Woodfox v. Foti et al

Filing 274

RULING granting 1 Petition for Writ of Habeas Corpus filed by Albert Woodfox. Signed by Judge James J. Brady on 2/26/2013. (LLH)

UNITED STATES DISTRICT COURT
MIDDLE DISTRICT OF LOUISIANA
ALBERT WOODFOX
CIVIL ACTION
VERSUS
NO. 06-789-JJB
BURL CAIN, WARDEN, LOUISIANA
STATE PENITENTIARY, ET AL
RULING
This matter is before the Court on Petitioner Albert Woodfox’s (“Woodfox”)
petition for habeas relief on the claim that Woodfox’s March 1993 indictment by a
West Feliciana Parish grand jury was tainted by grand jury foreperson
discrimination. An evidentiary hearing was held on May 29-31, 2012. Both
Woodfox and the State have filed post-hearing memoranda. (Docs. 259 and 258
respectively). Both Woodfox and the State have also filed post-hearing reply
memoranda. (Docs. 267 and 266 respectively). Finally, Woodfox filed a motion to
strike portions of the State’s post-hearing response memorandum and motion for
leave to file in reply. (Doc. 268). The State filed an opposition (Doc. 270), to
which Woodfox filed a reply. (Doc. 272). The Court will not strike any portions of
the State’s post-hearing response memorandum. For the foregoing reasons,
Woodfox’s habeas relief is GRANTED.

1

I.
In March 1993, Woodfox was indicted by a West Feliciana Parish grand
jury, and in December 1998, Woodfox was convicted of second-degree murder.1
On October 11, 2006, Woodfox filed a Petition for Habeas Corpus Relief in this
Court, challenging his 1998 conviction and sentence. On June 10, 2008,
Magistrate Judge Noland ruled that Woodfox had presented sufficient evidence
to support a prima facie case of grand jury foreperson discrimination.2 (Doc. 33 at
64). This Court adopted the Magistrate Judge’s findings and granted Woodfox’s
petition on July 8, 2008. (Docs. 35 and 48). The United States Court of Appeals
for the Fifth Circuit reversed this Court’s ruling on the issues presented for review
and remanded for a determination concerning the selection of the grand jury
foreperson. Woodfox v. Cain, 609 F.3d 774, 817-18 (5th Cir. 2010). Upon
remand, Woodfox presented his claim based upon statistics. On February 16,
2011, after oral argument, this Court concluded that AEDPA deference to the
Louisiana First Circuit Court of Appeals was unwarranted and ordered an
evidentiary hearing, which was held on May 29-31, 2012. (Doc. 100).
II.
To establish a prima facie case of grand jury foreperson discrimination, a
defendant must show: (1) the group to which the defendant belongs is a
1

Because the procedural history of this case is lengthy, the Court will not repeat it. A more complete
description of the procedural history can be found in the Magistrate Judge’s ruling (Doc. 33), which this
Court has previously adopted. (Doc. 48).
2
For the purposes of this ruling, this Court will not discuss the other issues in this case, which have been
resolved. This ruling focuses solely on the issue before the Court, the matter of whether there was grand
jury foreperson discrimination.

