Slip Op. 00 - 110
UNITED STATES COURT OF INTERNATIONAL TRADE
BEFORE: RICHARD W. GOLDBERG, JUDGE
LIBAS, LTD.,
Plaintiff,
v.
Court No. 95-01-00014
UNITED STATES,
Defendant.
[Upon remand, Customs’s test fails to meet the standards of
reliability articulated by the Court of Appeals for the
Federal Circuit. The goods shall be reliquidated under HTSUS
5208.42.10.]
Dated: August 29, 2000
Law Offices of Elon A. Pollack (Elon A. Pollack and Eugene P.
Sands), for plaintiff.
David W. Ogden, Assistant Attorney General; Joseph I.
Liebman, Attorney in Charge, International Trade Field Office,
Commercial Litigation Branch, Civil Division, United States
Department of Justice; Bruce N. Stratvert, Attorney,
International Trade Field Office, Commercial Litigation
Branch, Civil Division, United States Department of Justice;
Edward Maurer, of counsel, Office of Assistant Chief Counsel,
International Trade Litigation, United States Customs Service;
for defendant.
Libas, Ltd. v. United States, No. 95-01-00014 Page 2
OPINION
GOLDBERG, Judge: This classification case involves 32
bales of cotton fabric imported from India into the United
States in 1994. The United States Customs Service (“Customs”)
tested the fabric according to its “Methodology for the
[A]nalysis of Woven Fabric to Determine Whether Fabric had
been Power-loomed or Hand-loomed” (Customs’s test). Based on
the results, Customs determined the fabric was power-loomed
and classified it under subheading 5208.42.40 of the
Harmonized Tariff Schedule of the United States (“HTSUS”),
dutiable at a rate of 11.4% ad valorem and subject to a quota
restriction.
Plaintiff, Libas, Ltd., initiated this action in
1995 to challenge Customs’s classification. Plaintiff argued
that the fabric was hand-loomed, and should therefore have
been classified under HTSUS 5208.42.10, dutiable at a rate of
6% ad valorem. Plaintiff also argued that Customs was
required to accept the government of India’s certification
that the fabric was hand-loomed.
The Court held trial in May, 1996. In its
subsequent opinion, Libas Ltd. v. United States, 20 CIT 1215,
944 F. Supp. 938 (1996), aff’d in part and vacated in part,
Libas, Ltd. v. United States, No. 95-01-00014 Page 3
193 F.3d 1361 (Fed. Cir. 1999) (“Libas II”), the Court
sustained Customs’s classification. First, the Court held
that Customs was not required to accept as dispositive the
government of India’s certification that the fabric at issue
was hand-loomed; in the Court’s view, Customs acted within its
statutory authority when it independently assessed whether the
fabric at issue was hand-loomed or power-loomed. See 20 CIT
at 1218, 944 F. Supp. at 941. Second, based on the evidence
and testimony adduced at trial, the Court determined that the
fabric was properly classified as power-loomed. See 20 CIT at
1220, 944 F. Supp. at 942.
On appeal, the United States Court of Appeals for
the Federal Circuit (“Federal Circuit”) affirmed the Court’s
determination that Customs had the authority to independently
assess and reclassify fabric that had been certified as hand-
loomed by the Indian government. See Libas II, 193 F.3d at
1364. The Federal Circuit, however, vacated and remanded the
Court’s determination that the fabric was power-loomed. See
id. at 1369.
By statute, Customs’s classification of goods is
presumed to be correct. See 28 U.S.C. § 2639 (1994). The
presumption applies to every subsidiary fact necessary to
Libas, Ltd. v. United States, No. 95-01-00014 Page 4
support classification, see Commercial Aluminum Cookware Co.
v. United States, 20 CIT 1007, 1013, 938 F. Supp. 875, 881
(1996), including the “methods of weighing, measuring, and
testing merchandise used by customs officers and the results
obtained” therefrom. Exxon Corp. v. United States, 462 F.
Supp. 378, 381 (Cust. Ct. 1978) (quoting Consolidated Cork
Corp. v. United States, 54 Cust. Ct. 83 (1965)), aff’d 607
F.2d 985 (C.C.P.A. 1979). An importer may rebut the
presumption of correctness by “showing that [Customs’s]
methods or results are erroneous.” Id. at 382 (quoting same).
