UNITED STATES COURT OF APPEALS FOR VETERANS CLAIMS


NO . 03-1014

JEFFERY A. WELLS,                                              APPELLANT ,

       V.


ANTHONY J. PRINCIPI,
SECRETARY OF VETERANS AFFAIRS,                                 APPELLEE.


               Before KRAMER, Chief Judge, and FARLEY, IVERS, STEINBERG,
                         GREENE, KASOLD, and HAGEL, Judges.

                                             ORDER

        On November 3, 2003, the appellant, through counsel, filed a motion for an initial decision
by the full Court. He argues that full-Court consideration is necessary because the appeal presents
the opportunity to consider all aspects of the amendments to the Veterans Benefits Act of 2002, Pub.
L. No. 107-330, 116 Stat. 2820 (2002).

       Motions for a full-Court decision are not favored. Ordinarily they will not be granted unless
such action is necessary to secure or maintain uniformity of the Court's decisions or to resolve a
question of exceptional importance. See U.S. VET . APP . R. 35(c). In this matter, the appellant has
not shown that either basis exists to warrant a full-Court decision.

       Upon consideration of the foregoing, it is

       ORDERED that the appellant's motion for an initial decision by the full Court is denied.

DATED: May 11, 2004                                    PER CURIAM.

        STEINBERG, Judge, dissenting: I voted for the appellant's motion for initial en banc
consideration of this case in order for the full Court to address issues of exceptional importance
raised by the appellant (in the brief he submitted in October 2003) regarding the effect of section 401
of the Veterans Benefits Act of 2002 (VBA), Pub. L. No. 107-330, § 401, 116 Stat. 2820, 2832, and
especially for the full Court to examine collectively the Court's decisions in Gilbert v. Derwinski,

(1990), Roberson v. Principi, 

(2003) (Roberson III), and Mariano
v. Principi, 

(2003), in light of the enactment of VBA § 401. See U.S. VET . APP .
R. 35(c) (providing that full-Court consideration "[o]rdinarily . . . will not be granted unless such
action is necessary to secure or maintain uniformity of the Court's decisions or to resolve a question
of exceptional importance").
                                      I. Background: Roberson III

        The VBA amendments were enacted in December 2002. In June 2003, the Court issued a
per curiam order in Roberson III, which involved an inferred claim, for a Department of Veterans
Affairs (VA) rating of total disability based on individual unemployability (TDIU), that the Court
concluded "ha[d] never been addressed by [a VA regional office (RO)] or the Board of Veterans'
Appeals (Board or BVA)]", and that the Court thus remanded for initial consideration by the Board.
Roberson 

. In Roberson III, the Court made only two holdings as to matters
before it on appeal: (1) That the Court was bound by the opinion of the U.S. Court of Appeals for
the Federal Circuit (Federal Circuit) in Roberson v. Principi, 

(Fed. Cir. 2001), which
reversed this Court's decision that that appellant's prior claim for service connection had not raised
a claim for a TDIU rating and (2) that, because there was no VARO or BVA decision to review on
the claim for a TDIU rating, "we must decline the appellant's invitation to run roughshod over the
VA's adjudication process" by "adjudicat[ing] the matter in the first instance." Roberson 

.
Based on these holdings, the Court remanded "the matter of the appellant's eligibility for [a] TDIU
[rating] . . . to the Board", ibid., after first concluding that the Court had jurisdiction over the
unadjudicated TDIU-rating claim because it was "expressly presented to the RO and the Board", 

the Court in Roberson III had limited jurisdiction over the TDIU-rating claim
(insofar as it could have remanded it for the Board to refer it to the RO1), the Court lacked
jurisdiction to reach the merits of that claim because the Board had never addressed it and it was
never placed in appellate status before the Board by a Notice of Disagreement (NOD) that disagreed
either with the RO's adjudication of that claim or the RO's failure to adjudicate it. See 38 U.S.C.
§ 7105(a) (providing that BVA review initiated by an NOD); Buckley v. West, 

, 82
(1998) (concluding that Board’s jurisdiction "derives from a claimant’s NOD"); Velez v. West,

, 157 (1998) (recognizing that NOD is valid if it expresses "disagreement with an
RO's . . . failure to adjudicate [a] claim" raised to RO); Hazan v. Gober, 

, 516 (1997)
(same); Isenbart v. Brown, 

, 541 (1995) (same). This Court's reliance in Roberson
III on Fenderson v. West, 

(1999), and Holland v. Gober, 

(1997)
(per curiam order), for the Court's assertion of jurisdiction was misplaced because in both of those
cases there were NODs as to the claims unadjudicated by the Board. Roberson 

. Unlike in Fenderson and Holland, the Court in Roberson III, in order to correct the Board's
error in failing to consider the TDIU-rating claim, should simply have remanded the claim to the
BVA for it to refer the matter to the RO, because on remand the Board would not have had any
jurisdiction to adjudicate that TDIU-rating claim; the Board's jurisdiction would allow it merely to
refer that claim – not then in appellate status before the Board – to the RO.2

        1
           See Bruce v. West, 

, 408 (1998); Godfrey v. Brown, 

, 408-09 (1995); cf.
Manlincon v. West, 

, 240-41 (1999) (remanding, where claim was in appellate status before Board of
Veterans' Appeals (BVA or Board), claim unadjudicated by BVA and Department of Veterans Affairs (VA) regional
office (RO) for BVA to remand to VARO).

        2
            See supra note 1.

                                                       2
         Thus, the Court in Roberson III, having determined that it could not exceed "the authority
and primary responsibility of this Court . . . to review Board decisions", Roberson 

, had no basis for addressing the appellant's contention that the Court should adjudicate the
inferred TDIU-rating claim in the first instance – that argument was rendered moot by the former
determination.3 The Roberson III Court's extensive examination of the meaning and interpretation
of the effect of VBA § 401, primarily in terms of its effect on how this Court reviews BVA decisions
on the merits, Roberson 

-48, must therefore be considered dicta.4 However,
in Mariano this Court subsequently held, as Roberson III had concluded, 

under VBA
§ 401 our standard of review regarding BVA factfinding continued to be the "clearly erroneous"
standard, 38 U.S.C. § 7261(a)(4), and that "[t]he outcome of the Board's application of the section
5107(b) equipoise standard is a factual determination that this Court reviews under [that] 'clearly
erroneous' standard." 

; see also Mitchell v. Principi, __ Vet.App. __,
__, No. 01-1659, 

, at *1-2 (Apr. 7, 2004) (per curiam order) (Kramer, C.J.,
concurring) (reiterating holding from 

). That is where I believe the Court's caselaw
is to date in terms of the effect of VBA § 401. What has not yet been made clear, however, is exactly
how (by what criteria) the Court should review the Secretary's application of the section 5107(b)
equipoise standard. Because I believe that VBA § 401 was enacted to and did effect a substantial
change from prior law, I voted to grant full-Court consideration to this case in order to address this
matter of exceptional importance as to which our caselaw is much in need of clarification. See VET .
APP . R. 35(c). My conclusions on this matter follow.



         3
            The Court's decision in Roberson v. Principi, 

(2003) (Roberson III) to address the
amendments made by section 401 of the Veterans Benefits Act of 2002 (VBA), Pub. L. No. 107-330, § 401, 116 Stat.
2820, 2832, was also problematic because the Court had received virtually no briefing from the Secretary on the
questions that it addressed regarding the VBA amendments, largely because the Secretary had (in my view, correctly)
concluded that the Court could not properly reach those issues and because he (also, in my view, correctly) had asserted
that the amendments could not affect the standard of review in the Court over Board decisions denying claims of clear
and unmistakable error. Roberson III Secretary's Brief at 13-15.

         4
           It appears that the panel had reservations about the correctness of the policy judgment that would underlie
any effort by Congress to change this Court's standard of judicial review: Specifically, at the end of its analysis,
immediately before rejecting "the appellant's argument that the VBA somehow altered the landscape of judicial review",
the Roberson III panel stated:

                  If [Congress] decides that the record produced by a non-adversarial claims
                  adjudication process is, in its judgment, inadequate, then it can sharpen that process
                  by making it adversarial through elimination of the attorney fee prohibition during
                  claim development and adjudication. Merely changing this Court's standard of
                  review while doing nothing to enhance the record would compound rather than
                  correct any problems.

