Samuel Anstey, Petitioner Below, Petitioner v. David Ballard, Warden, Mt. Olive
Correctional Complex, Respondent Below, Respondent
Supreme Court No. 15-0067 FILED
June 2, 2016
released at 3:00 p.m.
Chief Justice Ketchum, dissenting: RORY L. PERRY, II CLERK
SUPREME COURT OF APPEALS
OF WEST VIRGINIA
Oliver Wendell Holmes once stated: “Certitude is not the test of certainty. We have
been cock-sure of many things that were not so.”1
That statement encapsulates the error of the majority in affirming the denial of habeas
relief. The error is twofold. First, the majority fails to realize that the State’s investigating
witnesses, based on group-think and the quantity of the evidence rather than its quality,
concluded, unscientifically, that the fire was intentionally set. Second, the majority has
accepted the determination of the habeas court that Anstey’s assertion of newly-discovered
evidence, i.e., the advancement in fire science represented by NFPA 921, as amended, does
not even warrant an omnibus habeas corpus hearing.
Given the complexity of the evidence, Anstey cannot reasonably present grounds for
a new trial unless an omnibus habeas corpus hearing is conducted. Therefore, the issue is not
whether this Court should award him a new trial. Instead, the issue is whether Anstey has
made a case for an evidentiary hearing and whether the habeas court abused its discretion in
1
Oliver Wendell Holmes, Natural Law, 32 Harv. L. Rev. 40 (1918).
1
denying him one. I am of the opinion that Anstey meets the standards which would entitle
him to an omnibus habeas corpus hearing. I, therefore, dissent.
I. Standards For Conducting
An Omnibus Habeas Corpus Hearing
The habeas court compared an eleven-day trial with the affidavits of two experts
submitted by Anstey and summarily concluded, without findings of fact and conclusions of
law, that “no testimony or other evidence is necessary.” However, Rule 9(a) of the Rules
Governing Post-Conviction Habeas Corpus Proceedings in West Virginia requires as
follows:
If the petition is not dismissed at a previous stage in the proceedings,
the circuit court, after the answer is filed, shall, upon a review of the record,
if any, determine whether an evidentiary hearing is required. If the court
determines that an evidentiary hearing is not required, the court shall include
in its final order specific findings of fact and conclusions of law as to why an
evidentiary hearing was not required.2
2
While a habeas court has discretion in deciding whether a hearing should be
conducted, the threshold warranting a hearing is low. W.Va. Code, 53-4A-7 [2008] of the
Post-Conviction Habeas Corpus Act, states, in part:
If it appears to the court from said petition, affidavits, exhibits,
records and other documentary evidence attached thereto, or the return or
other pleadings, or any such record or records referred to above, that there
is probable cause to believe that the petitioner may be entitled to some
relief and that the contention or contentions and grounds (in fact or law)
advanced have not been previously and finally adjudicated or waived, the
court shall promptly hold a hearing and/or take evidence on the contention
or contentions and grounds (in fact or law) advanced, and the court shall
2
The habeas court had before it Anstey’s petition, the affidavits of his experts, the
State’s answer to the petition, and Anstey’s response to the State’s answer. The positions of
the parties were, thus, joined and in sharp conflict over the evidentiary value of NFPA 921
as newly-discovered evidence. Nevertheless, the habeas court unfairly determined in a single
paragraph without the requisite findings that an omnibus hearing would not be conducted.
The habeas court then extrapolated findings and conclusions on the merits of Anstey’s
petition. As a result, this Court is left, on appeal, “greatly at sea without a chart or compass”3
in the absence of an evidentiary transcript from which the ultimate findings of the habeas
court may be reviewed. It was an abuse of discretion and a denial of Anstey’s right to due
process not to conduct an omnibus habeas corpus hearing in this case.
II. The Mandate of the
West Virginia Legislature
During the 1995 trial, State’s expert Harold Franck referred to the NFPA standards
as follows: “So it is a set of guidelines that have been developed over the past few years, and,
hopefully, the NFPA wants to have those as national standards at some point.” (emphasis
added). Franck’s surmise came true. NFPA 921 became the national authority for standards
pass upon all issues of fact without a jury.
