IN THE COURT OF CRIMINAL APPEALS
OF TEXAS
NO. WR-78,106-01
EX PARTE DOUGLAS TYRONE ARMSTRONG, Applicant
ON APPLICATION FOR A WRIT OF HABEAS CORPUS
FROM HIDALGO COUNTY
N EWELL, J., filed a dissenting opinion in which H ERVEY, J.,
joined.
The central dispute in this case is over what an applicant must prove
to succeed on an ineffective assistance claim based upon the failure to
investigate possible expert testimony on the mitigation special issue in a
death penalty case. Applicant has presented evidence from two expert
witnesses from outside of Texas that, according to the Court, provide new
mental health testimony that was not presented at trial. And, according
to the Court, this new testimony creates a reasonable probability of a
different outcome had it been presented at trial. The trial court found the
testimony credible, but it also found that Applicant failed to establish that
Armstrong Dissent – 2
any experts were available to testify because the trial court would not
have granted funds to pay for these experts over qualified local experts.
According to the Court, we can infer from this finding that some well-
qualified expert, local or otherwise, would have been available. Without
at least a finding of scientific reliability from the trial court regarding the
expert testimony, I am unwilling to make that leap. Before granting
Applicant a new trial, I would remand this case a second time so that the
trial court can clarify whether it believes the expert testimony at issue
would have been reliable.
Currently, the witness-availability rule appears to be an article of
faith in our ineffective assistance jurisprudence.1 Strickland v.
Washington makes no mention of a requirement that a defendant must
show that a witness was available to testify if he is to prevail upon a
claim of ineffective assistance based upon the failure to call a witness.2
The earliest example I found in our jurisprudence setting out this
requirement appears in Hunnicutt v. State. There, we simply dismissed
the defendant’s ineffective assistance claim by stating, “Likewise,
1
See, e.g., King v. State, 649 S.W .2d 42, 44 (Tex. Crim . App. 1983) (“Counsel’s
failure to call witnesses at the guilt-innocence and punishm ent stages is irrelevant absent a
showing that such witnesses were available and appellant would benefit from their
testim ony.”).
2
466 U.S. 668 (1984).
Armstrong Dissent – 3
appellant’s complaint that his counsel failed to call witnesses at the guilt
stage of the trial or character witnesses later is unwarranted in absence
of showing that such witnesses were available and appellant would
benefit from the presentation of such evidence.” 3 For this holding we
relied upon Coble v. State, an equally circumspect opinion on the issue.
But notably, the holding in Coble made no mention of the witness-
availability requirement; we simply stated, “Complaint of counsel’s action
in not presenting evidence in behalf of appellant is unwarranted in the
absence of a showing that appellant would have benefitted from the
presentation of such evidence.”4 Somewhere between Coble and
Hunnicutt, we seem to have added the witness-availability requirement
without explaining why. So while I am willing to continue to apply this
rule, I am reluctant to adhere to it too rigidly.
If I were to hazard a guess as to why such a showing should be
necessary, it would be that the prejudice prong of the Strickland test
requires a showing that there is a reasonable probability that, but for
counsel’s unprofessional errors, the result of the proceeding would be
3
Hunnicutt v. State, 531 S.W .2d 618, 625 (Tex. Crim . App. 1976). Hunnicutt dealt
with claim of ineffective assistance lodged against a retained attorney under a standard that
we ultim ately rejected. See Hurley v. State, 606 S.W .2d 887, 890 (Tex. Crim . App. 1980).
4
Coble v. State, 501 S.W .2d 344, 346 (Tex. Crim . App. 1973).
Armstrong Dissent – 4
different.5 I can see how the witness-availability rule might make sense
under this rationale with regard to fact witnesses. A witness who will
testify about discrete facts has unique knowledge about the case that
cannot be exactly replicated by a different witness. In such cases, the
availability of a unique witness is crucial to proving whether the result of
a proceeding would be different.