Roberson 

. Moreover, the Roberson III Court refers to Congress' having "amended section
7261(b) [by] adding an entreaty to the Court to 'take due account of the Secretary's application of [38 U.S.C.
§ ]5107(b).'" Roberson III, 17 V et.App. at 140. An entreaty, however, is "an earnest request; supplication; prayer",
W EBSTER 'S N EW W O RLD D ICTION ARY 454 (3d ed. 1998), whereas in the amended section 7261(b)(1) Congress m andated
("shall") that the Court carry out this new review. 38 U.S.C. § 7261(b)(1).

                                                            3
                                      II. VBA Amendments
                                         A. Text of Statute

         Section 401 of the VBA, effective December 6, 2002, amended 38 U.S.C. § 7261(a)(4) and
(b)(1). VBA § 401. Prior to the VBA, our caselaw provided (1) that the Court was authorized to
reverse a Board finding of fact only where "the only permissible view of the evidence [of record] is
contrary to that found by the BVA", Johnson (Brenda) v. Brown, 

, 10 (1996), i.e.,
supports the appellant's position, and (2) that a Board finding of fact must be affirmed where "there
is a 'plausible basis' in the record" for the Board's determination, Hurd v. West, 

, 451
(2000). See also Harder v. Brown, 

, 189 (1993) (citing Karnas v. Derwinski,

, 311 (1991)); Barnhill v. Brown, 

, 77 (1993).

         As a result of VBA § 401's amendments to section 7261(a)(4), the Court is now directed to
"hold unlawful and set aside or reverse" any "finding of material fact adverse to the claimant . . .
if the finding is clearly erroneous." 38 U.S.C. § 7261(a)(4), as amended by VBA § 401(a) (emphasis
indicates VBA amendments). Furthermore, new section 7261(b)(1) now provides in entirely new
language added to that subsection:

                  (b) In making the determinations under subsection (a), the Court
               shall review the record of proceedings before the Secretary and the
               [BVA] pursuant to [38 U.S.C. § ]7252(b) of this title[ 38] and shall –

                       (1) take due account of the Secretary's application of
               section 5107(b) of this title . . . .

38 U.S.C. § 7261(b)(1). The Secretary's obligation under section 5107(b), as referred to in
section 7261(b)(1), is as follows:

                  (b) BENEFIT OF THE DOUBT . – The Secretary shall consider all
               information and lay and medical evidence of record in a case before
               the Secretary with respect to benefits under laws administered by the
               Secretary. When there is an approximate balance of positive and
               negative evidence regarding any issue material to the determination
               of a matter, the Secretary shall give the benefit of the doubt to the
               claimant.

38 U.S.C. § 5107(b). In Gilbert, the Court characterized the benefit-of-the-doubt rule (then found
in 38 U.S.C. § 3007(b)) as mandating that "when . . . the evidence is in relative equipoise, the law
dictates that [the] veteran prevails" and that, conversely, a VA claimant loses only when "a fair
preponderance of the evidence is against the claim." 

; see also Shoffner
v. Principi, 

, 215 (2002) (citing Dela Cruz v. Principi, 

, 149 (2001),
for proposition that "[Veterans Claims Assistance Act of 2000, Pub. L. No. 106-475, 114 Stat. 2096]
did not change [the section 5107(b)] benefit-of-the-doubt doctrine").


                                                  4
         Reading amended sections 7261(a)(4) and 7261(b)(1) together, which must be done in order
to determine the effect of the VBA § 401 amendments, this Court is now directed, for the first time
as part of its scope-of-review responsibility under section 7261(a)(4), to undertake three new actions
in deciding whether BVA factfinding adverse to a claimant is clearly erroneous and, if so, what the
Court should hold as to that factfinding. See Reno v. Koray, 

, 56-57 (1995) (construing
statutory language in context of statutory scheme); Kilpatrick v. Principi, 

, 7 (2002)
(examining entire statutory scheme in interpreting meaning of statute), aff'd, 

(Fed.Cir.
2003). Specifically, the plain meaning of the amended subsections (a)(4) and (b)(1) requires the
Court (1) to review all evidence "before the Secretary and the Board", 38 U.S.C. § 7261(b); (2) to
consider "the Secretary's application of" the benefit-of-the-doubt rule in view of that evidence,
38 U.S.C. § 7261(b)(1); and (3) if the Court, after carrying out actions (1) and (2), concludes that an
adverse BVA finding of fact is clearly erroneous and therefore unlawful, to decide whether to set it
aside or reverse it, 38 U.S.C. § 7261(a)(4).

                                       B. Historical Context

        The significance and effect of the three new actions that the Court has been directed or
authorized (in the case of reversal) to undertake is best understood in the historical context of the
Court's exercise of its scope of review. Prior to the enactment of VBA § 401, the Court was
permitted by section 7261(a)(4) only to "set aside" a "clearly erroneous" BVA finding of fact. See
Veterans' Judicial Review Act (VJRA), Pub. L. No. 100-687, § 301(a), 102 Stat. 4105, 4115 (1988)
(enacting 38 U.S.C. § 4061(a)(4), renumbered as section 7261(a)(4)). Almost from this Court's
inception, its review of findings of fact under the "clearly erroneous" standard had been construed
as "parallel . . . [to] the standard used by [U.S.] Courts of Appeals in reviewing findings of material
fact made by [U.S.] District Courts," 

; see FED . R. CIV . P. 52(a) (providing,
at time of VJRA's enactment and presently, that courts of appeals review district court findings of
fact under "clearly erroneous" standard). The Court characterized this standard as meaning that "if
there [wa]s a 'plausible' basis in the record for the factual determinations of the BVA, even if this
Court might not have reached the same factual determinations, we c[ould] not overturn them",

-53 (citing United States v. United States Gypsum Co., 

, 395
(1948), Anderson v. City of Bessemer, 

, 573-74 (1985), and Danville Plywood Corp.
v. United States, 

(Fed. Cir. 1990)). As discussed in part II.E, below, however, this
formulation of the "clearly erroneous" standard seems at its inception to have strayed from the
formula applied in traditional Article III appellate review as articulated by the Supreme Court in the
three cases on which Gilbert purported to rely. In any event, more recently but prior to the VBA,
the Federal Circuit in Hensley v. West expounded as follows on this Court's standard of review as
enacted in the VJRA:

               On factual matters, the findings of the BVA may be overturned by the
               Court of Appeals for Veterans Claims only if they are clearly
               erroneous. See 38 U.S.C. § 7261(a)(4). . . . Furthermore, "in no
               event shall findings of fact made by the Secretary or [BVA] be
               subject to trial de novo by the Court [of Appeals for Veterans


                                                   5
                   Claims]." 38 U.S.C. § 7261(c). The statutory provisions are
                   consistent with the general rule that appellate tribunals are not
                   appropriate fora for initial fact finding.

Hensley, 

, 1263 (Fed. Cir. 2000).

        Under new section 7261(a)(4), in addition to the longstanding pre-VBA direction to the Court
to "set aside" a "clearly erroneous" finding, the Court is expressly permitted to "reverse" such an
adverse factfinding. As the ensuing discussion illustrates, the VBA § 401 amendments to
section 7261(a)(4) and (b) have fundamentally altered this Court's BVA-factfinding scope of review,
as evidenced by both the plain meaning of the amended language of these subsections as well as the
unequivocal legislative history of the amendments.