2

“recognizable, distinct class, singled out for different treatment under the laws, as
written or as applied”; (2) “the degree of underrepresentation must be proved, by
comparing the proportion of the group in the total population to the proportion
called to serve as grand jurors, over a significant period of time; and (3) “a
selection procedure that is susceptible of abuse or is not racially neutral supports
the presumption of discrimination raised by the statistical showing.” Castaneda v.
Partida, 430 U.S. 482, 494 (1977) (internal citations omitted). Once the
defendant has made a prima facie showing of grand jury foreperson
discrimination, the burden shifts to the State to rebut the showing. Id. Both the
Magistrate Judge and this Court have found that Woodfox made a prima facie
showing of grand jury foreperson discrimination. (Docs. 33, 100, and 237).
Woodfox is African-American, and African-Americans constitute a distinct,
cognizable class. Rose v. Mitchell, 443 U.S. 545, 565 (1979). Woodfox has also
shown substantial underrepresentation by comparing the proportion of AfricanAmericans in West Feliciana Parish to the proportion called to serve as grand
jurors over a substantial period of time. This Court has ruled that the relevant
period of time is 1980 through March 1993. (Doc. 237). During this time period,
there were 297 non-foreperson grand jurors, and the voter registrar and deputy
registrar were able to identify the race of 277 of those jurors. Out of the 277 nonforeperson grand jurors, 113 were African-American, or 40.8%. 5 out of the 27
forepersons appointed were African-American, or 18.5%. (Doc. 229). However, in
1990, the percentage of African-Americans in West Feliciana Parish, excluding
3

African-Americans incarcerated in Louisiana State Penitentiary at Angola, was
44%. Similarly, the percentage of African-Americans among registered West
Feliciana Parish voters between 1980 and 1993 was 43.5%. Finally, Louisiana’s
procedure for selecting grand jury forepersons prior to 1999 was “unquestionably
subject to abuse according to subjective criteria that may include race and
gender.” State v. Langley, 1995-1489 (La. 4/3/02); 813 So.2d 356, 371.
Upon making a prima facie case, the burden of proof is shifted to the State
to “dispel the inference of intentional discrimination.” Castaneda, 430 U.S. at
497-98. The State may rebut the prima facie case by showing “evidence that
objective, racially neutral criteria were used in the selection process.” Johnson v.
Puckett, 929 F.2d 1067, 1072 (5th Cir. 1991); see also Alexander v. Louisiana,
405 U.S. 625, 632 (1972) (finding that the burden of proof shifts to the State “to
rebut the presumption of unconstitutional action by showing that permissible
racially neutral selection criteria and procedures have produced the
monochromatic result.”); see also Guice v. Fortenberry, 722 F.2d 276, 280 (5th
Cir. 1984) (finding that the State “must show that the pattern of
underrepresentation proved . . . was the result of a ‘racially neutral selection
procedure.’”)(citing Alexander, 405 U.S. at 632). “Affirmations of good faith in
making individual selections are insufficient to dispel a prima facie case of
systematic exclusion.” Alexander, 405 U.S. at 632.

4

III.
In making its rebuttal argument, the State relied on two categories of
evidence to dispel the inference of discrimination. First, the State presented
expert statistics testimony to show that the data relied on by Woodfox in his
prima facie case does not support an inference of discrimination in the grand jury
foreperson selection process for the relevant period of time. (Doc. 258 at 4-5).
Second, the State presented evidence to show that judges in West Feliciana
Parish relied on racially neutral criteria in making their foreperson selections. (Id.
at 5).
The State presented the reports and testimony of its expert, Tumulesh
Solanky, PhD., to demonstrate that any observed racial disparity is statistically
insignificant. The State argued that courts have recognized that data, which
demonstrates a racial disparity on its face, can be misleading without deeper
analysis. The State cites Moultrie v. Martin, 690 F.2d 1078 (4th Cir. 1981) for the
proposition that courts should use statistical analysis methods to evaluate grand
jury discrimination claims. (Doc. 258 at 5). However, in Moultrie v. Martin, the
United States Court of Appeals for the Fourth Circuit was considering whether
the petitioner had made a prima facie showing through the use of statistical
evidence, and not whether the State had dispelled an inference of discrimination.
Moultrie, 690 F.2d at 1079 (finding that the petitioner’s claim was “based wholly
on statistics . . . [and because] he did not establish a prima facie case . . . [w]e
need not address the matter of rebuttal.” Id.). The State also cites Hillery v.
5