But expert witnesses are not like fact witnesses because they are,
at least arguably, interchangeable. For expert testimony to be
admissible, it must be based upon a reliable scientific foundation, and
relevant to the issues in the case.6 When the expert testimony involves
a “soft science,” such as in this case, there must be a showing that the
field of expertise is a legitimate one, the subject matter of the expert
testimony is within the scope of that field, and the expert’s testimony
properly relies upon or utilizes the principles involved in that field.7 The
overall focus is to determine whether the evidence has its basis in sound
scientific methodology such that testimony about “junk science” is
5
Perez v. State, 310 S.W .3d 890, 893 (Tex. Crim . App. 2010) (citing Strickland, 466
U.S. at 694).
6
Tillm an v. State, 354 S.W.3d 425, 435 (Tex. Crim . App. 2011).
7
Nenno v. Sate, 970 S.W .2d 549, 561 (Tex. Crim . App. 1998).
Armstrong Dissent – 5
weeded out.8 In contrast to the unique fact witness, the more unique the
expert witness, the more concern there is about the reliability of his or
her testimony. It is the substance of the testimony we focus upon, not
the identity of the expert giving that testimony.
This is why I regard the trial court’s findings on the availability of
the expert witnesses in this case insufficient. On the one hand, the trial
court determined the experts to be credible. And the experts testified
that they were available to testify. But the trial court also stated that
“there is no reason why Applicant’s trial attorneys would have found the
out-of-state expert witnesses whom habeas counsel later retained and it
is unlikely that this court would have granted requests for funds to pay
said individuals, as opposed to qualified local experts.”9 At first blush, the
8
Jordan v. State, 928 S.W.2d 550, 555 (Tex. Crim . App. 1996).
9
The prejudice prong of Strickland presum es that a judge acts according to law.
Strickland, 466 U.S. at 694-95 (“A defendant has no entitlem ent to the luck of a lawless
decisionm aker, even if a lawless decision cannot be reviewed. The assessm ent of prejudice
should proceed on the assum ption that the decisionm aker is reasonably, conscientiously,
and im partially applying the standards that govern the decision.”). In this regard, the trial
court’s determ ination that Applicant failed to establish prejudice sim ply because the trial
court would not have authorized funding for the experts seem s to apply too strict a
standard. The appropriate question would be whether the trial court would have been
required to appoint the specific experts as a m atter of due process. Ake v. Oklahom a, 470
U.S. 68, 77 (1985) (“W e recognized long ago that m ere access to the courthouse doors does
not by itself assure a proper functioning of the adversary process, and that a crim inal trial is
fundam entally unfair if the State proceeds against an indigent defendant without m aking
certain that he has access to the raw m aterials integral to the building of an effective
defense.”). W hether Applicant was entitled to the appointm ent of these experts or sim ilar
experts in the first place is an additional reason for rem anding the case.
Armstrong Dissent – 6
trial court appears to be determining that the out-of-state experts’
testimony was reliable because it was interchangeable with that of
qualified local experts. But in the context of a recommendation to deny
relief, it is also possible that the trial court wanted proof that qualified
local experts would have reached the same conclusion because it believed
the out-of-state experts’ testimony was not sufficiently reliable.10
Consequently, I would remand the case again for the trial court to
make findings regarding whether the expert testimony in this case was
reliable under Nenno.11 If the trial court determines that the testimony
is reliable, then it is fair to make the inference that a local expert would
have been available to provide the same or similar testimony. Because
the Court does not, I respectfully dissent.
Filed: November 15, 2017
Do Not Publish
10
Notably, in one of the findings, the trial court refers to the out-of-state experts as
“hired guns” suggesting that he did not regard their conclusions as sufficiently reliable.
11
The trial court sets out in great detail in its findings the substance of the expert
testim ony presented. It prefaces each finding with a statem ent that this testim ony was
“credible.” I acknowledge we could sim ply perform the reliability analysis ourselves.
However, we only asked the trial court to consider the issues of credibility and availability,
not reliability, which I believe is the key inquiry regarding the failure to investigate and
present expert testim ony. So in this instance, I would let the trial court have an opportunity
to conduct that inquiry first rather than reject the trial court’s recom m endation based upon
a determ ination that the trial court never had a chance to m ake.