                                C. Examination of New Statutory Language

        In examining the text of the § 401 amendments, I turn first to the addition in
section 7261(a)(4) of "or reverse" as to BVA factfinding that is "adverse to the claimant", one of the
three new actions directed or authorized to be undertaken by the Court pursuant to amended
section 7261. As a general rule of statutory construction, meaning must be attributed to each
amendment made by Congress: "The [Supreme] Court has often said that 'every clause and word
of a statute' should, 'if possible,' be given 'effect.'"5 The plain meaning of the language used,
"reverse", signifies a major shift for the Court's review of BVA factfinding. Although
section 7252(a) has included, since the enactment of the VJRA in 1988, general authority for the
Court to "reverse" a Board decision, VJRA § 301(a), 102 Stat. at 4113 (enacting 38 U.S.C.
§ 4052(a), now section 7252(a) ("the Court shall have power to affirm, modify, or reverse a decision
of the Board")), it appears (in light of 

) that that authority could have pertained to
reversal solely where factfinding was not required, i.e., where reversal was required as a matter of
law.6 Whereas "setting aside" a finding of fact is discarding or invalidating it and allowing the
agency or court from which the appeal is heard to readjudicate the issue, "reversing" a finding of fact
is to change it.7 Furthermore, consistent with the proclaimant nature of the VA adjudication system,


         5
            Chickasaw Nation v. United States, 

, 93 (2001) (quoting United States v. Menasche, 

, 538-539 (1955), and Montclair v. Ramsdell, 

, 152 (1883) ("[i]t is the duty of the court to give effect,
if possible, to every clause and word of a statute, avoiding, if it may be, any construction which implies that the
legislature was ignorant of the meaning of the language it employed")); see Duncan v. Walker, 

, 174 (2001)
(quoting Market Co. v. Hoffman, 

, 115 (1879), for

proposition that "[a]s early as in Bacon's Abridgment, sect. 2, it was said that 'a statute ought, upon the whole, to be so
construed that, if it can be prevented, no clause, sentence, or word shall be superfluous, void, or insignificant'").

         6
             See Taylor and Andrulonis, infra note 7.

         7
           This distinction is well presented in Icicle Seafoods, Inc. v. Worthington (on which the U.S. Court of Appeals
for the Federal Circuit (Federal Circuit) had relied in Hensley v. West, 

, 1263 (Fed. Cir. 2000)), where the
Supreme Court described the general, traditional role of Article III appellate-court review (which, of course, is not

                                                            6
see Hodge v. West, 

, 1362-63 (Fed. Cir. 1998), and the availability of appeal to this
Court only by the appellant, see 38 U.S.C. § 7252(a), Congress in VBA § 401 provided the authority
to "reverse" (or "set aside") only those findings that are "adverse to the claimant".8 Moreover, as set
forth below, the legislative history bolsters the plain meaning of the statute by making clear that
Congress intended for the Court to take a more proactive and less deferential role in its BVA-
factfinding review. See infra notes 15-18 and accompanying text.

        As to new section 7261(b)'s requirement that this Court "review the record of proceedings",
that provision further supports the conclusion that VBA § 401 brought about a major expansion of
this Court's responsibilities as to review of BVA factfinding. 38 U.S.C. § 7261(b) ("In making the
determinations under subsection (a) of this section, the Court shall review the record of proceedings
before the Secretary and the [BVA] pursuant to section 7252(b)."). The "record of proceedings"
referred to in new section 7261(b)(1) is the record on appeal (ROA) designated by the parties and
transmitted to the Court pursuant to Rules 10 and 11 of this Court's Rules of Practice and Procedure.
See Homan v. Principi, 

, 3-4 (2003) (per curiam order). Although section 7252(b) has
always provided that "[r]eview in the Court shall be on the record of proceedings before the
Secretary and the Board", the Federal Circuit in Hensley implicitly held that our review under section
7261(b) was limited to the evidence used by the Board. 

; see also Winters
v. Gober, 

, 1380 (Fed. Cir. 2000) (holding that Court had exceeded its statutory
authority and prejudiced appellant by deciding case on grounds "not relied on" by Board that may
have required "improper de novo findings of fact"). The addition of that requirement to section
7261(b), in light of Hensley, signals that the Court's review should be based on all the evidence
before the Board, rather than just the evidence used by the Board and suggests a congressional




carried out pursuant to the same standard of review made applicable to this Court by the VBA § 401 amendments to
section 7261(a)(4) and (b)(1)):

                  If the Court of Appeals . . . was of the view that the findings of the District Court
                  were "clearly erroneous" . . . , it could have set them aside on that basis. If it
                  believed that the District Court's factual findings were unassailable, but that the
                  proper rule of law was misapplied to those findings, it could have reversed the
                  District Court's judgment. But it should not simply have made factual findings on
                  its own.

Icicle Seafoods, Inc., 

, 714 (1986) (emphasis added). Compare, e.g., Fed. Election Comm'n v. Akins,

, 25 (1998) ("If a reviewing court agrees that the agency misinterpreted the law, it will set aside the agency's
action and remand the case – even though the agency (like a new jury after a mistrial) might later, in the exercise of its
lawful discretion, reach the same result for a different reason." (emphasis added)), with Taylor v. Chater, 

,
1278 (8th Cir. 1997) (reversing district court decision and ordering district court to grant summary judgment to party),
and Andrulonis v. United States, 

, 1236 (2d Cir. 1994) (reversing part of district court decision that
awarded payment of interest through March 11, 1991, and "direct[ing] the district court to award interest through
February 13, 1992" (emphasis added)).

         8
          Cf. Morgan v. Principi, 

, 232 (2002) ("Neither this Court nor the . . . Federal Circuit has held
whether this Court can disturb a favorable Board determination.").

                                                            7
rejection of both Hensley and Winters. See infra note 17 and accompanying text (discussing
legislative history suggesting Hensley was overruled by VBA § 401).9

        Finally, the effect of the addition to the Court's scope-of-review authority in section 7261 of
a direction ("shall review") to this Court to "review the record of proceedings before the Secretary
and the Board . . . pursuant to section 7252(b)" can be fully understood only in the context of the
specific mandate in section 7252(b) that "[t]he extent of [this Court's] review shall be limited to the
scope provided in section 7261." 38 U.S.C. § 7252(b); see Mayer v. Brown, 37 F.3d, 618, 620 (Fed.
Cir. 1994) (stressing separate purposes of section 7252(b) and section 7261), overruled on other
grounds by Bailey (Harold) v. West, 

(Fed. Cir. 1998). That is, section 7252(b) says
in effect that unless a matter is set forth in section 7261, that matter is not part of the Court's scope
of review; the VBA § 401(b) amendment to section 7261(b) now meets that criterion.10


         9
            I do not understand how the Court could reverse (as distinguished from setting aside) a finding of material
fact without in some fashion finding facts itself based on evidence of record and, hence, I conclude that, by virtue of this
new statutory authority, 

, no longer applies to prevent this Court from reversing under section 7261(a)(4)
and, in some manner, weighing evidence in order to take due account of the Secretary's application of the equipoise rule
as well as, under section 7261(b)(2), to take due account of the rule of prejudicial error. In this regard, the list of
precedential cases, cited by the Court in Roberson 

, in which the Court had previously reversed
BVA factfinding divides into two categories: (1) Those cases preceding Hensley that were not bound by its no-
factfinding rule and (2) the two cases issued post-Hensley, namely Pentecost v. Principi and Harth v. West. Harth's
reversal was based on a de novo application of the law to the facts in order to find well grounded a service-connection
claim for post-traumatic stress disorder (PTSD), Harth, 

, 5-6 (2000), and the opinion does not mention
section 7261(a)(4). Pentecost's reversal (although section 7261(a)(4) is referred to) is also as to a PTSD claim and seems
to be based on the Board's too-strict reading of both the stressor-corroboration requirement in 38 C.F.R. § 3.304(f)
(1999) and the "controlling precedent" of Suozzi v. Brown, 

(1997). Pentecost, 

24, 128-29
(2002) (quoting Suozzi,10 Vet.App. at 311, in which the Court had "rejected" a narrow definition of "corroboration" that
would require "corroboration of every detail[,] including the appellant's personal participation").

         10
              About this section 7261(b)(1) insertion, the Roberson III order stated:

                    The Secretary's application, or lack thereof, of the benefit[-]of[-]the[-]doubt rule is
                    part of any decision of the B oard, and was therefore already within the Court's
                    review power under [38 U.S.C. §] 7252(a). The command that the Board "shall
                    review the record of proceedings before the Secretary and the Board . . . pursuant
                    to section 7252(b)" wraps the statutory text around on itself. Section 7252(b) states,
                    in pertinent part, that "[r]eview in the Court shall be on the record of proceedings
                    before the Secretary and the Board" and that "[t]he extent of the review shall be
                    limited to the scope provided in section 7261." In other words, section 7261 states
                    that the Court shall review the record of proceedings which in turn states that such
                    review is limited by section 7261. Review is thus trapped between these two
                    mutually referential provisions, which is to say that there is no clear reading of this
                    provision.

