to muster.” N.T.,
June 7, 2011, at 25. It is Appellant’s core assertion that his attorney failed to advance
an obvious challenge to that evidence by invoking the requirements for admission of
novel scientific evidence per Frye v. United States, 293 F. 1013 (D.C. Cir. 1923). By
reason of this dereliction, Appellant argues, he was deprived of the ability to capitalize
on the most positive piece of exculpatory evidence available to him -- i.e., the “threat
note” -- which he has contended throughout explains much of the prosecution’s
circumstantial evidence about his behavior preceding the fire.
Presumably in light of the prosecution’s own position that the canine DNA
evidence was pivotal to its cause, the Commonwealth has not taken up the
contradictory position that the evidence was non-prejudicial, nor did the PCRA court so
find. In these circumstances, I strongly differ with the majority’s sua sponte resort to a
prejudice-based disposition. See Majority Opinion, slip op. at 19-22. Moreover, this
approach, in my view, has the effect of negating the broader import of the case in terms
of affording essential cautionary guidance concerning the admission of novel scientific
evidence in the courtroom.
One has only to survey the current media to learn that there are grave concerns
being raised concerning the government’s use of novel scientific techniques and
analysis in obtaining convictions, and about liberality on the part of the courts in
permitting such use. See, e.g., FBI National Press Release, FBI Testimony on
Microscopic Hair Analysis Contained Errors in at Least 90 Percent of Cases in Ongoing
Review (Apr. 20, 2015) (explaining that the FBI is engaged in an ongoing review of
more than 2500 cases in which criminal convictions were attained, in part, based on
scientifically flawed evidence presented through FBI agent-analysts or laboratory
reports) (available at http://www.fbi.gov/news/pressrel/press-releases/fbi-testimony-on-
microscopic-hair-analysis-contained-errors-in-at-least-90-percent-of-cases-in-ongoing-
review) (last visited May 29, 2015). Recently, the National Research Council published
a report discussing “documented ills of the forensic science disciplines” and a
continuing tendency of courts to “rely on forensic evidence without fully addressing the
limitations of different forensic science disciplines.” NATIONAL RESEARCH COUNCIL,
[J-49-2013] [M.O. – Eakin, J.] - 2
STRENGTHENING FORENSIC SCIENCE IN THE UNITED STATES, A PATH FORWARD 85 (Nat’l
Academies Press 2009). In particular, the report emphasized the need to “limit the risk
of having the reliability of certain forensic science methodologies condoned by the
courts before the techniques have been properly studied and their accuracy verified.”
Id. at 109.
In light of such considerations, this Court has maintained that “[s]trict application
of the Frye standard when [novel] scientific proof is offered is essential if the defendant
is to receive a just and fair trial[.]” Commonwealth v. Topa, 471 Pa. 223, 232, 369 A.2d
1277, 1282 (1977). In the present case, I agree with Appellant’s core position that strict
application of the Frye standard did not occur on account of material and prejudicial
derelictions on the part of Appellant’s trial counsel. My reasoning follows. Although my
statement of the background overlaps with the majority’s presentation, I find it
necessary to lay the groundwork for my conclusions in considerable detail.
Appellant has been adjudged to have set fire to his own home on March 9, 2001,
while his girlfriend Denise Riddle and his two-year-old daughter Jessica slept. As the
structure burned, Appellant and Ms. Riddle escaped, but Jessica remained in her crib
until firefighters were able to remove and unsuccessfully attempt to revive her.
Appellant was charged with criminal homicide, multiple forms of arson, and reckless
endangerment, and the Commonwealth gave notice of an intention to pursue imposition
of the death penalty.
As related by the trial court, “[t]he Commonwealth’s theory of the case was that
[Appellant] had taken a number of steps to make it appear that someone else started
the fire.” Commonwealth v. Trieber, Nos. 842A&B of 2001, slip op. at 5 (C.P. Erie Feb.
27, 2003). The most remarkable of these measures was to contrive a threatening letter,
which, the prosecution asserted, Appellant surreptitiously attached to his own mailbox
[J-49-2013] [M.O. – Eakin, J.] - 3
and arranged for Ms. Riddle to find about six weeks before the fire. See N.T., Oct. 2,
2002, at 85-86; N.T., Oct. 7, 2002, at 95-98.1 At that time, Appellant turned the note
over to law enforcement, the Commonwealth contended, in an effort to cast suspicion
away from himself and onto unknown others. See N.T., Oct. 3, 2002, at 39-42; N.T.,
Oct. 7, 2002, at 96. In the weeks preceding the fire, he regularly called police to ask
about the progress of their investigation into the source of the letter. See N.T., Oct. 3,
2002, at 103.
Although initial examinations of the threat letter in a police laboratory produced
no forensic evidence, eventually, a scientist found two hairs protruding from dried glue
affixed to the paper.2 At least one of these, he concluded, was canine. See id. at 14-
17. Police sought to determine whether either hair had originated from Appellant’s
dogs, since this would connect him to the preparation of the threat letter. Accordingly,
they sent the evidence samples to Joy Halverson, DVM, a California veterinarian and
epidemiologist by formal education, who also had practical experience in molecular
biology and held herself out as a forensic canine DNA analyst. See id. at 109-12.
Along with this submission, police provided reference blood and saliva samples taken
from Appellant’s dogs -- most significantly one taken from the body of his dog Janie,
which had perished in the fire. The district attorney also advised Appellant’s attorney
that testing was to occur.
Appellant’s lawyer then initiated some inquiries, telephoning several individuals
knowledgeable in animal DNA analysis, including Marcia Eggleston, PhD, who oversaw
the genetic testing of animals at the University of California, Davis. The main subject of
1
The letter was addressed to “Steve,” and said, “Get rid of the dogs or I will kill them
and burn you out again.” N.T., Oct. 2, 2002, at 88-89.
2
The note’s text was comprised of letters cut from printed materials, which had been
glued onto the paper.
[J-49-2013] [M.O. – Eakin, J.] - 4
these conversations was to determine whether a defense expert should be present
during the testing, particularly since it appeared that, given the miniscule sizes, the
evidence samples would likely be consumed. See N.T., Aug. 10, 2009, at 32-34. Ms.
Eggleston told counsel that Ms. Halverson was qualified to undertake the testing and
that the procedures were straightforward. See id at 39-40. Based on such
conversations -- and prior to the actual testing or preparation of any report -- counsel
largely abandoned further inquiry into scientific validity and ceased efforts to obtain a
defense expert,3 in favor of a strategy of advancing the proposition that the hairs found
on the threat letter were merely contaminants, which came in contact with the letter well
after the time it was prepared. See id. at 40-41.4
In pre-trial discovery, the Commonwealth produced to the defense reports by Ms.