“If a Prima facie case is made out, the presumption is
destroyed and the Government has the burden of going forward
with the evidence.” Id. (quoting same).
In this case, Customs’s classification of the fabric
as power-loomed and the test Customs used to arrive at that
determination were both presumed to be correct. The Federal
Circuit found that Customs’s presumption of correctness had
been overcome, however, because “Libas’ [sic] argument at
trial against the reliability of [Customs’s] test was
sufficient to rebut the statutory presumption of correctness
accorded Customs classifications.” Libas II, 193 F.3d at 1366
n.2.
Libas, Ltd. v. United States, No. 95-01-00014 Page 5
Given this posture, the Federal Circuit found
wanting the Court’s determination that Customs’s
classification was correct. In the Federal Circuit’s view,
the Court relied solely on the results of Customs’s test,
filed as part of the official record, to conclude that the
fabric was power-loomed, see id. at 1365, but “did not
ascertain whether, or explain why, the Customs test was
reliable according to appropriate standards.” Id. at 1367.
To assess the reliability of Customs’s test, the
Federal Circuit stated that the Court should have employed the
standards articulated by the United States Supreme Court in
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579
(1993). See Libas II, 193 F.3d at 1366-67. The Daubert
standards are: (1) whether a theory or technique, such as
Customs’s test, has been tested; (2) whether it has been
subjected to peer review and publication; (3) its known or
potential rate of error; and (4) whether it is generally or
widely accepted. See Daubert, 509 U.S. at 593-94.
Importantly, the Federal Circuit counseled that the Daubert
standards bear not only on whether evidence is admissible, but
also on how much or how little weight the Court should accord
Libas, Ltd. v. United States, No. 95-01-00014 Page 6
such evidence.1 See Libas II, 193 F.3d at 1366.
In light of the Daubert standard, the Federal
Circuit found the record before it “insufficient . . . to make
a determination of . . . [the] reliability [of Customs’s test]
with any confidence,” and advised that “[f]urther evidentiary
hearings are probably called for.” Id. at 1369. In accordance
with those instructions, the Court conducted a hearing to
assess the reliability of Customs’s test.
At the hearing, defendant failed to establish that
its test satisfied any of the Daubert standards cited by the
Federal Circuit. And while the Daubert factors are not a
1
The Court makes no judgment as to whether the
Federal Circuit’s determination that Daubert bears on weight
as well as admissibility is limited to cases in which, as
here, the Court is acting as the trier of fact and the
evidence at issue is already part of the record. Cf. Exxon
Corp. v. United States, 45 Fed. Cl. 581, 682 n.206 (Fed. Cl.
1999) (internal citations omitted) (“Daubert and Kumho Tire
are, of course, concerned with the admissibility of expert
opinion testimony under Federal Rule of Evidence 702. Here at
bar, in contrast, we address the sufficiency of expert opinion
testimony already in the record . . . However, our application
of the Daubert standard of evidentiary reliability is
consistent with the ‘hard look’ doctrine, under which the
district courts have a duty to evaluate the reliability of
expert opinion testimony, even after such testimony is in the
record, in order to determine whether the case should go to
the jury. Here at bar, sitting as the trier of fact, this
court thinks that it is clear beyond cavil that the Daubert
reliability standard may properly be taken into consideration
in evaluating the probative weight of expert opinion testimony
already in the record.”).
Libas, Ltd. v. United States, No. 95-01-00014 Page 7
“definitive checklist or test,” Daubert, 509 U.S. at 593,
defendant also failed to demonstrate that its test bears any
other indicia of reliability.
According to Daubert, one of the “key question[s]”
the Court should consider is whether a theory or technique
“can be (and has been) tested.” 509 U.S. at 593. Daubert
also directs the Court to “consider the known or potential
rate of error” of a theory or technique. Id. at 594. In this
case, defendant’s three expert witnesses stated that, in their
opinion, Customs’s test was a reliable method for
distinguishing between hand-loomed and power-loomed fabric.
Defendant failed, however, to demonstrate that Customs’s test
(1) measures what it purports to, and (2) does so within an
acceptable rate of error.