Roberson 

. I must confess not to find any such "wrap[ping] . . . around" or "trapp[ing]" of review.
Although section 7252(b)'s pre-VBA direction that the Court review the "record of proceedings" included, in a general
way, the Board's consideration of the equipoise standard, the direct command in new section 7261(b) that the Court, in
completing its review, "take due account of the Secretary's application of section 5107(b)" has made a meaningful
change to our scope of review. 38 U.S.C. § 7261(b)(1) (emphasis added). The first matter under section 7252(a) is a

                                                              8
        Perhaps the most dramatic of the three Court actions directed or authorized by VBA § 401
is the mandate that the Court "take due account of the Secretary's application of section 5107(b)",
which, for many years preceding the enactment of the VJRA, had been known as the "benefit-of-the-
doubt rule". 38 U.S.C. § 7261(b)(1). The "application of section 5107(b)" in VBA § 401, refers to
the question whether "there is an approximate balance of positive and negative evidence" – a
criterion that has been known since Gilbert as the "equipoise rule" but is more aptly called the
"equipoise standard", see 

. Prior to the enactment of the VBA and at
the time of 

, the scope-of-review provisions enacted in section 4061 (now
section 7261) by the VJRA in 1988 made no reference whatsoever to section 3007(b) (the
predecessor of current section 5107(b)) but, rather, referred to review of BVA factfinding only in
terms of the "clearly erroneous" standard. In light of the lack of specificity in those VJRA-enacted
provisions, the Court in Gilbert had a sound basis for applying a deferential formulation of the
"clearly erroneous" standard and for concluding, in essence, that the Court's review of the Board's
compliance with then-section 3007(b) was limited to whether the Board had provided an adequate
statement of reasons or bases (under 38 U.S.C. § 4004(d)(1) (now 38 U.S.C. § 7104(d)(1))) for its
conclusion that the evidence preponderated against the claim or against the claimant's position on
a finding of material fact. 

("[w]here findings of material fact by the Board
are properly supported and reasoned, and the Board concludes that a fair preponderance of the
evidence weighs against the claims of a veteran, it would not be error for the Board to deny the
veteran the benefit of the doubt"). Although the Court later intensified its reasons-or-bases scrutiny
somewhat in Williams (Willie) v. Brown, that scrutiny did not include an evaluation of how the Board
actually applied the equipoise standard but only how it said it had applied it. Williams (Willie),

, 273-74 (1993) ("[i]n a case where there is significant evidence in support of an
appellant's claim, as there is here, the Board must provide a satisfactory explanation as to why the
evidence was not in equipoise").11 It is against this rather relaxed standard of review that, through
VBA § 401, Congress has now required the Court to review the entire ROA and to examine the
Secretary's determination as to whether the evidence presented was in equipoise on a particular
material fact.12


matter of jurisdiction and the latter matter under section 7261(b) defines how the Court will conduct review within its
jurisdiction. Furthermore, that new section 7261(b) directive must be read in the context of the full section 7261(a),
which commands such "tak[ing] account" in the context of evaluating BVA findings – including whether adverse
findings of fact are clearly erroneous. As explained in the text immediately following this footnote, I believe that section
7261(b) requires the Court to review the Secretary's application under that same "clearly erroneous" standard; that
requirement surpasses that imposed by the broad language of section 7252(a), which does not mention section 5107(b)
by name and which speaks only to what matters are included in the Court's jurisdiction to review.

         11
            See also Jackson v. West, 

, 425, 429 (1999) (applying Williams (Willie) v. Brown,

, 273-74 (1993)); Gaines v. West, 

, 359 (1998) (same).

         12
             See infra note 17 and accompanying text (legislative history clarifying that VBA § 401 amendments with
respect to the benefit-of-the-doubt rule require Court to provide less deferential and more searching review of Board
decisions and stating that that rule will, under VBA § 401, be given "full force"); cf. Gregory C. Sisk, The Trial [sic]
Courts of the Federal Circuit; Diversity by Design, 13 Fed. Cir. B.J. 241, 263 (2004) ("the actual textual changes to the
statute enacted . . . [in the VBA] appear to mandate moderate but nonetheless meaningful adjustment by [this Court] on
a case-by-case basis, rather than revolutionary change[;] . . . a fundamental principle of statutory interpretation is that

                                                             9
       In light of this background, it appears that this new mandate supercedes the conclusion in
Gilbert that we were to review the Board's application of the equipoise standard to determine
whether that application was "arbitrary, capricious, an abuse of discretion, or otherwise not in
accordance with law", pursuant to section 7261(a)(3)(A) (then section 4061(a)(3)(A) (1988)).

. Specifically, by inserting into section 7261(b)(1) the direction that the
Court evaluate the Board's application of the equipoise standard to findings of material fact,
Congress seems to have been suggesting that the section 7261(a)(4) "clearly erroneous" standard of
review should be applied as to that evaluation.13 As indicated in part I, above, the Court so
concluded in Mariano:

                   The outcome of the Board's application of the section 5107(b)
                   equipoise standard is a factual determination that this Court reviews
                   under the "clearly erroneous" standard. See 

("[The Court] is not authorized to make the determination as to
                   whether the evidence is in equipoise and apply the
                   benefit[-]of[-]the[-]doubt doctrine; the Court is empowered only to
                   ensure that the Secretary's determination in that regard is not clearly
                   erroneous."). Indeed, Congress, by adding section 7261(b)(1),
                   directed that the Court, "[i]n making the determinations under
                   subsection (a)," including a determination as to whether a finding of
                   material fact adverse to the claimant is clearly erroneous, shall take
                   due account of the Secretary's application of section 5107(b)'s
                   equipoise standard. 38 U.S.C. § 7261(b)(1); see 38 U.S.C.
                   § 7261(a)(4).

.14


a change in statutory language presumptively effects a change in meaning").

         13
              The Roberson III assertion that "[t]he Secretary's application, or lack thereof, of the benefit[-]of[-]the[-
]doubt rule . . . was . . . already within the Court's review power under section 7252(a)" does not take into consideration
the difference between that which is permissible and that which is now mandated ("shall"), according to the two Veterans'
Affairs Committees, to provide for "more searching appellate review of BVA decisions" in order to give "full force to
the 'benefit of the doubt' provisions." 148 C ONG . R EC . S11337, H9003 (daily ed. Nov. 18, 2002). If the Court had
always given such "due regard", it would seem that the Roberson III order would cite some precedent on that point, but
it does not (perhaps because that "regard" has been, since Gilbert v. Derwinski, 

(1990), only a reasons-or-
bases check, 

text accompanying note 11).

         14
             Further, our review of Board decisions under this standard and the direction to this Court to engage in a
limited form of factfinding in reviewing the Board's factfinding should be a double-edged sword: In reviewing the record
before the Secretary and the Board, the Court is not only empowered, by section 7261(a)(4) and (b)(1), to reverse a
clearly erroneous finding of fact, but is also directed, by section 7261(b)(2), to affirm a Board decision with an error that
is rendered nonprejudicial because the evidence is so overwhelmingly against the claim in question that a remand based
on that error would be superfluous. See Soyini v. Derwinski, 

, 546 (1991); see also Valiao v. Principi,

, 232 (2003) ("W here the facts averred by a claimant cannot conceivably result in any disposition of
the appeal other than affirmance of the Board decision, the case should not be remanded for development that could not

                                                            10
                                            D. VBA Legislative History

        The legislative history supports the plain meaning of these provisions discussed in part II.C,
above, by strongly evidencing the intent of Congress to bring about decisive change in the scope of
this Court's review of BVA factfinding.15 That legislative history is set forth at this point essentially
in full. The House and Senate Committees on Veterans' Affairs described the new provisions
enacted by VBA § 401 as follows in an Explanatory Statement that they prepared regarding their
compromise agreement:

                                                       Senate bill

                           Section 501 of S. 2237 would amend section 7261(a)(4) . . .
                   to change the standard of review [the U.S. Court of Appeals for
                   Veterans Claims (CAVC)] applies to BVA findings of fact from
                   "clearly erroneous" to "unsupported by substantial evidence."
                   Section 502 would also cross-reference section 5107(b) in order to
                   emphasize that the Secretary's application of the "benefit of the
                   doubt" to an appellant's claim would be considered by CAVC on
                   appeal.