Halverson and statistician Christopher Basten, PhD, concerning the evidence samples
taken from the threat letter. Ms. Halverson’s reports disclosed that the sample which
she labeled “Golden2” was “suboptimal” and produced “low template quantity or quality”
DNA. Reports of Joy Halverson, DVM, dated Oct. 2, 2001, and Aug. 26, 2002,
Petitioner’s Evidentiary Hearing Exhibits, Vol. IX, tabs 2 & 3. Indeed, profiling was
incomplete at ninety percent of the loci analyzed. See id.; accord N.T., Oct. 29, 2009, at
3
It appears that, at a later point in time, counsel considered further consultation with an
expert, since, before trial, the trial court granted a continuance to provide the defense
with the opportunity to do so. See Order dated Oct. 24, 2001, in Commonwealth v.
Treiber, Nos. 842 A&B of 2001 (C.P. Erie) (“[T]his Court finds that the defense requires
additional time in which to review scientific evidence recently developed by the
Commonwealth and requires an opportunity to obtain an expert witness to review the
findings of the Commonwealth’s expert(s).”). However, no further consultation
occurred. See N.T., Aug. 10, 2009, at 38-39, 158-60.
4
Obviously, the prosecution’s rejoinder emphasized that the hairs were embedded in
dried glue, thus evidencing that they were reposited on the letter at a time when the
glue was still wet, i.e., during the letter’s preparation. See, e.g., N.T., Oct. 7, 2002, at
96-97.
[J-49-2013] [M.O. – Eakin, J.] - 5
136-37. Nevertheless, through her laboratory analysis identifying eight alleles in the
Golden2 sample matching the reference sample taken from Appellant’s dog Janie and
no excluding alleles, Ms. Halverson declared a match.5 Via a particularized application
of the product rule,6 Ms. Halverson opined that
[a]ssuming that [the utilized] data group is representative of
the general dog population and is the correct data set for
comparison in this case, the product rule shows that the
likelihood that the evidence sample Golden2 and reference
blood sample are from different dogs and match by random
chance exceeds 1 in 1.6 million.
Report of Joy Halverson, DVM, dated Oct. 2, 2001, Petitioner’s Evidentiary Hearing
Exhibits, Vol. IX, tab 2.
In a supplemental report, Ms. Halverson indicated that, having recently
completed coursework in forensic DNA testing, she attained a “greater understanding of
the use of the likelihood ratio for estimating the significance” of a match between
evidence and reference samples. Supplemental Report of Joy Halverson, DVM, dated
Aug. 26, 2002, Petitioner’s Evidentiary Hearing Exhibits, Vol. IX, tab 3. According to the
report, she had come to appreciate the necessity of abandoning a straightforward
application of the product rule, in favor of applying a likelihood ratio which could correct
for “population substructure found in dog breeds and for the data missing from profiles
of suboptimal DNA samples, such as the hair in this case.” Id. Applying such ratio, Ms.
Halverson then made more than a thousand-fold downward adjustment of her previous
5
As described in Ms. Halverson’s analysis, alleles reflect consistency or variances
relative to a given genetic marker. See N.T., Oct. 3, 2002, at 125.
6
The product rule is a principle of statistical probability which is applied in DNA analysis
to determine the likelihood of a random match between evidence and reference
samples. See generally Commonwealth v. Blasioli, 552 Pa. 149, 161-62, 713 A.2d
1117, 1123-24 (1998).
[J-49-2013] [M.O. – Eakin, J.] - 6
probability estimate, while cross-referencing the supportive report of the prosecution’s
statistical expert, Mr. Basten. See id.
Upon review of the Halverson and Basten reports, Appellant’s trial attorney did
not consult with a DNA expert, despite obtaining a continuance of the trial for the
express purpose of doing so. See supra note 3. Nor did the lawyer lodge a pre-trial
challenge to the admissibility of their testimony. Rather trial counsel simply maintained
the strategy of asserting post-preparation contamination of the threat letter. See N.T.,
Aug. 10, 2009, at 37-39.
At trial, Ms. Halverson testified, consistent with her report, that her testing
revealed a match between the DNA template derived from Golden2 and that of the dog
Janie’s blood. According to Ms. Halverson’s testimony, it was “a thousand times more
likely that they match because they came from the same dog than because they came
from two dogs by coincidence.” N.T., Oct. 3, 2002, at 129. This thousand-times figure
was reinforced through testimony from Mr. Basten. See id.
On cross-examination, Appellant’s attorney asked no questions of Ms. Halverson
concerning her qualifications and offered no objection to her acceptance as an expert in
the field of DNA analysis and comparison. See N.T., Oct. 3, 2002, at 116. Rather, the
lawyer told Ms. Halverson, “I’m certainly not going to quibble with you about your
findings.” Id. at 132. Consistent with this representation, counsel made no inquiries
about her protocols, the actual testing, or the ensuing analysis; rather, counsel briefly
had Ms. Halverson confirm that she was unfamiliar with the source of the evidence
samples of hair she had tested,7 then asked a few questions about the extraneous hair
7
Parenthetically, Ms. Halverson’s attestation to such unfamiliarity conflicted sharply with
testimony she later gave in the post-conviction proceedings. See, N.T., Oct. 29, 2009,
at 119-20.
[J-49-2013] [M.O. – Eakin, J.] - 7
which had yielded no DNA profile. See id. at 132-34. Appellant’s attorney also offered
no cross-examination whatsoever of Mr. Basten. See id. at 143.
In the defense case, Appellant testified, denying having created the threat letter
or having committed arson or murder. See N.T., Oct. 5, 2002, at 36-37, 61. Consistent
with trial counsel’s strategy to focus on contamination, the defense did not present a
DNA expert.
In arguments to the jury, trial counsel again affirmatively conceded the DNA
match declared by Ms. Halverson and advanced his contamination argument. See,
e.g., N.T., Oct. 7, 2002, at 75 (reflecting trial counsel’s remark to the jury in closing that
“[w]e know because the DNA lady came and told us that the dog hair, the single dog
hair matched Janie the dog that was in the basement of the fire. But I’m going to
suggest to you from that evidence, so what?”). The prosecutor, for his part, stressed
that the forensic evidence strongly connected Appellant to the threat note and evinced
his preparation for arson and murder. He explained:
[C]riminals make mistakes, ladies and gentlemen. The
things of life, the commonplace things that happen everyday
that criminals, no matter how brilliant or how evil or how
careful they plan, never imagine. . . .
In this case a hair, a dog hair . . .. And it’s actually important
when the hair is discovered, because the time the hair is
discovered the dog from whom it came is dead and burnt.
Janie’s hair was put there under the [letter] “o” when Janie
was still alive. Okay. And we know it was put there before
or on January 30th because that hair is found embedded in
the glue. . . . This was not some hair floating in the air later
on. This hair was put on that letter, under that letter when
the glue was attached, when the letter was attached.