(emphasis added)
3
See State ex rel. HCR ManorCare, LLC v. Stucky, 235 W.Va. 677, 687, 776
S.E.2d 271, 281 (2015), citing Workman v. Workmen’s Compensation Comm’r, 160
W.Va. 656, 662, 236 S.E.2d 236, 240 (1977).
3
in fire science and investigation years later in 2000 upon its endorsement by the United States
Department of Justice.4
Since recognition by the Department of Justice, NFPA standards have been included
in the law of this State by the West Virginia Legislature. W.Va. Code, 29-3-5(b) [2010], of
the West Virginia Fire Prevention and Control Act, states in part: “Whenever any new or
revised code or standard is adopted by the fire codes published by the National Fire
Protection Association, the State Fire Commission may propose and promulgate revised rules
reflecting such updated codes and standards[.]” (emphasis added). See W.Va. Code, 29-3
16a [2012] (adopting NFPA standards for the installation of smoke detectors and sprinkler
systems in one- and two-family dwellings, including any “manufactured home”); Title 87 of
the Code of State Rules: § 87-1-2 [2014] (adopting NFPA’s National Fire Codes); and § 87
4-4 [2013] (adopting certain NFPA provisions with respect to the State Building Code). See
4
Since 2000, a number of jurisdictions have recognized that NFPA 921 is an
accepted reference, if not the “gold standard,” for fire investigators: Fireman’s Fund Ins.
Co. v. Canon U.S.A., Inc., 394 F.3d 1054, 1057-58 (8th Cir. 2005) (NFPA 921 is a reliable
standard which is endorsed professionally); U.S. v. Aman, 748 F.Supp.2d 531, 536 (E.D.
Va. 2010) (“NFPA 921 is sufficiently reliable to pass muster under Daubert.”); Tunnell v.
Ford Motor Co., 330 F.Supp.2d 707, 725 (W.D. Va. 2004) (Many courts have recognized
NFPA 921 as a peer reviewed and generally accepted standard.); McCoy v. Whirlpool
Corp., 214 F.R.D. 646 (D. Kan. 2003) (“The ‘gold standard’ for fire investigations is
codified in NFPA 921, and its testing methodologies are well known in the fire
investigation community and familiar to the courts.”); Farmland Mut. Ins. Companies v.
Chief Indus., Inc., 170 P.3d 832, 836 (Colo. Ct. App. 2007) (A number of courts have
held that NFPA 921 is an accepted reference for fire investigators).
4
also § 103-3-3.18 [2015] (pertaining to the State Fire Marshall, adopting standards and
requirements as published by the NFPA).
Significantly, Series 8 of Title 87, designated “Volunteer Firefighters’ Training,
Equipment, and Operating Standards” (§ 87-8-1, et seq.), references a number of NFPA
provisions concerning training levels, curriculum approval and equipment standards. The
unit initially responding to and investigating the trailer fire in this case was the Oak Hill
Volunteer Fire Department.
Syllabus point 5 of Smith v. W.Va. Human Rights Comm’n, 216 W.Va. 2, 602 S.E.2d
445 (2004), holds: “A regulation that is proposed by an agency and approved by the
Legislature is a ‘legislative rule’ as defined by the State Administrative Procedures Act,
W.Va. Code, 29A-1-2(d) [1982], and such a legislative rule has the force and effect of law.”
Accord syl. pt. 2, State ex rel. Maple Creative LLC v. Tincher, 226 W.Va. 118, 697 S.E.2d
154 (2010).
Accordingly, insofar as NFPA standards have been recognized by the West Virginia
Legislature, the State Fire Commission and the State Fire Marshall, NFPA standards enjoy
5
a higher status than merely an expert opinion on what the proper standard ought to be.5 The
NFPA standards necessarily include NFPA 921. In view of the extent to which NFPA
standards permeate West Virginia law, it cannot be said that NFPA 921 is excluded. For
example, given the requirement of W.Va. Code, 29-3-16a [2012], that smoke detectors or
sprinkler systems are to be installed pursuant to NFPA standards in manufactured homes, it
makes no sense to ignore NFPA 921 where a fire in a manufactured home resulting in death
has occurred.