                                                       House bill

                             The House Bills contain no comparable provision.

                                               Compromise agreement

                          Section 401 of the Compromise Agreement follows the
                   Senate language with the following amendments.




possibly change the outcome of the decision.").

         15
              The legitimacy of using legislative history as a method of assisting in finding the meaning of statutory
provisions whose meaning is not free from doubt is well established. See Crosby v. Nat'l Foreign Trade Council, 

, 376 n.9 (2000) (relying on statements by sponsors of act to support conclusion as to statutory construction);

(same); 

(same); Duffield v. Robertson Stephens & Co., 

, 1192 (9th Cir.
1998) (criticizing U.S. Court of Appeals for the Fourth Circuit for "ignor[ing] the reasoning of eight Justices on the
subject of statutory analysis, [and] rely[ing instead] on a separate opinion by Justice Scalia, and partially on the basis
of that reasoning decid[ing] to disregard the legislative history"); Oregon v. Ashcroft, 

, 1089 (D.
Or. 2002) ("'The legitimacy of legislative history as a means of interpreting statutes, at least when they are unclear, is,
rightly or wrongly, well established. Other than Justice Thomas, no Justice seems interested in adopting Justice Scalia's
rejection of legislative history or his rejection of the notion of legislative intent.'" (quoting W illiam Funk, Review Essay
Faith in Texts - Justice Scalia's Interpretation of Statutes and the Constitution: Apostasy for the Rest of Us?, 49 A D M IN .
L. R EV . 825 (1997)). It is interesting, nonetheless, that the Roberson III 

, cites Justice Scalia's
concurring opinion in 

, rather than the majority opinion there, cited at the outset of this footnote.

                                                             11
                          The Compromise Agreement would modify the standard of
                  review in the Senate bill in subsection (a) by deleting the change to
                  a "substantial evidence" standard. It would modify the requirements
                  of the review the Court must perform when it is making
                  determinations under section 7261(a) . . . . Since the Secretary is
                  precluded from seeking judicial review of decisions of the [BVA], the
                  addition of the words "adverse to the claimant" in subsection (a) is
                  intended to clarify that findings of fact favorable to the claimant may
                  not be reviewed by the Court. Further, the addition of the words "or
                  reverse" after "and set aside" is intended to emphasize that the
                  Committees expect the Court to reverse clearly erroneous findings
                  when appropriate, rather than remand the case.

                          New subsection (b) [of section 7261] would maintain
                  language from the Senate bill that would require the Court to
                  examine the record of proceedings before the Secretary and BVA
                  and the special emphasis during the judicial process on the benefit
                  of the doubt provisions of section 5107(b) as it makes findings of
                  fact in reviewing BVA decisions. This would not alter the formula
                  of the standard of review on [sic] the Court, with the uncertainty of
                  interpretation of its application that would accompany such a change.
                  The combination of these changes is intended to provide for more
                  searching appellate review of BVA decisions, and thus give full
                  force to the "benefit of doubt" provision.

148 CONG . REC. S11337, H9003 (daily ed. Nov. 18, 2002) (emphasis added) (Explanatory Statement
printed in Congressional Record as part of debate in each body immediately prior to final passage
of compromise agreement).

       At the time of final action on S. 2237, the bill that was ultimately enacted as the VBA, VBA
§ 401 was quite extensively explained by Senator Rockefeller, who was the Chairman of the Senate
Committee, the floor manager of the bill in the Senate, and the principal author of VBA § 401.16 In
explaining VBA § 401, he emphasized, as did the two Committees in their Explanatory Statement,
148 CONG . REC. S11337, H9007, that "the combination of" the new requirements that the Court
"examine the . . . [ROA]", consider the benefit-of-the-doubt rule, and "make[] findings of fact in
reviewing BVA decisions" is "intended to provide for more searching appellate review of BVA


         16
            Amendments to section 7261 (dealing with all the same elements as did VBA § 401) were included in S. 2079
as introduced by Senator Rockefeller on April 9, 2002. S. 2079, 107th Cong., 2d Sess., § 2. The special responsibilities
of a Committee Chairman in explaining a bill emanating from that committee have been recognized even by the Supreme
Court's greatest skeptic as to the utility of legislative history. Conroy v. Aniskoff, 

, 521-22 (1993) (Scalia,
J., concurring in the judgment) ("This comment cannot be dismissed as the passing remark of an insignificant Member,
since the speaker was the Chairman of the House Judiciary Committee, the committee that reported the bill to the House
floor.").

                                                           12
decisions and thus give full force to the 'benefit[-]of[-]the[-]doubt' provision." 148 CONG . REC.
S11334 (emphasis added). Chairman Rockefeller concluded that the Court should "reverse clearly
erroneous findings when appropriate, rather than remand the case", and declared that the "new
language in section 7261 would overrule . . . Hensley."17 148 CONG . REC. S11334. His statement
is particularly significant (1) because only the Senate had passed provisions to amend the Court's
section 7261 scope-of-review provisions (in S. 2237),18 and the Committees on Veterans' Affairs
explained that VBA § 401 generally "follows the Senate language", and (2) because there is no
legislative history that is inconsistent with his statement. 147 CONG . REC. S11337, H9003.
Representative Evans, the ranking minority member of the House Committee, spoke "in strong
support of the bill S. 2237" and explained that "the bill . . . clarifies the authority of the Court of
Appeals for Veterans Claims to reverse decisions of the [BVA] in appropriate cases and requires
[that] the decisions be based upon the record as a whole, taking into account the pro-veteran rule
known as the 'benefit of the doubt.'" 148 CONG . REC. H9003 (emphasis added).

                                       E. Meaning of "Clearly Erroneous"

        Based on the plain language of the section 7261(a)(4) and (b)(1) amendments and their
emphatic legislative history, I conclude (1) that Congress has tasked the Court with the responsibility
of evaluating BVA factfinding adverse to a claimant by reviewing all evidence in the ROA in order


         17
              Senator Rockefeller stated in full regarding VBA § 401:

                              Section 401 of the Compromise Agreement would maintain the current
                    "clearly erroneous" standard of review, but modify the requirements of the review
                    the court must perform when making determinations under section 7261(a) of
                    title 38. CAVC would be specifically required to exam ine the record of
                    proceedings – that is, the record on appeal – before the Secretary and BVA.
                    Section 401 would also provide special emphasis during the judicial process to the
                    "benefit of the doubt" provisions of section 5107(b) as CAVC makes findings of
                    fact in reviewing BVA decisions. The com bination of these changes is intended
                    to provide for m ore searching appellate review of BVA decisions, and thus give
                    full force to the "benefit of doubt" provision. The addition of the words "or
                    reverse " after "and set aside" in section 7261(a)(4) is intended to emphasize that
                    CAVC should reverse clearly erroneous findings when appropriate, rather than
                    remand the case. This new language in section 7261 would overrule the recent
                    U.S. Court of Appeals for the Federal Circuit decision of Hensley v. West,[ 

(Fed. Cir. 2000),] which emphasized that CAVC should perform only
                    limited, deferential review of BVA decisions, and stated that BVA fact-finding "is
                    entitled on review to substantial deference." However, nothing in this new language
                    is inconsistent with the existing section 7261(c), which precludes the court from
                    conducting trial de novo when reviewing BVA decisions, that is, receiving evidence
                    that is not part of the record before BVA.

148 C O N G . R EC . S11334 (remarks of Sen. Rockefeller) (emphasis added).

         18
              S. 2237 was passed by the Senate on September 26, 2002. 148 C O N G . R EC . S9556, S9559 (daily ed. Sept. 26,
2002).