[Appellant] wrote himself a threatening letter, cutting out
letters, like TV. And did that over a month before he killed
his daughter. He never figured on that. He couldn’t see the
[J-49-2013] [M.O. – Eakin, J.] - 8
hair. He may have been clever enough not to lick the
envelope. They found no DNA from anybody. And certainly
he touched the letter after it gets “discovered.” So that
explains any stray fingerprints that might be there. But he
didn’t figure on the dog hair.
The dog hair tells you that on January 30th he was already
planning this fire.
N.T., Oct. 7, 2002, at 96-98.
The jury returned verdicts of guilt relative to first-degree murder, arson, and
reckless endangerment. After a penalty hearing, Appellant received a sentence of
death. Appellant pursued a direct appeal, represented by trial counsel, and relief was
denied. See Commonwealth v. Treiber, 582 Pa. 646, 874 A.2d 26 (2005).
The present collateral proceedings followed, per the Post Conviction Relief Act.
See 42 Pa.C.S. §§9541-9546 (the “PCRA”). Appellant raised a series of challenges,
including an assertion that his attorney at trial was ineffective for failing to challenge the
Commonwealth’s low-template, canine DNA evidence. Amended PCRA Petition of July
9, 2007, (“Amended Petition”) ¶¶52-219. Among other lines of argumentation, Appellant
contended that counsel had failed to adequately investigate the evidence and to file and
litigate a challenge to its admissibility per Frye v. United States, 293 F. 1013 (D.C. Cir.
1923).
Appellant explained that the Frye test, adopted by this Court as the standard
governing the admissibility of novel scientific evidence in Pennsylvania, requires general
acceptance in the relevant scientific community of the underlying scientific theory and
methodology. See Grady v. Frito-Lay, Inc., 576 Pa. 546, 555, 839 A.2d 1038, 1043-44
(2003); Commonwealth v. Blasioli, 552 Pa. 149, 153, 713 A.2d 1117, 1119 (1998);
Commonwealth v. Topa, 471 Pa. 223, 231, 369 A.2d 1277, 1281 (1977). See generally
Blum ex rel. Blum v. Merrell Dow Pharm., Inc., 564 Pa. 3, 6, 764 A.2d 1, 3 (2000) (“Frye
requires the scientific community to reach some consensus as to reliability then relies
[J-49-2013] [M.O. – Eakin, J.] - 9
on such consensus to determine the admissibility of the challenged scientific
evidence.”). Appellant stressed that, at the time of his trial, none of the hallmarks of
general acceptance -- such as scientific study, validation, and peer review -- were
present relative to canine DNA evidence. See, e.g., Amended Petition ¶55 (“[T]here
was not one peer-reviewed article in any scientific journal regarding the forensic
application of canine DNA identification procedures at issue in [Appellant’s] case or a
single court decision in the entire country upholding the admissibility of such evidence
under Frye.”). Indeed, according to various declarations appended to Appellant’s
petition, far from being generally accepted, the theory and methodology underlying the
Commonwealth canine DNA evidence explicitly was rejected by multiple experts in
forensic genetics. See id. ¶64 (citing declarations of Laurence D. Mueller, PhD, Randell
Libby, PhD, and Ms. Eggleston).
By way of background pertaining to human DNA theory and methodology,
Appellant explained:
The most common form of human DNA testing used today is
called the short tandem repeats, or STR, test. STRs are
regions of DNA that contain a series of short repeated units.
See, e.g., . . . [Jane Campbell] Moriarty, [2] Psychological
and Scientific Evidence in Criminal Trials, [] §§ 11:6 & 11:22
[1997)] (“Moriarty”). This test can be performed using trace
amounts of DNA, which are then amplified, or replicated,
using a method called polymerase chain reaction, or PCR.
Id., at §§ 11:6 & 11:20 . . .. Importantly, . . . “[t]he forensic
science community in the United States has standardized
[human] DNA typing using a set of 13 core STR loci that
have relatively high degrees of variation in the population as
a whole. This set of 13 core STR loci is used for entry into
the national DNA profiling database known as CODIS, which
is managed by the FBI.” Michael R. Bromwich, Final Report
of the Independent Investigator for the Houston Police
Department Crime Laboratory and Property Room, June 13,
2007 at 119 (available at
[J-49-2013] [M.O. – Eakin, J.] - 10
http://www.hpdlabinvestigation.org). Because of their
extensive variability, these highly polymorphic loci are used
by forensic scientists as “markers” to detect genetic
variations, that is, distinguish among individuals. Moriarty at
§ 11:6. . . . Commercial kits amplify DNA from those
locations using PCR. As the kits amplify the DNA, they label
the specific locations on the DNA molecule to be tested with
colored dyes. An automated test process then “reads” the
DNA and generates graphs which show the alleles at each
location tested. An analyst then interprets the graphs to
determine whether or not the samples have matching alleles.
Id. at § 11:25.
Amended Petition ¶58 n.5. See generally KENNETH S. BROUN, 1 MCCORMICK ON
EVIDENCE §205 (2013) (discussing the use of PCR-based STR profiling and
electrophoresis to detect the number of repeats, at given locations, for different alleles
on a graph known as an electropherogram).8
Appellant acknowledged that Ms. Halverson extended methods applicable to
human DNA analysis into the canine DNA arena. He explained, however, that the
underlying robust, studied, and verified infrastructure for profiling human DNA simply
was not present relative to canine DNA. In this regard, Appellant asserted, there was
no reliable, verified standardized marker kit supporting canine forensic DNA
identification and no reliable, verified canine database to utilize for the sake of
comparison. See Amended Petition ¶58 n.5. According to Appellant, Ms. Halverson
had simply adapted her own company’s kit containing proprietary markers previously
used in parentage testing and assembled a database of convenience consisting
primarily of DNA samples from local (California) dogs.
8
While this Court’s Crews and Blasioli decision involved procedures known as
Restriction Fragment Length Polymorphism (“RFLP”) testing, see, e.g., Blasioli, 552 Pa.
at 157, 713 A.2d at 1121, PCR-based STR procedures, such as were used by Ms.
Halverson, are more discriminating and efficient and, thus, have come to “dominate
forensic testing.” BROUN, 1 MCCORMICK ON EVIDENCE §205.
[J-49-2013] [M.O. – Eakin, J.] - 11
From the outset, Appellant stressed that his position, in this regard, was
vindicated in State v. Leuluaialii, 77 P.3d 1192 (Wash. App. Ct. 2003), in which a state
appellate court determined that canine DNA evidence presented through Ms.