The Federal Circuit noted
that the reliability of the test has not
been established by the obvious and natural
method of double-blind testing. That would
involve running the Customs test on fabric,
the source of which was known in some other
way, perhaps by direct observation, and
determining whether testers who themselves
had no knowledge of whether test samples
were hand-loomed or power-loomed could
reliably distinguish power-loomed from hand-
loomed fabric within a respectable rate of
error. Testing a methodology in this manner
would satisfy two of the Daubert factors,
verification and known error rate, and for
Libas, Ltd. v. United States, No. 95-01-00014 Page 8
this reason would enhance confidence in the
reliability of the test.
Id. at 1368. When asked by the Court whether Customs’s test
could be tested in the manner described above, Dr. Irene Good,
a specialist in textile and fiber analysis and one of
defendant’s experts, answered in the affirmative. Yet,
defendant presented no evidence that Customs’s test had ever
been tested in this manner.
In fact, Dr. Desiree Koslin, one of defendant’s
experts from the trial proceedings, testified that she had
tested Customs’s test. Yet, prior to applying Customs’s test
to a given piece of fabric, she knew in advance whether that
fabric was hand-loomed or power-loomed. Because she was not
“blind” to the correct answer, the Court attaches less weight
to Dr. Koslin’s testimony that Customs’s test is a reliable
method for distinguishing between hand-loomed and power-loomed
fabrics.2 Cf. Ruffin v. Shaw Indus., Inc., 149 F.3d 294, 300
(4th Cir. 1998) (finding testimony inadmissible because, among
other things, examiner was not “blinded” to which mice were
2
Dr. Koslin also testified that her personal method
of examination “dovetails” with Customs’s test. Notably,
however, Customs did not offer evidence that Dr. Koslin’s
methodology has itself been tested for accuracy, i.e. that Dr.
Koslin is able to identify, without advance knowledge of the
correct result, fabric as hand-loomed or power-loomed within
an acceptable rate of error.
Libas, Ltd. v. United States, No. 95-01-00014 Page 9
control group members and which were experimental).
Moreover, both Dr. Koslin and Ms. Mary Carrillo,
Textile Analyst at the United States Customs Laboratory and
one of defendant’s witnesses at the trial proceedings,
testified that Customs’s test was 100% foolproof and thus had
an error rate of zero. Ms. Carrillo further testified that
Customs’s test is completely error-free because multiple
analysts, each with years of experience and training, apply it
to each sample. Ms. Carrillo testified that in her
experience, Customs’s test has never led to inconclusive
results, nor have analysts disagreed as to the origin of a
particular fabric.
The Court attaches little weight to Dr. Koslin and
Ms. Carrillo’s testimony on this point. First, little
credence can be accorded the witnesses’ belief that Customs’s
test is foolproof, when the accuracy of Customs’s test has
never been measured in any scientific way. Second, that
Customs analysts never reach different conclusions as to
whether fabric is hand-loomed or power-loomed is questionable
in light of the same experts’ testimony that a number of the
criteria in Customs’s test are “qualitative” or “subjective.”
Indeed, Customs analysts must use their judgment to determine
Libas, Ltd. v. United States, No. 95-01-00014 Page 10
a number of criteria, such as whether cut fringe is “uniform,”
variation between areas is “minimal,” yarn is “complex,” and
knots are “minimal.” See Def.’s Ex. 1. For the foregoing
reasons, Customs’s test fails to meet Daubert’s standards of
testability and error rate.
Under Daubert, “[w]idespread acceptance can be
an[other] important factor in” assessing the reliability of a
theory or technique. Daubert, 509 U.S. at 594. In an attempt
to demonstrate such acceptance, defendant offered Exhibit 13,
a “Check-Sheet for Identification of Handloom Items.”
Defendant claimed that the “check-sheet” was used by the
Government of India (“GOI”) to distinguish between hand-loomed
and power-loomed fabrics. Another exhibit purported to show
the concordance between the “check-sheet” and Customs’s test.
See Def.’s Ex. 1.
Defendant was unable to authenticate the “check-
sheet,” however. Although the words “Government of India
(GOI)” were handwritten at the top of the document, the
government witness testifying at the time, Mr. Richard
Crichton of the U.S. Customs Service, did not know who wrote
them. The “check-sheet” was also undated and unsigned. Given
its uncertain lineage and defendant’s failure to offer an
Libas, Ltd. v. United States, No. 95-01-00014 Page 11
affidavit certifying that it was an accurate representation of
the actual test used by the GOI, the Court excluded the
“check-sheet.” See Fed. R. Evid. 901 (requiring
authentication or identification).