The incorporation of NFPA standards into West Virginia law, including NFPA 921,
was never addressed by the habeas court. This State’s incorporation of those standards
constitutes an additional reason why an omnibus habeas corpus hearing should have been
conducted.6
5
Even before recognition by the Department of Justice in 2000, the fire codes
published by the NFPA were a part of the law of this State. See W.Va. Code, 16-3A-2(a)
[1985] (requiring the State Department of Health to establish a list of hazardous materials
and give consideration to “any list, publication, regulation, report, guideline or other
compilation” of the NFPA); W.Va. Code, 16-5N-7(d) [1997] (requiring residential care
communities to comply with the “life safety code” promulgated by the NFPA and adopted
by the State Fire Commission.); W.Va. Code, 21-3-3a [1965] (adopting standards of the
National Electric Code as promulgated by the NFPA); Reed v. Phillips, 192 W.Va. 392,
395, 452 S.E.2d 708, 711 (1994) (The State Fire Commission “has adopted the ‘National
Fire Codes’ published by the National Fire Protection Association (‘NFPA’) as a part of
the West Virginia Fire Code.”); Teller v. McCoy, 162 W.Va. 367, 375, 253 S.E.2d 114,
120 (1978) (The State Fire Commissioner’s regulations include the National Fire Code
published by the National Fire Protection Association).
6
In the absence of scientific investigation guidelines, “many arson investigators
believed that what they did was more of an art than a science - a blend of experience and
6
III. Certitude Is Not Certainty
Without doubt, the fire investigation conducted by the volunteer firefighters at the
trailer is questionable. The appendix record before this Court includes no reports or exhibits
of any kind with regard to the underlying trial. All we have is the trial transcript. However,
it is undisputed that the toaster and the trailer’s breaker switches were manipulated during
the first stages of the investigation. One of the State’s witnesses in charge of the Oak Hill
Volunteer Fire Department’s investigation unit moved the toaster’s plunger up and down
before photographing the toaster.7 Moreover, although volunteer firefighters had flipped the
trailer’s main electrical breaker to the off position, that same witness flipped three other
breaker switches back and forth to the off and on positions to determine whether those
switches had tripped during the course of the fire. His testimony in that regard seems
uncertain:
Q. Okay. Did you put them back basically where they had been, or did you put
them in a different position?
intuition.” David Grann, “Trial by Fire,” The New Yorker (Sept. 7, 2009). “People
investigated fire largely with a flat-earth approach . . . . It looks like arson
therefore, it’s arson.” Id.
7
The trial transcript supports Anstey’s assertion that this witness, who noticed the
toaster on the countertop and saw the plunger in the down position “raised the plunger
until it released” and then pushed it back down. Moreover, the witness moved or lifted
the toaster to examine it, placed it back “in what he believed was its original position,”
and only then photographed the toaster. Anstey asserted: The mishandling of the toaster
“is in direct contravention to the procedures that NFPA 921 requires for the preservation
of evidence.”
7
A. I would say I put them back to the off position. I flipped them off to
– as sort of a reminder when I was doing a report or something what three
breakers were tripped, in my opinion.
The evidence at trial concerning the electrical circuitry in the trailer was directly
relevant to whether a living room lamp short-circuited resulting in the fire and tripping a
breaker switch which, in turn, shut off the power to the electric smoke alarm.
In addition, unlike in the living room and kitchen areas of the trailer, the ceiling in the
bedroom where the victim was found was still in place. Though never conclusively
determined, Anstey’s evidence indicated that ceiling tile was found among the debris on the
vent in the victim’s bedroom. That evidence supported Anstey’s assertion that the debris was
tracked into the bedroom by volunteer firefighters and that, therefore, there was no second
fire in the trailer. Responding to the State’s two-fire theory, Anstey maintains that the State’s
determination that a separate fire occurred in the victim’s bedroom was conclusory and “in
direct contravention with NFPA 921's requirement that all data pertaining to the
determination of a fire’s origin be analyzed.”
Finally, it is worth noting that Anstey called two witnesses during the underlying trial
who testified that the victim herself had threatened to set the trailer on fire. One of those
witnesses, an acquaintance of the victim, stated: “She said on several occasions she would
set fire to herself and burn everything up.”