Halverson’s testimony in a 1998 trial failed to meet the requirement of general
acceptance. See id. at 1197. Appellant highlighted the Leuluaialii court’s determination
that canine DNA evidence “clearly involved novel scientific theory: the forensic
identification with high statistical probabilities of a specific dog,” as well as its conclusion
that “[a] Frye hearing was absolutely necessary in this case.” Id. As to general
acceptance, Appellant referenced Leuluaialii’s observations that “[c]urrent canine DNA
testing and mapping focuses on the goals of paternity testing, pure breed testing, and
cancer and disease research studies,” id. at 1200, and “the study of canine DNA has
not progressed to the point of the study of human DNA sufficient to permit an expert to
testify to a match between a sample and a specific dog.” Id. at 1194. Additionally,
Appellant quoted the Leuluaialii court’s admonition to other judicial tribunals, as follows:
“we would suggest that other courts tread lightly in these waters and closely examine
canine DNA results before accepting them at trial.” Id. at 1201.
Another facet of Appellant’s challenge to Ms. Halverson’s methodology centered
upon her approach to the interpretation of the partial DNA profile attained from the
Golden2 evidence sample. In this regard, Appellant challenged, inter alia, Ms.
Halverson’s use of subjective criteria to support her finding of a match and her methods
for accounting for the the incompleteness of the Golden2 profile generated by her
testing, including her reliance on Mr. Basten’s likelihood ratios to account for alleles
which were not revealed (or allelic dropout). See Amended Petition ¶¶108, 127-28.9
9
Appellant also challenged many other aspects of Ms. Halverson’s methodology. It is
sufficient, for present purposes, to focus on general acceptance relative to Ms.
(continued…)
[J-49-2013] [M.O. – Eakin, J.] - 12
In terms of the stewardship of trial counsel, Appellant asserted that, had counsel
undertaken any sort of a reasonable investigation, he would readily have appreciated
the patent lack of general acceptance of the methodology underlying the
Commonwealth’s canine DNA evidence. Had counsel simply filed a Frye motion putting
this information before the court, Appellant maintained, there was a reasonable
probability that the canine DNA evidence would have been suppressed.
In the evidentiary phase of the post-conviction proceedings, Appellant offered
expert testimony from Messrs. Mueller and Libby and Ms. Eggleston, whose
declarations he had presented with his petition. They all indicated that the genetic and
statistical methods underlying the trial testimony of Ms. Halverson and Mr. Basten were
not generally accepted in the relevant scientific communities. See, e.g., N.T., Aug. 10,
2009, at 161 (Eggleston); Aug. 11, 2009, at 117 (Libby); Aug. 12, 2009, at 64 (Mueller).
In response, the Commonwealth presented testimony from Ms. Halverson and
Mr. Basten to the effect that their methods were generally accepted. See, e.g., N.T.,
Oct. 29, 2009, at 10-11 (Halverson); id. at 151 (Basten). On cross-examination,
however, Ms. Halverson affirmed that, at the time of Appellant’s trial, she was the only
person in the United States using her company’s proprietary set of canine genetic
markers to support forensic DNA analysis, and that such kit was no longer commercially
available in this country. See id. at 67-68, 77. Although Ms. Halverson had stated on
direct examination that her markers were validated, she acknowledged on cross-
examination that the validation studies to which she referred were internal to her private
laboratory. See id. at 139-40. In terms of her subjective judgments about the partial
DNA template for Golden2, Ms. Haverson repeatedly conceded that these could not be
(…continued)
Halverson’s markers, her database, and her and Mr. Basten’s approach to low-template
DNA.
[J-49-2013] [M.O. – Eakin, J.] - 13
tested or verified and “we just have to take [her] word for it.” Id. at 106-09; accord id. at
128 (reflecting Halverson’s post-conviction testimony that, “[i]n a suboptimal sample like
the hair in this case one peak might be a homozygous allele or one peak might be an
indication of allelic dropout. We can’t know.”). Further, she agreed that, if any of
several of her subjective judgments in such respects were mistaken, an exclusion would
be present, see id. at 109-10, such that Appellant’s dog Janie could not have been the
source of the Golden2 sample.10
Upon direct examination by the Commonwealth of Mr. Basten, he defended the
general acceptance of his statistical methods “currently” based on the fact that his
methodology was published in a paper and had never been rebutted. N.T., Oct. 29,
2009, at 161. On cross-examination, however, he admitted that such paper was
published three years after Appellant’s trial and, as of the time of trial, no published
papers existed relative to his methodology for accounting for the partial, canine DNA
profile revealed through Ms. Halverson’s testing. See id. at 181. Moreover, Mr. Basten
acknowledged that the computer software utilized to account for missing information in
the Golden2 DNA profile at ninety percent of the loci analyzed was proprietary and, as
such, its schematic was published nowhere. See id. at 187.
On consideration of the evidence, the PCRA court denied relief. The court
reasoned, because Ms. Eggleston had “vouched for Halverson’s testing procedure and
Halverson’s qualifications to conduct the test,” this placed counsel “in the unenviable
position of having to virtually concede the strength of the Commonwealth’s canine DNA
10
Appellant’s post-conviction counsel sought to pursue additional lines of cross-
examination with Ms. Halverson; however, the PCRA court terminated the questioning,
dubbing the subject matter of Appellant’s proffers “cumulative, adequately covered, and
goes to weight.” See, e.g., id. at 145.
[J-49-2013] [M.O. – Eakin, J.] - 14
evidence.” See Treiber, Nos. 842 A&B of 2001, slip op. at 31-33.11 The court also
highlighted that Ms. Eggleston did not refer counsel to any other expert who might
assist him. See id. at 32.12
In any event, relying on the overlap between human and canine DNA theory and
methodology, the post-conviction court pronounced that canine DNA evidence simply is
not novel for purposes of Frye. See id. at 28-29. In terms of novelty and otherwise, the
court did not discuss the aspects of Appellant’s challenges focusing on low-quality
sample testing and analysis of a materially incomplete DNA profile.
As to specifics, the PCRA court offered a series of observations focusing mostly
upon the genetic markers employed by Ms. Halverson. First, the court opined that
Appellant “failed to demonstrate that Halverson’s markers were unreliable.” Id. at 30.13
Then, the court noted that Ms. Eggleston never told trial counsel that the markers were
invalid, see id. at 32 (citing N.T., Aug. 10, 2009, at 219); in 1996, Ms. Halverson had
discussed her markers at an international conference, see id. at 31 (citing N.T., Aug. 10,
2009, at 215); Halverson had published the markers in the international community, see
11
As further discussed below, the PCRA court’s rationale in this regard rests upon a
non-sequitur, since the Commonwealth’s incriminating low-template, canine DNA
analysis had not yet come into being as of the time when counsel spoke with Ms.
Eggleston.