Aside from its attempt to demonstrate the purported
similarities between Customs’s test and the GOI’s, defendant
did not offer evidence to show similarities to any other test.
Upon examination, Dr. Koslin testified that countries other
than the United States and the GOI, such as France, must
distinguish between hand-loomed and power-loomed fabrics, yet
defendant did not submit a copy of the test used by any other
country. And other than the testimony of Dr. Koslin that the
methodology she uses and teaches is based on similar factors,
defendant did not offer documentation of a test used by any
other entity, such as a museum, auction house, conservation
organization, private lab, or the American Society for Testing
and Materials. In this way, defendant failed to demonstrate
that its test enjoyed widespread acceptance among other
countries and organizations concerned with distinguishing
between hand-loomed and power-loomed goods.
Finally, defendant failed to show that its test has
been published and subjected to peer review. See Daubert, 509
Libas, Ltd. v. United States, No. 95-01-00014 Page 12
U.S. at 593. The Supreme Court acknowledged that “[s]ome
propositions . . . are . . . of too limited interest to be
published,” id., and that “[i]t might not be surprising in a
particular case . . . that a claim made by a scientific
witness has never been the subject of peer review, for the
particular application at issue may never previously have
interested any scientist.” Kumho Tire Co., Ltd. v.
Carmichael, 526 U.S. 137, 151 (1999). It may well be that the
community concerned with distinguishing between hand-loomed
and power-loomed fabrics is extremely limited. Nonetheless,
the Court comments on this factor because, while not
dispositive, “submission to the scrutiny of the scientific
community is a component of ‘good science,’ in part because it
increases the likelihood that substantive flaws in methodology
will be detected.” Daubert, 509 U.S. at 593.
According to Ms. Carrillo, Customs’s test has never
been published outside of the Customs Technical Bulletin.
Further, Ms. Carrillo was not aware of any publication that
discussed Customs’s test, nor did she believe that it had been
the subject of peer review. While Ms. Carrillo believed that
other labs used tests similar to that of Customs, she could
not point to peer review or publication of those tests. And
Libas, Ltd. v. United States, No. 95-01-00014 Page 13
while Dr. Koslin testified that she had tested Customs’s test,
she did not publish her results.
The bulk of the evidence presented by defendant at
trial focused on the test Customs used to determine that the
fabric at issue was power-loomed. See Libas, 20 CIT at 1218-
20, 944 F. Supp. at 942-43. Upon reconsideration, Customs’s
test does not meet any of the Daubert factors, nor did
defendant point to any other indicia of reliability.
Therefore, the Court now accords the test little weight.
Apart from Customs’s test, defendant’s experts testified that,
in their opinion, the fabric was power-loomed. Their personal
methodologies for determining this, however, have themselves
never been tested, have no known error rate, have never been
published, and have never been subjected to peer review.
Without reliable evidence, defendant fails to prove that the
fabric at issue was power-loomed.
At the original trial, several of plaintiff’s
witnesses offered compelling testimony that, based on first-
hand experience, the fabric at issue was hand-loomed.
Plaintiff’s main witness, Mary Jane Leland, Professor Emeritus
at California State University at Long Beach, testified that
the fabric at issue is typical of fabric produced on a hand-
Libas, Ltd. v. United States, No. 95-01-00014 Page 14
powered fly shuttle loom in the Madras area of India.
Professor Leland testified that she has observed “skilled
master weaver[s who] can loom fabric by hand with results that
cannot be distinguished from those obtained by a machine
loom.” Libas, 20 CIT at 1220, 944 F. Supp. at 942. Further,
Mr. S. Ponnuswamy, partner in JLC International (“JLC”) of
Madras, India, testified that JLC purchased the fabric at
issue from two master weavers located in Kovur, India. Mr.
Ponnuswamy testified that he personally observed similar
fabric being hand-loomed in Kovur under the supervision of the
master weavers.
Because Customs’s test does not meet the standards
for reliability, the weight of the evidence now supports the
conclusion that the fabric is hand-loomed. Accordingly, the
fabric shall be reliquidated under HTSUS 5208.42.10.
__________________________________
Richard W. Goldberg
JUDGE
Dated: August 29, 2000
New York, New York.