8
A case strikingly similar to the current matter was before the Court of Appeals of
Indiana in Bunch v. State, 964 N.E.2d 274 (Ind. Ct. App. 2012). In Bunch, the defendant was
convicted of the murder of her son involving a fire in their mobile home. The son was found
in a bedroom and died at the scene from smoke inhalation. During the defendant’s 1996 trial,
the State relied largely on expert testimony which described the presence of an accelerant and
two separate fires - one in the bedroom where the son was found and another in a doorway.
In Bunch, the State used its two-fire theory to assert that the fires were intentionally set. The
defendant’s expert witness testified that the mobile home fire should have been classified as
undetermined. The murder conviction was affirmed on direct appeal.
In 2006, the defendant, in Bunch, filed a petition for post-conviction relief claiming
newly-discovered evidence in the form of advances in the field of fire science. Unlike the
present case, the habeas court in Bunch conducted an evidentiary hearing during which the
defendant presented the testimony of experts who discussed the scientific advances in
relation to the mobile home fire. The habeas court, however, denied relief on the basis that
the conclusions of the defendant’s experts, that the fire was undetermined, was the same as
that of the defendant’s expert in the underlying trial. The habeas court, in Bunch, further
concluded that the testimony of the defendant’s experts merely tended to impeach the State’s
witnesses and did not constitute newly-discovered evidence.
9
The Court of Appeals of Indiana, in Bunch, reversed the decision of the habeas court
and awarded the defendant a new trial, concluding that the new evidence was neither
cumulative nor solely for purposes of impeachment. One of the grounds of reversal was the
testimony of the defendant’s post-conviction expert witness, Jamie McAllister. McAllister
concluded that the fire started accidentally in a confined space (between the bedroom ceiling
and the roof) and was not caused by an accelerant. McAllister’s conclusion was based, in
part, on the toxicology and autopsy reports indicating that the son, inhaling smoke and soot,
reached an 80% carbon monoxide saturation level without thermal damage to his respiratory
system, suggesting an under-ventilated fire rather than a well-ventilated fire in an open room.
Granting the defendant post-conviction relief, the Court of Appeals in Bunch stated:
McAllister also testified the consideration of a fire victim’s
physiological condition did not become a recognized component of the fire
origin analysis until after 2001, the first time a chapter on the fire-related
deaths appeared in the National Fire Protection Association 921 Guide for Fire
and Explosion Investigations (“NFPA 921"), which is “a peer reviewed and
generally accepted standard in the fire investigation community.” Travelers
Prop. & Cas. Corp. v. General Elec. Co., 150 F.Supp.2d 360, 366 (D. Conn.
2001). * * * The post-conviction court’s finding that [the defendant’s]
post-conviction evidence is just “different packaging” for the same conclusion
does not give appropriate due to the science which has emerged since [the
defendant’s] trial to support that conclusion.
964 N.E.2d at 287, 289.8
8
In Bunch, the Court of Appeals of Indiana noted that the Oklahoma, Nebraska
and Arizona legislatures passed resolutions supporting judicial review of cases in which
faulty science is alleged to have contributed to an arson conviction, with the Oklahoma
10
In Bunch, a post-conviction, evidentiary hearing was conducted. During the hearing,
the defendant was able to develop the record with new expert testimony of a complex nature
concerning the fire in the mobile home. Consequently, an expanded record and
comprehensive order from the post-conviction court were before the Court of Appeals of
Indiana for review. By contrast, no evidentiary hearing was conducted by the habeas court
with regard to Anstey’s petition which, as in Bunch, alleged newly-discovered evidence
pursuant to an amended NFPA 921. The principles set forth in NFPA 921 played no part in
Anstey’s 1995 trial.9
resolution stating, in part, that “the Oklahoma State Senate urges the judicial branch, law
enforcement agencies, and other relevant government entities in Oklahoma to employ
NFPA 921 when conducting fire investigations.” 964 N.E.2d at 288
9
Here, the habeas court concluded that the testimony of Anstey’s new experts
would be the same as the testimony of his experts in the underlying trial. Moreover, the
use of the new experts in a new trial would serve no other purpose than to impeach the
State’s witnesses. Both of those conclusions were rejected in Bunch by the Court of
Appeals of Indiana which had the benefit of a fully developed record. The post-
conviction record present in Bunch highlights, a fortiori, the absence of such a record in
this proceeding.