12
The court, however, did not reconcile its reliance on Ms. Eggleston’s ostensible
omission in this regard with its own recognition, otherwise, that Ms. Eggleston herself
had offered to provide further expert assistance herself by “review[ing] the results of the
forthcoming tests.” See Treiber, Nos. 842 A&B of 2001, slip op. at at 31.
13
It was not Appellant’s burden to do so, however, per Frye, which, as explained, turns
upon general acceptance in the relevant scientific community. The court’s allusions to
proof of actual unreliability, as opposed to general acceptance, harkens to the federal
approach to the admission of scientific evidence embodied in Daubert v. Merrell Dow
Pharm., Inc., 509 U.S. 579, 592, 113 S. Ct. 2786, 2796 (1993).
[J-49-2013] [M.O. – Eakin, J.] - 15
id. (citing N.T., Aug. 10, 2009, at 216);14 Ms. Eggleson previously had analyzed Ms.
Halverson’s markers and had referred work to her, see id. at 31-32 (citing N.T., Aug. 10,
2009, at 147, 216-17); Ms. Eggleston had testified on post-conviction that the fact that
different laboratories use different markers did not render canine DNA evidence
unreliable, see id. at 32 (citing N.T., Aug. 10, 2009, at 222); and she herself had used
three or four of the markers for a three to four year period, see id. (citing N.T., Aug. 10,
2009, at 226).
As to the database used by Ms. Halverson, the PCRA court briefly credited her
testimony, and that of Mr. Basten, that it was of sufficient size and reliability to support
DNA comparison analysis. See id. at 34 n.21.
Furthermore, the PCRA court discerned a “fundamental problem” in the
testimony of Appellant’s post-conviction experts in that portions were “predicated to a
great extent on information that did not exist at the time of the trial.” Id. at 32 n.20. The
court couched this line of evidence as “a restrospective analysis” and discounted it as
having “very little value.” Id.
The court made no mention of the contrary position taken by the Leuluaialii court,
including the Washington court’s salient analysis and conclusion that canine DNA
14
The PCRA court did not discuss, however, whether such publishing encompassed
information which could be used by other scientists to accomplish critical review. See
generally Barbara van Asch & Filipe Pereira, State-of-the-Art and Furture Prospects of
Canine STR-Based Genotyping, OPEN FORENSIC SCI. J. 45, 49 (2010) (asserting, with
reference to the genetic markers used by Halverson that the “kit did not include [a
conventional reference tool to increase reliability] nor was a nomenclature of the
included markers ever published: therefore it is not surprising that the commercial
success of the product was greatly compromised from the very beginning” (emphasis
added)). Notably, in the post-conviction hearings, the court had foreclosed Appellant
from pursuing such material line of inquiry as “cumulative, adequately covered, and
goes to weight.” N.T., Oct. 29, 2009, at 145-46. Accordingly, fairness was lacking in
the court’s implicit assumption that the materials published in the relevant time period
were sufficient to support critical scientific review.
[J-49-2013] [M.O. – Eakin, J.] - 16
evidence is novel and does not enjoy general acceptance in any scientific community.
See Leuluaialii, 77 P.3d at 1197.
This appeal followed. Although counsel is presumed to have rendered effective
assistance in the first instance, see, e.g., Commonwealth v. Beasley, 600 Pa. 458, 470
n.5, 967 A.2d 376, 383 n.5 (2009), the PCRA court should have recognized (and this
Court should now acknowledge) that Appellant has negated such presumption in the
present case.15
Initially, as I have otherwise observed, see supra note 11, the PCRA court’s
reliance on the pre-testing advice trial counsel received from Ms. Eggleston as forcing
counsel to “concede,” as he did, “the strength of the Commonwealth’s canine DNA
evidence” is unsustainable. Treiber, Nos. 842 A&B of 2001, slip op. at 33. Any
competent trial lawyer should know that, regardless of whether particular scientific
testing procedures may or may not be routine or the qualifications of the tester, the
ensuing analysis and conclusions advanced by an expert witness may present fertile
grounds for challenge. Indeed, the trial court granted a continuance of trial for the
express purpose of permitting Appellant’s attorney to secure expert review of the
Halverson/Basten reports. See supra note 3. The relevant question, then -- which the
PCRA court simply did not address -- is whether there was any reasonable basis
supporting trial counsel’s failure to follow through with such review.16
15
The review standards governing resolution of Appellant’s claims of deficient
stewardship -- evaluated in Pennsylvania according to the three-prong inquiry into
arguable merit, reasonable strategy, and prejudice -- are well developed in the
decisional law. See, e.g., Commonwealth v. Jones, 572 Pa. 343, 364-65 & n.14, 815
A.2d 598, 611 & n.14 (2002). The cases also discuss the overlap between federal
constitutional law and Pennsylvania practice in such regard. See id.
16
While not confronting the matter squarely in its opinion, the PCRA court accepted as
a fact, during the post-conviction hearings, that trial counsel did not review Ms.
(continued…)
[J-49-2013] [M.O. – Eakin, J.] - 17
Trial counsel’s sole explanation was that he had elected to pursue a strategy of
advancing the proposition that Golden2 sample came in contact with the threat letter
well after the time it was prepared. See N.T., Aug. 10, 2009, at 40-41. Such strategy,
however, suffered from an obvious drawback, given the unlikelihood that contaminants
contacting the threat letter after its preparation would be embedded in the dried glue.
Trial counsel, nonetheless ignored this apparent weakness, see N.T., Oct. 7, 2002, at
75-77, which the Commonwealth highlighted to the jury unanswered, see id. at 97-98.
Given the facially evident drawbacks of trial counsel’s contamination strategy, he was
obviously remiss in selecting such approach before exploring other available options.
See Strickland v. Washington, 466 U.S. 668, 690-91, 104 S. Ct. 2052, 2066 (1984)
(explaining that defense lawyers must conduct a thorough pre-trial investigation or make
reasonable decisions rendering particular investigations unnecessary, and strategic
choices made following a less-than-complete investigation are reasonable only to the
extent that reasonable professional judgment supports the limitation of the
investigation).
I recognize, as the PCRA court remarked, that it is not always necessary for the
defense to consult an expert to evaluate and counter scientific evidence proffered by the
prosecution. See Treiber, Nos. 842 A&B of 2001, slip op. at 30-31 (citing
Commonwealth v. Chmiel, 612 Pa. 333, 387-88, 30 A.3d 1111, 1143 (2011),
Commonwealth v. Copenhefer, 553 Pa. 285, 307, 719 A.2d 242, 253 (1998), and
Commonwealth v. Smith, 544 Pa. 219, 238, 675 A.2d 1221, 1230 (1996)). Each of the
decisions referenced by the post-conviction court, however, focuses on whether the
defense had some other reasonable avenue for confronting the Commonwealth’s
(…continued)
Halverson’s or Mr. Basten’s reports with a defense expert. See, e.g., N.T., Aug. 10,
2009, at 158-60.