                                                             13
to assess the Board's application of the equipoise standard and (2) that if the Court, based on that
assessment, concludes that Secretary's section 5107(b) application was clearly erroneous and thus
unlawful then the Court is expressly empowered to reverse that finding – that is, not only to declare
it wrong but also to correct it.19 See 

(reversing two BVA findings
of fact as "clearly erroneous application[s] of the section 5107(b) equipoise standard"); cf. 

to another BVA finding of fact, setting aside as "clearly erroneous" but not reversing "Secretary's
application of the section 5107(b) equipoise standard"). This process of applying the "clearly
erroneous" standard to the Board's application of the section 5107(b) equipoise standard, see
Mariano and Roberson III, 

, raises the question (not previously addressed) whether the
"plausible basis in the record" and "only permissible view of the evidence" tests set forth in 

-53, remain viable, if they ever were viable, under the "clearly erroneous" standard
as proper measures for this Court's review of BVA factfinding in light of the three new section 7261
mandates.

         Although at the advent of this Court's exercise of judicial review the Court had good reason
to attempt to follow precedent regarding the scope of review for Article III appellate courts under
Rule 52(a) of the Federal Rules of Civil Procedure, the post-VBA statutory scheme in chapter 72 of
title 38, U.S. Code, is unique to this Court. Specifically, the VBA amendments, particularly that this
Court apply the "clearly erroneous" standard in our review of the Board's application of the
section 5107(b) equipoise standard, see Mariano and Mitchell, 

, render the "plausible
basis in the record" and "only permissible view of the evidence" formulas from Gilbert unsuited for
application to the Board's process of assessing the weight of positive and negative evidence. A
reexamination of Gilbert, some of our precedential decisions since Gilbert that have engrafted new
language into its "plausible basis" holding, the very Supreme Court precedents on which Gilbert had
relied, and recent Supreme Court and Federal Circuit precedent has led me to conclude that our
"clearly erroneous" review should be carried out in accordance with the primary (less deferential)
definition that Gilbert drew from the Supreme Court's opinion in United States Gypsum Co.:

                 A finding is "clearly erroneous" when although there is evidence to
                 support it, the reviewing court on the entire evidence is left with the
                 definite and firm conviction that a mistake has been committed.

(emphasis added) (quoting U.S. Gypsum 

). This criterion is the
one used by both the Supreme Court and the Federal Circuit in the Zurko litigation, which was
concluded in 2000, only two years before the VBA's enactment, and was cited in the Senate
Committee report on S. 2237. S. Rep. No. 107-234, 107th Cong., 2d Sess. 18 n.1. In that litigation,
the Supreme Court explained the "clearly erroneous" standard as meaning "whether a reviewing
judge has a 'definite and firm conviction' that an error has been committed." Dickinson v. Zurko,

, 162 (1999); see In re Zurko, 

, 1449 (Fed. Cir. 1998) (en banc) (stating



        19
          See Atl. Coast Line R.R. Co. v. St. Joe Paper Co., 

, 833 (5th Cir. 1954) (defining "reverse a
judgment" as "to overthrow it by a contrary decision, to make it void, to undo or annul it for error").

                                                        14
that "we affirm decisions as long as we lack a definite and firm conviction that a mistake has been
made").20

         Using the definite-and-firm-conviction criterion, rather than the "no plausible basis in the
record" standard, would correct the caselaw creep that began with Gilbert itself and has made it
increasingly difficult for the Court to conclude that a BVA finding of fact is "clearly erroneous"
under section 7261(a)(4). An example of the unfortunate direction that the Court's caselaw has taken
is provided by Hicks v. Brown, which, five years after its debut in Gilbert, characterized the "no
plausible basis in the record" criterion as requiring that the evidence be "uncontroverted in the
appellant's favor" and that there be "absolutely no plausible basis" in the record for the Board's
decision before a BVA finding of fact may be reversed. Hicks, 

, 422 (1995). These
quoted words and the corresponding heightened standard in Hicks were derived from Hersey v.
Derwinski, 

, 95 (1992), but Hersey does not support the use of those criteria as a basis
for denying reversal. Rather, in Hersey the Court reversed a Board decision denying a rating of total
disability based on individual unemployability (TDIU) and, in doing so, noted that "[t]here is
absolutely no plausible basis for the BVA's statement that [the veteran's] disability was 'not of
sufficient severity as to preclude hi[s] engaging in substantially gainful employment consistent with
his education and occupational experience'" and that the BVA's denial of a TDIU rating was "clearly
erroneous in light of the uncontroverted evidence in [the] appellant's favor." 

(quoting
BVA decision). However, the Court in Hersey (1) had already defined the "clearly erroneous"
standard as "the definite and firm conviction that a mistake has been committed" criterion (quoting
U.S. Gypsum 

, as it had been quoted in 

), and (2) had also already
explained that a finding of fact cannot be found clearly erroneous "[i]f the [factfinder]'s account of
the evidence is plausible in light of the record viewed in its entirety" or where such a review of the
evidence shows that "there are two permissible views of the evidence", 

(quoting 

, as quoted in 

). It is thus clear that the use in Hersey of the
"absolutely no plausible basis" and "uncontroverted" evidence terminology was intended only to
illustrate how unfounded the Board's factfinding was in that case, rather than, as Hicks used it, to
establish a new, more stringent standard for "clearly erroneous" review.

       Between the issuance of Hersey and Hicks, the Court put all the Gilbert language together
in Harder v. Brown, stating:


         20
               Utilizing this definition is consistent (1) with the characterizations of the "clearly erroneous" standard as less
deferential than the "substantial evidence . . . on the record" standard in the Administrative Procedure Act (APA),
5 U.S.C. § 706(2)(E), and (2) with the fact that the the original Senate-passed version of S. 2237 had adopted the
"substantial evidence of record" standard drawn from the APA and that that standard was rejected in the final
compromise that became VBA § 401. S. 2237, 107th Cong., 2d Sess., § 501 (amending section 7261(a)(4)); 148 C O N G .
R EC . S9558-9 (Senate passage of reported bill); see also 148 C O N G . R EC . S11337, H9006 (Committees on Veterans'
Affairs Explanatory Statement explaining that compromise agreement "would not alter the formula of the standard of
review [(that is, the "clearly erroneous" formula)] on [sic] the Court, with the uncertainty of interpretation of its
application that would accompany such a change"); S. Rep. No. 107-234, 107th Cong., 2d Sess. 18 n.1 (recognizing that
Supreme Court in Dickinson v. Zurko, 

, 162 (1999), had "interpreted" the "'substantial evidence' formula
. . . to be slightly more deferential than a traditional 'clearly erroneous' standard").

                                                              15
                       After a review of the record, we conclude that there is no
               plausible basis for the BVA's decision. "[B]ecause there is no
               evidence to support the BVA determination, it is obvious that a
               mistake has been committed, the finding is not plausible, there can be
               only one permissible view of the evidence, and, thus, the finding is
               clearly erroneous." Karnas v. Derwinski, 

, 311
               (1991); see also Caldwell [v. Derwinski], 1 Vet.App. [466,] 470
               [(1991)].

Harder, 

, 189 (1993). As in Hersey, this articulation is appropriate when there is no
evidence against the claimant's position on a particular question of material fact, because it
demonstrates the gravity of the Board's error. This articulation should not be mistaken for the
appropriate standard of review, however; for example, in Mariano,17 Vet.App. at 314-17, the Court
reversed two findings of fact even though the record contained evidence against the claimant's
position, and thus the evidence in support of the claimant's position could not be considered
"uncontroverted", 

; see U.S. 

(calling for a "clearly erroneous"
determination "although there is evidence to support" a lower court's finding); see also 

(calling for review "in light of the record . . . in its entirety").

        Moreover, a close reading of the Supreme Court's actual language in Anderson, like a close
reading of the wording in U.S. Gypsum, makes clear that the phrase "if there is a 'plausible' basis in
the record" in the Gilbert holding omitted the critical words "in light of the record viewed in its
entirety" from 

, and "although there is evidence to support [the lower
court's finding]" from U.S. 

. Thus, it appears that Gilbert had already strayed afield
from the Supreme Court's actual language and paved the way for the Hicks/Hersey detour and the
evolution of a practice in some cases of this Court's focusing only on a single or a few pieces of
negative evidence as providing a basis for affirmance regardless of the overall strength of the
evidence in favor of a claim. The Court's opinion in 

, however, showed the error of
such an approach.