Furthermore, as long as the testimony of Anstey’s experts otherwise meets the
elements of newly-discovered evidence, it is of no moment that their testimony will
impeach the State’s witnesses. Under the syllabus point in State v. Frazier, 162 W.Va.
935, 253 S.E.2d 534 (1979), a new trial “will generally be refused when the sole object of
the new evidence is to discredit or impeach a witness on the opposite side.” That
consideration, however, is not to be mechanically applied. In State v. Stewart, 161 W.Va.
127, 239 S.E.2d 777 (1977), this Court stated that while new impeachment evidence “will
not normally mandate a new trial,” a new trial should be granted where all other
considerations are met and the new impeachment evidence concerns the key prosecution
witness. 161 W.Va. at 136, 239 S.E.2d at 783. In Stewart, this Court stated that it
disapproved of any interpretation “to the effect that a new trial will never be warranted if
the newly-discovered evidence merely impeaches a witness.” Id. (emphasis added)
11
Manifestly, the publication of NFPA 921 and its periodic amendments do not, per se,
constitute newly-discovered evidence in the absence of a nexus between its provisions and
the facts of a particular case. Here, the habeas court summarily concluded that “no testimony
or other evidence is necessary.” I disagree. The granting of an omnibus habeas corpus
hearing was warranted by the following: (1) the standards for conducting a hearing under the
habeas rules and the Post-Conviction Habeas Corpus Act, (2) the recognition of NFPA 921
by the West Virginia Legislature, (3) the complexity and scientific aspects of the trial and
post-trial evidence, and (4) the problems concerning the investigation at the scene.10
IV. The NFPA 921 Affidavits
Ultimately, the question in this case is whether Anstey’s affidavits from two new
experts on fire investigation constitute newly-discovered evidence and justify an omnibus
habeas corpus hearing. The experts are Dr. Gerald Hurst, a consulting chemist on fires and
explosions, and Mark Goodson, a consulting engineer. Obviously, the fact that Hurst and
Goodson’s conclusions may coincide with the conclusions of Anstey’s experts at trial does
not mean that a jury will return the same verdict as before. As Anstey suggests, if a similar
conclusion about the fire is more logically sound and can be shown to be scientifically valid
10
In at least one instance, the expert testimony of the defense during the 1995 trial
was problematic. Anstey’s expert in electrical engineering prepared a “Toaster Circuit
Diagram” concerning the toaster found in the trailer. At trial, the expert acknowledged
that one of the toaster’s components of some importance shown in the diagram was
inaccurately depicted.
12
based on NFPA 921 criteria, as amended, the conclusion is likely to be more persuasive and
readily acceptable by a jury.11
Hurst’s affidavit notes that, since its initial publication in 1992, new editions of NFPA
921 have been issued every three or four years and that the 2011 edition is the current
version. Concluding that the trailer fire remains undetermined, Hurst indicates that “to a
reasonable degree of scientific certainty” the State’s investigation “did not conform to
recommended fire investigative protocol, was not conducted in a methodical and reliable
fashion, and did not utilize the scientific method to determine origin and causation.”
Specifically, Hurst determined that the State’s hypothesis, that a fire originated with the
toaster and also in the victim’s bedroom, was not scientifically valid. Instead, according to
Hurst, the State’s hypothesis fell victim to “expectation bias,” rather than validation through
11
NFPA 921 sections 1.3 and 4.1, included in the appendix record, state in part:
1.3 These procedures represent the judgment developed from the
NFPA consensus process system that if followed can improve the
probability of reaching sound conclusions. Deviations from these
procedures, however, are not necessarily wrong or inferior but need to be
justified.
4.1 The use of a systematic approach often will uncover new factual
data for analysis, which may require previous conclusions to be
reevaluated.
(emphasis added)
13
the scientific method.12
Mark Goodson notes in his affidavit that NFPA 921 incorporates the scientific method
into the field of fire investigation and has become the standard for assessing the reliability
of expert testimony. With regard to the trailer fire, Goodson concluded:
It is clear based upon my review of the transcripts of the State’s origin
and cause witnesses that they did not follow the scientific method in
connection with their investigation of the trailer fire. Rather, the State’s
experts relied upon outdated methods that had permeated the fire investigation
community for years prior to the time of their investigation and testimony, and
those methods are no longer accepted within the fire investigation community
today.