[J-49-2013] [M.O. – Eakin, J.] - 18
evidence, often reducing to the question of whether counsel has prepared for and
engaged in effective cross-examination. See, e.g., Chmiel, 612 Pa. at 388, 30 A.3d at
1143. In the present case, however, trial counsel engaged in no cross-examination
whatsoever of either Ms. Halverson or Mr. Basten relative to their expert opinions
regarding the Golden2 sample. Rather, the lawyer affirmatively conceded the validity of
the prosecution’s damaging scientific evidence and expert testimony. See N.T., Oct. 3,
2002, at 132; N.T., Oct. 7, 2002, at 75.
Certainly, there are circumstances in which the defense has little or no choice but
to acknowledge the validity of scientific evidence presented by the government,
particularly where such evidence is indisputable. In the present case, however, both the
reports of Ms. Halverson and Mr. Basten contained ample grounds for further, critical
review. For example, Ms. Halverson’s observation that the Golden2 sample was
“suboptimal” and produced “low template quantity or quality” DNA merited exploration.
Reports of Joy Halverson, DVM, dated Oct. 2, 2001, and Aug. 26, 2002, Petitioner’s
Evidentiary Hearing Exhibits, Vol. IX, tabs 2 & 3. This information was not previously
available to counsel when he spoke with Ms. Eggleston and was of a type which would
implicate further investigation by a competent defense attorney intent on ascertaining
deficiencies in the Commonwealth’s proofs. Significantly, in this regard, this Court’s
previous approval of the use of forensic DNA evidence in the courtroom was dependent
of high template quality producing results amenable to objective assessment. See
Commonwealth v. Crews, 536 Pa. 508, 520, 640 A.2d 395, 401 (1994) (positing that
“scientists are almost certain to agree . . . that two DNA samples do or do not match at
a given number of critical loci”). The Court has not reviewed the scientific controversies
over tests of minimal or degraded DNA samples producing marked stochastic effects
increasing the subjectivity of typing assessments, such as pervasive allelic dropout
[J-49-2013] [M.O. – Eakin, J.] - 19
throughout an electropherogram. See generally Erin Murphy, The Art in the Science of
DNA: A Layperson’s Guide to the Subjectivity Inherent in Forensic DNA Typing, 58
EMORY L.J. 489, 503-08 (2008) (discussing such phenomena in terms of the introduction
of subjectivity in DNA analysis).
Furthermore, Ms. Halverson’s more-than-thousand-fold adjustment to her
probability estimates based on her continuing education efforts fails to inspire
confidence in her results and further demonstrates cause for probing inquiries. See
Supplemental Report of Joy Halverson, DVM, dated Aug. 26, 2002, Petitioner’s
Evidentiary Hearing Exhibits, Vol. IX, tab 3. Remarkably, as well, data disclosed on the
face of Basten’s report showed different alleles at a particular locus as between the
Golden2 sample and the reference sample taken from the dog Janie’s blood, thus
manifesting an affirmative exclusion of Janie as the source of Golden2. See Report of
Christopher J. Basten, dated June 19, 2002, Petitioner’s Evidentiary Hearing Exhibits,
Vol. XI, tab 12. This had to be explained by the Commonwealth’s experts, on post-
conviction, as a mistake, see, e.g., N.T., Oct. 29, 2009, at 114, but, again, there can be
little excuse for a defense attorney’s inapprehension that the prosecution’s own
evidence, on its face, failed of its own accord. Cf. Driscoll v. Delo, 71 F.3d 701, 709
(8th Cir. 1995) (explaining that “a reasonable defense lawyer would take some
measures to understand the laboratory tests performed and the inferences that one
could logically draw from the results” linking his client to a murder weapon).
For its part, the United States Supreme Court has recognized that “[c]riminal
cases will arise where the only reasonable and available defense strategy requires
consultation with experts or introduction of expert evidence.” Harrington v. Richter, 562
U.S. 86, 106, 131 S. Ct. 770, 788 (2011). Where, as here, the defense possessed no
other reasonable strategy relative to a strongly-incriminating line of the prosecution’s
[J-49-2013] [M.O. – Eakin, J.] - 20
evidence, it was plainly unreasonable for counsel to forego further analysis and/or
consultation relative to the Halverson/Basten reports.
I also agree with Appellant’s assertion that the Commonwealth’s low-template,
canine DNA evidence strongly implicated a Frye challenge. In terms of novelty, this
Court has explained that it is appropriate to ascribe a broad meaning to the concept,
subsuming scenarios in which it is asserted that an expert has not applied an accepted
scientific methodology in a conventional fashion in reaching his or her conclusions. See
Betz v. Pneumo Abex, LLC, 615 Pa. 504, 545, 44 A.3d 27, 53 (2012). Here, counsel
could have presented evidence that Ms. Halverson’s extension of methodologies used
for human DNA typing to the analysis of low-template, canine DNA entailed her
utilization of unstudied proprietary genetic markers, a database of convenience, and
unverified statistical measures for accounting for the partial profile obtained. Each of
these has sufficiently novel attributes to implicate careful judicial review as predicate to
admission into evidence in a criminal case. Accord Leuluaialii, 77 P.3d at 1197
(indicating that “[a] Frye hearing was absolutely necessary” relative to Ms. Halverson’s
methodology for canine DNA typing).
As to general acceptance, this Court has looked to treatment by other
jurisdictions, meaningful experiments and studies published in scientific journals
allowing for critical review by other scientists, and the resolution of controversies
through ongoing scientific discourse. See, e.g., Blasioli, 552 Pa. at 166-68, 713 A.2d at
1126-27. At the time of Appellant’s trial, however, there was a dearth of case law
relative to the use of forensic canine DNA in the courtroom. See Leuluaialii, 77 P.3d at
1197 (“There are no published United States cases that involve the use of canine DNA
markers for forensic purposes or examine the validity of the specific markers used
here.”). Years after Appellant’s trial, researchers still describe the emergence of a
[J-49-2013] [M.O. – Eakin, J.] - 21
forensic community performing non-human analysis as a “pioneering area.” van Asch &
Pereira, State-of-the-Art and Future Prospects of Canine STR-Based Genotyping, OPEN
FORENSIC SCI. J. at 45-46, 49 (observing that canine DNA typing has had a “difficult
coming of age,” and “canine derived evidence is not frequently analyzed in forensic
caseworks and, consequently, is seldom reported in the literature”); Edward J.