        Accordingly, I believe that, as a logical evolution of Roberson III and Mariano, 

,
the Court should issue an en banc opinion holding that under the VBA § 401 amendments the
Board's application of section 5107(b) is clearly erroneous when, after reviewing the entire evidence
of record, the Court has "a definite and firm conviction that a mistake has been committed" by the
Board in making a material finding of fact adverse to the claimant – that is, when we have a definite
and firm conviction that the evidence on that question did not preponderate against the claimant's




                                                 16
position.21 In that situation, the Court should declare the BVA factfinding unlawful and decide
whether to set it aside and remand or to reverse it. 38 U.S.C. § 7261(a)(4). In my view, the Court
can make that decision only by assessing the evidence in the context of the section 5107(b) equipoise
standard in order to determine whether it is appropriate in a given case for the Court to reverse a
finding of fact – that is, declare the correct finding of fact. See supra note 17 and accompanying
text. This assessment is, in effect, the process that the Court recently followed in Mariano, when,
as noted above, it held that the Board's application of the section 5107(b) equipoise standard as to
three findings of fact was clearly erroneous; the Court reversed two of those findings and set aside
(but did not reverse) and remanded one finding because evidence in the claimant's favor on that
question was not present. Compare 

and 

I note that, to the extent that there is ambiguity in the statutory language, as the
Roberson III order 

, the analysis in this separate statement was
formulated in light of the direction of the Supreme Court to resolve ambiguity in a veterans benefits
statute in favor of the claimant. Brown v. Gardner, 

, 118 (1994) (directing that
reasonable doubt in statutory interpretation is to be ''resolved in the veteran's favor"); 

(quoting Brown v. 

); see also Jones (Ethel) v. West, 

,
1299 n.2 (Fed. Cir. 1998); Allen (William) v. Brown, 

, 446 (1995) (en banc).
Moreover, it is a longstanding maxim of statutory interpretation that remedial legislation, such as
VBA § 401, is to be interpreted broadly in order to effectuate its basic purpose. See Smith (William)
v. Brown, 

, 1525 (Fed. Cir. 1994) ("courts are to construe remedial statutes liberally to
effectuate their purposes . . . [and v]eterans benefits statutes clearly fall in this category"). Thus, I
see no basis for the Court to choose to construe narrowly the VBA provisions given their proclaimant
remedial purpose. See Splane v. West, 

, 1068-69 (Fed. Cir. 2000) ("canons of
construction . . . require us to give effect to the clear language of a statute and avoid rendering any
portions meaningless or superfluous"). When Congress writes legislation that is less than completely
clear, it is the judiciary's role to make the best of the language that is enacted and to seek to find a
reasonable interpretation of the statutory text consistent with the goals that Congress has indicated
it sought to achieve with that legislation. Only if, after undertaking this analysis, the only reasonable
conclusion is that Congress, notwithstanding its intention, failed to provide statutory language that
can be fairly interpreted as achieving its basic legislative purpose should a court tell Congress "nice
try, but you haven't done the job you apparently intended to do."

                                                    III. Conclusion



         21
             In applying the "clearly erroneous" standard, the Board's determinations about the credibility of live-witness
testimony before the Board (such as that presented by the appellant here (Record at 443-64)) must be afforded substantial
deference. See Anderson v. City of Bessemer, 

, 575 (1985) (holding that trial judge's choice "if not
internally inconsistent", between credible testimony of live witnesses, "each of whom has told a coherent and facially
plausible story that is not contradicted by extrinsic evidence, . . . can virtually never be clear error"); Zenith Radio Corp.
v. Hazeltine Research, Inc., 

, 123 n.18 (1969); United States v. Yellow Cab Co., 

, 341-42
(1949).

                                                             17
         Because I believe that the question of the meaning of the VBA § 401 amendments has not
been clearly and definitively addressed by our Court in light of our decisions in Gilbert, Roberson
III, and Mariano, and has been forcefully and effectively argued by the appellant in his brief, I voted
for full-Court examination of these matters. At a minimum, I would require a response from the
Secretary before the Court votes on the appellant's motion for initial full-Court consideration. See
U.S. VET . APP . R. 35(g).

         KASOLD, Judge, dissenting: I respectfully dissent from the denial of en banc consideration
in this case. The appellant seeks en banc review so that the Court might address (1) our application
of the "clearly erroneous" standard of review; (2) the benefit-of-the-doubt rule; and (3) the doctrine
that once this Court finds an issue warranting remand, we generally will not address other issues (aka
the Best rule). I believe en banc review is warranted because these issues are of exceptional
importance to our judicial review and there is a lack of uniformity or clarity in our caselaw in each
of these areas. See U.S. VET . APP . R. 35(c) (motions for full-Court decision will ordinarily not be
granted "unless such action is necessary to secure or maintain uniformity of the Court's decisions or
to resolve a question of exceptional importance").

                            A. "Clearly Erroneous" Standard of Review

        The meaning of the term "clearly erroneous" was first set out by this Court in Gilbert v.
Derwinski, 

, 52-53 (1990). Gilbert cites to United States v. United States Gypsum Co.,

, 395 (1948), for the definition of this term: "A finding is 'clearly erroneous' when
although there is evidence to support it, the reviewing court on the entire evidence is left with the
definite and firm conviction that a mistake has been committed." Gilbert then cites to Anderson v.
City of Bessemer City, 

, 573-74 (1985), to illustrate when a court might not have a firm
conviction:

       This standard plainly does not entitle a reviewing court to reverse the finding of the
       trier of fact simply because it is convinced that it would have decided the case
       differently. The reviewing court oversteps the bounds of its duty under Rule 52(a)
       if it undertakes to duplicate the role of the lower court. "In applying the clearly
       erroneous standard to the findings of a district court sitting without a jury, appellate
       courts must constantly have in mind that their function is not to decide factual issues
       de novo." Zenith Radio Corp. v. Hazeltine Research, Inc., 

, 123 (1969).
       If the district court's account of the evidence is plausible in light of the record viewed
       in its entirety, the court of appeals may not reverse it even though convinced that had
       it been sitting as the trier of fact, it would have weighed the evidence differently.
       Where there are two permissible views of the evidence, the factfinder's choice
       between them cannot be clearly erroneous. United States v. Yellow Cab Co., 

, 342 (1949); see also Inwood Laboratories, Inc. v. Ives Laboratories, Inc.,
       

(1982).

(quoting 

).


                                                  18
         Although Gilbert favorably cited to the Supreme Court’s definitions of the "clearly
erroneous" standard, Gilbert subsequently stated that "if there is a plausible basis in the record for
the factual determinations of the BVA, even if this Court might not have reached the same factual
determinations, we cannot overturn them." 

is this standard that is primarily used
throughout our cases. Whether or not the Gilbert standard was designed to be different from that
enunciated by the Supreme Court, the fact remains it is. The Supreme Court stated that a "finding
is 'clearly erroneous' when, although there is evidence to support it, the reviewing court on the entire
evidence is left with the definite and firm conviction that a mistake has been committed." 

(emphasis added), quoted favorably in Concrete Pipe & Prods. of Cal., Inc., v.
Constr. Laborers Pension Trust for S. Cal., 

, 622-23 (1993). Thus, under the standard
established by the Supreme Court, even when there is evidence to support a lower tribunal’s finding,
the reviewing court can reverse it if it has a definite and firm conviction that a mistake has been
made, based on the entire record. By contrast, under the Gilbert standard, if there is a plausible basis
in the record for the decision below, it cannot be overturned.

        The Supreme Court also stated that a decision below was not clearly erroneous when that
decision provided an "account of the evidence [that] is plausible in light of the record viewed in its
entirety." 

(emphasis added). This requires both (1) an explanation of the evidence
and (2) that the explanation be plausible "in light of the record viewed in its entirety." 

basis in the record is not enough to preclude a reversal of a finding; there must be an
explanation that is plausible "in light of the record viewed in its entirety." Of course, if there is a
plausible explanation in light of the entire record, the Court cannot substitute its own, alternate
plausible basis for that of the lower tribunals. Id.; see also Yellow 

.

        Although the Gilbert standard is generally used in this Court, it is not uniformly applied.
There are a number of cases applying an even stricter, dual-based definition of the "clearly
erroneous" standard that permits reversal only when "there is absolutely no plausible basis" for the
BVA decision and where that decision "is clearly erroneous in light of the uncontroverted evidence."
See, e.g., Kay v. Principi, 16 Vet.App 529, 533 (2002); Shoffner v. Principi, 16 Vet.App 208, 212
(2002); Pentecost v. Principi, 16 Vet.App 124, 129 (2002).