Attached to Hurst and Goodson’s affidavits is an appendix which lists the materials,
in compact disc form, provided for Hurst and Goodson’s review. The materials consist
primarily of testimony and exhibits from Anstey’s 1995 trial. Although a large quantity of
material was provided, Goodson’s affidavit suggests that the data is not a reliable source to
12
For example, Hurst observed:
Mr. Franck testified that the presence of patterns inside the toaster’s
cover indicated that the toaster’s power cord had been stuffed inside the
toaster. There is no evidence beyond Mr. Franck’s opinion-based assertion,
however, in support of the conclusion that the cord was stuffed up inside
the toaster. Although such an observation would have been unusual and
worthy of noting and photographing, neither of the fire department officials
who first identified and seized the toaster at the scene made any mention of
the cord being inside the toaster.
14
use in trying to determine how the fire started.13 For example, Goodson states in his
affidavit: “It is standard engineering practice to x-ray breakers from circuits that feed
suspected fire causative appliances (in this case, the toaster) and to test the breakers
electrically for proper functioning. This was not done.”
It has been said that expert opinion “is only an ordinary guess in evening clothes.”14
While that statement appears descriptive of the evidence in Anstey’s 1995 trial, it is not
reasonably applicable to the new evidence, backed up by NFPA 921, which Anstey seeks to
present in this habeas corpus proceeding. At the very least, the affidavits of Hurst and
Goodson rightly form the basis for granting an omnibus habeas corpus hearing.
V. Conclusion
In my view, Anstey has shown newly-discovered evidence in the form of the
advancement in fire science and arson investigation in the intervening years since his
conviction. The advancement is represented by the National Fire Protection Association 921
13
The appendix indicates that the testimony of firefighters, fire investigators and
expert witnesses, as well as exhibits, for both the prosecution and the defense were
provided to Hurst and Goodson. One of the compact discs contains in excess of eighty
exhibits, mostly photographs. Among the many items photographed are the kitchen area;
the toaster with the plunger in the down position; the inside of the toaster; the coffee pot,
microwave and stove; ceiling damage in the living room; the lamp in the living room;
smoke and heat damage; Anstey’s bedroom; the victim’s bedroom; the breaker box; and
debris in the victim’s bedroom.
14
Earl M. Kerstetter, Inc., v. Commw., 404 Pa. 168, 171 A.2d 163, 165 (1961).
15
Guide for Fire and Explosion Investigations (NFPA 921). NFPA 921, as amended, is now
recognized by the Department of Justice and by this State through the mandate of the West
Virginia Legislature.
Moreover, Anstey has correctly shown that the habeas court’s refusal to grant him an
evidentiary omnibus habeas corpus hearing prevented him from demonstrating the substantial
impact of NFPA 921 on the expert testimony presented during his 1995 trial. The evidence
Anstey proposes to introduce may very well place his conviction in a different light by
exposing the conclusions of the State’s witnesses as objectively unreliable. However, he
cannot present that evidence as grounds for a new trial unless a hearing is conducted. The
circuit court abused its discretion in denying Anstey the opportunity to develop the record.
Due process requires an omnibus habeas corpus hearing in this case.15 Therefore, I
respectfully dissent.
15
On February 6, 1998, Anstey filed an unrelated petition for a writ of habeas
corpus in the Circuit Court of Fayette County styled State ex rel. Anstey v. Trent, 98-C-46
(1998). The circuit court dismissed the petition on February 11, 1998. Anstey appealed
the dismissal to this Court, alleging that the habeas court abused its discretion by denying
him “an opportunity to develop his contentions at an omnibus hearing.” Anstey stated
that numerous errors in his underlying trial needed to be developed, including
disqualification of the trial judge, prosecutorial misconduct, and ineffectiveness of
defense counsel. Anstey’s appeal to this Court from the denial of habeas relief was
refused on December 16, 1998, and his motion to reconsider this Court’s refusal was
denied on January 21, 1999.
16