Imwinkelried, Canine DNA, 46 No.4 CRIM. L. BULLETIN 6 (2010) (explaining, long after
Appellant’s trial, that “there are major differences between the state of the art of human
DNA testing and the current state of canine DNA testing”). In contrast to the widely
used and studied genetic markers and database employed for human DNA typing, i.e.,
the Federal Bureau of Investigation’s Combined DNA Index System, 17 by her own
admission Ms. Halverson was the only person in the United basing forensic DNA testing
on the specific canine typing infrastructure she conceived. See N.T., Oct. 29, 2009, at
67-68. Furthermore, the discrete methods used by Ms. Halverson and Mr. Basten for
accounting for low-template DNA were, at the very least, controversial. See, e.g.,
United States v. McCluskey, 954 F. Supp. 2d 1224, 1276-77 (D.N.M. 2013) (“When
there is too small a sample, the DNA testing . . . may yield unreliable and non-
reproducible results because of the significant increase in stochastic effects.” (citing
Peter Gill, Application of Low Copy Number DNA Profiling, 42(3) CROATIAN MED. J. 229,
229–30 (2001))).18
17
See generally State v. Russell, 882 P.2d 747, 759-68 (Wash. 1994) (recognizing that
DNA identification of humans is generally accepted by the scientific community and is
based on knowledge about the extent of genetic variation in humans; identification of
discrete, variable polymorphic DNA loci; and accurate probability estimates, all of which
were developed through years of extensive research and validation by the scientific
community).
18
Accord David L. Faigman, Jeremy A. Blumenthal, Edward K. Cheng, Jennifer L.
Mnookin, Erin E. Murphy & Joseph Sanders, 4 MOD. SCI. EVID. §31:32 (2013-2014) (“It
is fair to say that [low copy number DNA] typing is the subject of great dispute among
(continued…)
[J-49-2013] [M.O. – Eakin, J.] - 22
In this case, the PCRA court was simply wrong to equate general acceptance of
a closely-studied and controlled infrastructure for routine typing of adequate samples of
human DNA with general acceptance of an unstudied, uncontrolled infrastructure for
analyzing low-template, non-human DNA. To the degree the PCRA court believed that
the genetic markers used by Ms. Halverson, her database, and her approaches to
addressing low-template DNA analysis were simply minor scientific premises which did
not merit scrutiny for legitimate general acceptance in their own right, I disagree.
Accord DAVID H. KAYE, DAVID E. BERNSTEIN & JENNIFER L. MNOOKIN, NEW W IGMORE: A
TREATISE ON EVIDENCE §9.2.3 (2014) (discussing the differences between major
premises and case-specific expert testimony relative to Frye inquiries).
Even if the Commonwealth could have surmounted all of the above, the fact
remains that its statistical expert was only able to justify the “general acceptance” of his
(…continued)
some of the leading lights in the forensic community.”); Broun, 1 MCCORMICK ON
EVIDENCE §205 (“Even with existing systems, efforts to push PCR to its limits in copying
only DNA fragments from a few cells (low template or touch DNA samples) have
generated controversy.”); Bruce Budowle, Arthur J. Eisenberg & Angela van Daal, Low
Copy Number Typing Has Yet to Achieve “General Acceptance”, FORENSIC SCIENCE
INT’L: GENERAL SUPPLEMENT SERIES 2 551-52 (2009); cf. I. Pfeiffer, I. Volkel, H. Taubert
& B. Brenig, Forensic DNA-Typing of Dog Hair: DNA-Extraction and PCR Amplification,
141 FOR. SCI. INT’L 149, 149-50 (2004) (“DNA-typing from one up to 10 dog hairs is
often problematic in forensic science. Single dog hairs contain very small amounts of
DNA and have often no roots at all. Genetic typing from nuclear DNA isolated from dog
hair often failed.”).
Even courts which have been willing to allow the admission of low-template DNA
analysis as sufficiently reliable under the Daubert regime generally rely on the close
adherence to nationally recognized protocols and standards in determining reliability.
See, e.g., United States v. Williams, 979 F. Supp. 2d 1099, 1103 (D. Haw. 2013)
(explaining that “the existence of null alleles is a known phenomenon which ‘is taken
into account on a case-by-case and profile-by-profile basis and is managed according to
validated methodology, standard operating procedures, [Scientific Working Group on
DNA Analysis Methods] guidelines and the FBI Quality Assurance Standard”) (citation
omitted). Ms. Halverson, however, did not consult such standards. See N.T., Oct. 29,
2009, at 91.
[J-49-2013] [M.O. – Eakin, J.] - 23
proprietary method for accounting for missing information in the partial DNA profile
obtained from Golden2 and peculiarities of the canine population infrastructure based
on a paper he first published three years after Appellant’s trial. See N.T., Oct. 29, 2009,
at 181. Indeed, putting aside that the mere publishing of a single paper in a scientific
arena outside mainstream scrutiny cannot establish general acceptance in the first
instance, as well as the question of whether the document contained sufficient
information to support critical scientific inquiry into the underlying methodology given the
proprietary character which was maintained, see N.T., Oct. 29, 2009, at 187, 19 the
expert’s concession as to the timing of his contribution to the literature demonstrates
very succinctly that his and Ms. Halverson’s claims to general acceptance three years
earlier were entirely hollow ones.
In my judgment, the previously-discussed concerns expressed by the National
Research Council are highly resonant, see NATIONAL RESEARCH COUNCIL,
STRENGTHENING FORENSIC SCIENCE IN THE UNITED STATES, A PATH FORWARD 85, and this
case serves as a ready example of the problem.20
In light of the above, I conclude that the PCRA court’s treatment of the Frye test
was unduly loose and requires correction. Moreover, the above analysis establishes
that Appellant’s claim of deficient stewardship relative to the failure to pursue a Frye
19
See generally State v. Schwartz, 447 N.W.2d 422, 428-29 (Minn. 1989) (holding that,
because a DNA laboratory failed to make information about its methodology and probes
publicly available, this prevented independent assessment of its methods, supporting a
determination that specific test results were inadmissible).
20
The report also stressed the need for accredited laboratories, certification for forensic
practitioners, and sound, standardized operational principles and procedures. See id. at
110. In this respect, I note that Ms. Halverson’s private laboratory was not accredited,
she enjoyed no relevant certification, the laboratory had no documented quality
assurance program in place, and she did not consult nationally-recognized forensic-
testing guidelines. See N.T., Oct. 29, 2009, at 60-62, 89, 91.
[J-49-2013] [M.O. – Eakin, J.] - 24
challenge to the Commonwealth’s low-template, canine DNA evidence has arguable
merit and trial counsel lacked a reasonable strategy for foregoing such attack. See
supra note 15 (referencing the criteria governing resolution of claims of deficient
stewardship, in terms of arguable merit, lack of reasonable strategy, and prejudice).