        The genesis and development of this two-pronged standard is a classic example of the
evolution of case meanings over time. The phrase "absolutely no plausible basis" was first used in
our cases in Hersey v. Derwinski, 

, 95 (1992). After restating the Supreme Court’s
definitions of the "clearly erroneous" standard, as well as noting it as the Gilbert standard,1 the
Hersey Court then noted that in the case before it "[t]here is absolutely no plausible basis for the
BVA’s statement" and then found that the BVA decision "was clearly erroneous in light of the
uncontroverted evidence in appellant’s favor." 

"absolutely no plausible basis" and


          1
             The Hersey court made no mention of the standard actually enunciated in Gilbert and applied in most of our
cases, i.e., "if there is a plausible basis in the record for the factual determinations of the BVA, even if this Court might
not have reached the same factual determinations, we cannot overturn them," Gilbert v. Derwinski, 

, 53
(1990). See Hersey v. Derwinski, 

, 93-95 (1992).

                                                            19
the "uncontroverted evidence" comments were not intended to be a standard of review. Rather, they
were the Court’s view of the evidence before it; i.e., there was not only no plausible basis in the
record as a whole for the Board’s finding, but there was also absolutely no basis for it given the
uncontroverted evidence against it. For whatever reasons, numerous single-judge decisions began
citing Hersey for this two pronged, stricter proposition. See, e.g., Abrams v. Brown, No. 92-435,

(Vet. App. Sept. 22, 1993); Moon v. Brown, No. 91-2044, 

(Vet.
App. Feb. 01, 1994). This proposition later made its way into the panel decisions noted above
(Pentecost, Shoffner, and Kay, 

).

         I suspect that the Gilbert standard and the evolutionary Hersey standard were both
unintentional deviations from the Supreme Court’s definition of "clearly erroneous." Regardless,
this is an exceptionally important issue on which we do not have uniformity within the Court, and
therefore en banc review is warranted.2

                                           B. The Benefit of the Doubt

         Section 5107(b) of title 38, U.S. Code, requires the Secretary to consider all the information
and evidence in a veteran's case and, if "there is an approximate balance of positive and negative
evidence regarding any issue material to the determination of a matter," to give the benefit of the
doubt to the veteran/claimant. This Court established early on that the Secretary’s application of the
benefit-of-the-doubt rule, as finally applied by the Board, would be reviewed for whether it was
"'arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.'" 

.

        What was once well settled, however, is now open to question as a result of the panel
decisions in Roberson v. Principi, 

, 146 (2003), and Mariano v. Principi,
17 Vet.App 305, 313 (2003), which state the standard of review of the benefit-of-the-doubt rule as
"clearly erroneous." Neither case cites a basis for changing the standard of review, with Mariano
simply citing Roberson. See 

. Indeed, neither case cites Gilbert. Moreover,
Roberson gives the impression that "clearly erroneous" was always the standard of review for the
Secretary's application of the benefit-of-the-doubt rule. 

conflict in our caselaw on this
significant issue and it should be resolved en banc.

                                               C. The Best Doctrine

       The panel decision in Best v. Principi, 

, 19 (2001) (per curiam order),
proclaimed that "[f]rom the outset, it has been the practice of this Court that when a remand is
ordered because of an undoubted error that requires such a remedy, the Court will not, as a general


         2
           Although I believe the differences between the Gilbert and Hersey standards of review are significant, to the
extent some view them as subtle this should not preclude review. The Supreme Court noted there was a subtle difference
between the "substantial evidence" and the "clearly erroneous" standards of review, but still found it important to apply
the correct standard. Dickinson v. Zurko, 

, 162-63 (1999).

                                                          20
rule, address other putative errors raised by the appellant." See also Mahl v. Principi, 

(2001) (per curiam order) (acknowledging departures from the Best doctrine, but maintaining they
are the exception rather than the rule). In support of its proposition, Best cites directly to Dunn v.
West, 

, 467 (1998), and further refers the reader to Aronson v. Brown, 

, 155 (1994), and Mokal v. Derwinski, 

(1990). A closer look, however, finds that
none of these cases supports the broad proposition announced in Best. Indeed, Dunn supports a
much narrower proposition and almost inverse general rule.

        Dunn fully addressed the appellant’s assertion on appeal that the Board had erred and
affirmed the Board on three claims and remanded the fourth. 

. The
fourth claim was remanded because the Court found that the Board had failed to actually obtain
records that appeared to be relevant to that claim. These records were deemed to be constructively
before the Board, and therefore it was error not to consider them; remand was appropriate. 

would have to be obtained and considered on remand, the additional assertions
that the Secretary failed in his duty to assist by failing to secure the records or that the veteran was
denied due process by the Secretary's failing to secure those records were rendered moot. Nothing
in Dunn stands for the proposition that this Court had a practice of not addressing other claims
simply because one claim warranted a remand; Dunn suggests only that those issues mooted by the
remand are not to be considered.

        It also seems that Best actually turned the Dunn approach on its head, with its "recognition"
of a general rule that if a finding or holding on one claim or issue causes remand of a case, none of
the other claims or issues generally should be decided at that time by the Court. Although Dunn
refused to address additional issues that were mooted by decision on one issue, application of the
Best rule does not depend on mooting an issue; it simply remands all claims and issues in the case
to the Board where they all can be reconsidered. See Fletcher v. Derwinski, 

, 397
(1991) ("A remand is meant to entail a critical examination of the justification for the decision. The
Court expects that the [Board] will reexamine the evidence of record, seek any other evidence the
Board feels is necessary, and issue a timely, well-supported decision in this case.").

        The other cases relied on by Best similarly do not support the general rule as announced in
Best. Aronson stands for the sole proposition that when a claim is rendered moot, the Court no
longer has jurisdiction over that claim or any matters appurtenant to that claim. 

. Mokal stands for the proposition that this Court has adopted the case-or-controversy
jurisdictional restraints imposed by Article III of the U.S. Constitution. 

.
Both cases support the general Dunn approach, i.e., not addressing issues mooted by a remand, but
not the broad rule enunciated in Best that if one issue warrants remand none of the others generally
should be addressed.

        Best further cites to a number of other courts that "regularly decline to address the remaining
allegations of error if the court orders a remand and a new trial based on any one allegation of error."

. Reliance on these cases for adopting the broad principle enunciated in Best
is misplaced. For one, a new trial is far different than the remand of a case from this Court to the


                                                  21
Board. In a new trial, the entire process is redone, whereas during new proceedings after a remand
from this Court, the record below remains intact; it can be augmented, but it is not redone.
Moreover, with the exception of the issue on which the remand was based and those, as in Dunn, that
were mooted by the remand, I fail to see the basis for believing that the Board will render a different
decision on any remaining issues. This leaves the veteran with the prospect that his case will be
remanded as many times as he has separate claims or independent issues, before he can get a final
decision. See Brambley v. Principi, 

, 25-28 (2003) (Steinberg, J., concurring).
Unlike the moot issues in Dunn, the issues and claims remanded under the Best construct remain
very much alive.

         Finally, in addition to being, at best, a stretch from the Dunn case, the Best principle is
applied inconsistently and without uniformity in the Court. Compare, e.g., Best and Mahl, 

, with Pelegrini v. Principi, 

, 423-24 (2004) (remanding for compliance with
38 U.S.C. § 5103(a) but also addressing another argument raised in case), and Moran v. Principi,

49, 155-56 (2003) (remanding for inadequate reasons or bases, but also addressing
appellant's arguments regarding 38 C.F.R. § 3.304(f)); see also 

(Steinberg, J.,
concurring) (enumerating other remandable errors, outside of those addressed by majority option,
that without having been so addressed might reoccur on readjudication). This issue is important and
there is a lack of uniformity in this area within our cases, both warranting clarification by the en banc
Court.3

         For the foregoing reasons, I respectfully dissent.




         3
           See also Mahl v. Principi, 

, 40-47 (2001) (per curiam order) (Steinberg, J., dissenting) (noting
additional infirmities in the Best rule).

                                                           22