Regarding prejudice, it might be said that, regardless of the merits of a Frye
challenge as discussed above, the trial judge would have rejected such attack, since the
same judge did so at the post-conviction stage. While this argument carries some
weight so far as it goes, it does not account for the post-conviction court’s errors as
delineated above -- had those mistakes also been made prior to trial, they should have
been corrected at the direct-appeal stage.
As to the prejudicial impact of the Commonwealth’s low-template, canine DNA
evidence at trial, Appellant asserts that this was the only forensic evidence linking
Appellant to the arson and murder, while highlighting this Court’s recognition that expert
testimony and scientific proofs carry the potential to “assume a posture of mystic
infallibility in the eyes of a jury of laymen.” Topa, 471 Pa. at 232, 369 A.2d at 1282
(quoting United States v. Addison, 498 F.2d 741, 744 (D.C. Cir. 1974)). Further, and as
previously related, Appellant observes that the prosecution itself has explained that the
canine DNA evidence was “crucial to the Commonwealth to prove identity, intent and
lack of accident.” Commonwealth’s Reply to Defendant’s Post-Sentence/New Trial
Motion, dated December 3, 2002, at 2. Along these lines, Appellant develops, the trial
prosecutor reaffirmed the importance of this evidence on national television, asserting
that “this type of DNA match [is] the type of thing that a jury wraps itself around,” and
that the evidence “destroyed any kind of defense that the Defendant tried to muster,”
[J-49-2013] [M.O. – Eakin, J.] - 25
N.T., June 7, 2011, at 25, 21 and that the trial prosecutor maintained precisely such
position in the post-conviction proceedings. See id. It is Appellant’s core assertion that,
because his attorney failed to advance an obvious challenge to the evidence, he was
deprived of the ability to capitalize on the most positive piece of exculpatory evidence
available to him. Finally, Appellant contends that, “without the dog DNA evidence, the
Commonwealth is left with a weak circumstantial case: there was no confession or
eyewitnesses, and the testimony of key Commonwealth witness Jamie Pianta is
questionable.” Brief for Appellant at 12.
Again, neither the PCRA court nor the Commonwealth has addressed prejudice
in terms of the degree of impact of the low-template, canine DNA evidence on the
verdict.
Initially, I differ with Appellant’s position that, in the absence of the canine DNA
evidence, the Commonwealth presented only a weak, circumstantial case against him.
In point of fact, I recognize that the prosecution presented strong circumstantial
evidence concerning his guilt. Furthermore, as Appellant acknowledges, the
21
Appellant’s case was profiled on the Animal Planet Media Network television show
“Animal Witness,” which has summarized its coverage as follows:
Investigators in tiny Mill Creek, Pennsylvania uncover
evidence of arson after a house fire claims the lives of a
toddler and the family Rottweiler. From the start, detectives
target homeowner Stephen Treiber as their chief suspect.
But they'll need something stronger than just a hunch to
make their case. To do this, they'll need to rely on an
unusual witness to solve the crime. From beyond the grave,
Treiber's own dog - a Rottweiler named Janie - will help
investigators and prosecutors link her owner to the
devastating fire through DNA evidence.
ANIMAL PLANET, Programs: Animal Witness, http://press.discovery.com/asia-
pacific/apl/programs/animal-witness/ (last viewed July 16, 2014).
[J-49-2013] [M.O. – Eakin, J.] - 26
prosecution adduced testimony from Jamie Pianta attesting that, several days before
the fire, he overheard Appellant discussing his plans commit arson with Ms. Riddle’s
son, Eric Keith. See N.T., Oct. 2, 2002, at 207-10. According to Pianta, Appellant said
that he desired to murder Jessica on account of his child-support obligation. See id. at
209-10.
Despite my appreciation that the Commonwealth’s case against Appellant was
formidable, undisputed evidence on this record establishes the prejudicial impact of the
Commonwealth’s unchallenged low-template, canine DNA evidence in terms of greatly
elevating the Commonwealth’s proofs of identity and pre-planning in the eyes of a lay
jury and gutting the defense, as recognized by the trial prosecutor publicly and on the
post-conviction record. 22 Such impact of evidence which the Commonwealth itself
regarded as “critical” to the prosecution is too severe to permit the conclusion that
Appellant received a fair trial despite his attorney’s unreasonable failure to mount a
challenge to the evidence.23
22
The majority’s depiction of the prosecutor’s public commentary, in terms of his merely
“wax[ing] eloquent on national television,” Majority Opinion, slip op. at 19 n.13, fails to
account for the trial prosecutor’s sworn reaffirmation of his account on the post-
conviction record. See N.T., June 7, 2011, at 25.
23
The force of Pianta’s testimony, relative to the prejudice assessment, is diminished to
a substantial degree given that, after Appellant’s trial, the Commonwealth dismissed
conspiracy charges against Keith, based on “serious problems” with Pianta’s testimony.
The Commonwealth did so despite Keith’s own videotaped confession to aiding
Appellant in preparing for and perpetrating arson and murder. See N.T., Aug. 4, 2011,
at 12-13, 18.
Parenthetically, Appellant attempted to call Keith as a witness for the defense at his
trial, but Keith asserted his Fifth Amendment right and did not testify. See N.T., Oct. 5,
2002, at 3-7. Keith’s examination on post-conviction was curtailed by the PCRA court,
but he essentially recanted his confession, testifying that he had pled guilty to gain
leniency on the advice of his counsel, he told law enforcement officials what they
wanted to hear, and he had done nothing wrong. See N.T., Oct. 21, 2010, 74-75, 78-
79. Keith also affirmed the accuracy of the content of a written declaration containing a
(continued…)
[J-49-2013] [M.O. – Eakin, J.] - 27
The United States Supreme Court has explained that “[s]ome errors [on the part
of a defense attorney] will have had a pervasive effect on the inferences to be drawn
from the evidence, altering the entire evidentiary picture, and some will have had an
isolated, trivial effect.” Strickland, 466 U.S. at 695-96, 104 S. Ct. at 2069. The trial and
post-conviction records make clear that trial counsel’s dereliction in the present case
was of the former variety, thus undermining confidence in the verdict and supporting the
conclusion that Appellant has demonstrated, by a preponderance of the evidence, that,
but for the ineffectiveness of his counsel, there is a reasonable probability of a different
outcome. See generally Commonwealth v. Pierce, 515 Pa. 153, 158-60, 527 A.2d 973,
975-76 (1987) (elaborating on the prejudice component of the ineffectiveness inquiry).
In summary, it is my considered position that relief in the form of a new trial is
warranted in this capital post-conviction appeal based on the failure of Appellant’s
lawyer, at trial, to challenge the Commonwealth’s presentation of low-template, canine
DNA evidence.
(…continued)
fuller recantation of his confession. See id. at 62; Petitioner’s Evidentiary Hearing
Exhibits, Vol. XIV, tab 2.
[J-49-2013] [M.O. – Eakin, J.] - 28