Pulley, 563 F.Supp. 1228 (E.D. Cal. 1983) for support. Similarly, in Hillery, the
court analyzed statistical evidence to determine whether the petitioner had made
out a prima facie case of discrimination, not whether the State had dispelled such
an inference. Hillery, 563 F.Supp. at 1249 (finding that the petitioner’s statistical
evidence and analysis “make out a prima facie case of intentional discrimination”
and further finding that “[o]nce a prima facie showing is made out, the burden of
proof [shifts] to the State to dispel the inference of intentional discrimination.”)
(internal quotations and citations omitted). Although this Court has already found
that Woodfox made a prima facie showing, the Court will consider the statistical
evidence put forward by both the State and Woodfox in assessing whether the
State has met its burden.
IV.
As a threshold matter, the Court will first assess the proper baseline to be
used when determining whether the observed racial disparities were statistically
significant. The State argues that the proper baseline is 36.62%, whereas
Woodfox argues that the proper baseline is 40.8%.
The State’s Proposed Baseline of 36.62%
The State’s statistical analysis “was based on a comparison between an
observed figure to an expected figure.” (Doc. 258 at 6). The observed number is
5 out of 27, or 18.5%, of African-American grand jury forepersons selected
between 1980 and March 1993. The State correctly points out that in order to
establish a prima facie case, the petitioner may compare the “proportion of the
6

group in the total population to the proportion called to serve as grand jurors.”
Castaneda, 430 U.S. at 494. Woodfox did just that: he compared the proportion
of African-Americans in the total population, which was 44%, to the proportion of
African-Americans selected as forepersons, which was 18.5%. However, the
State argues in its rebuttal, that the correct figure for assessing whether there
was discrimination should be what the State refers to as the “expected figure,”
instead of the “observed figure.” The State explains that general population
figures distort the analysis because not every member of the general population
is eligible to serve on a grand jury, noting that Louisiana law does not permit
felons or illiterates to serve. See La. Code Crim. Proc. Art. 401.
The State posits that in order to determine whether African-Americans
were underrepresented in the grand jury foreperson selection process, it is
incorrect to compare the percentage of total African-Americans in the parish to
the percentage of African-Americans selected to serve as grand jury forepersons.
Rather, the State argues that it is more accurate to compare the percentage of
total African-Americans in the parish that were eligible to serve to the percentage
of African-Americans that actually served. However, because there is no
recorded evidence showing how many African-Americans were eligible to serve
as grand jury forepersons in West Feliciana Parish during the relevant time
period, the State developed an estimate of eligible African-Americans based on
available data, and concluded that 36.2% is the proper baseline.

7

In reaching the baseline value of 36.62%, the State’s expert, Dr. Solanky,
first looked to West Feliciana voter rolls because the voter registration process
“screens out certain individuals” who would be ineligible to serve as grand jurors,
such as the mentally incompetent and felons. However, as the State points out,
this is not conclusive because “illiterates may vote but are not eligible to serve as
grand jurors,” and notes that Dr. Solanky accounted for illiterates when
estimating eligibility. (Doc. 258 at 8). Dr. Solanky obtained illiteracy data for the
years 1980-85 and 1988-93 via public records. Dr. Solanky utilized the following
numbers to reach its conclusion of a proper baseline of 36.62%.
 1980-85 Illiteracy data available: This data was broken down by race, and
African-Americans constituted between 97.8 and 98% of illiterate voters.
Dr. Solanky determined that the percentage of eligible African-Americans
based on literacy was 34.8% during these years.
 1988-93 Illiteracy data available: This data was not broken down by race
and Dr. Solanky used the data from 1980-85 to estimate what percentage
of illiterate voters from 1988-93 were African-American. He selected 97.8%
(the smaller number from the 1980-85 data set) and concluded that
between 1988-93, 36.47% would have been eligible.
 1986-87 No illiteracy data available: Dr. Solanky used a regression
analysis and concluded that in 1986, there was 615 total illiterates and in
1987, there were 643 total illiterates.

8

 36.62% eligibility: Dr. Solanky combined data from the three time periods
and concluded that the overall percentage of eligible African-Americans
was 36.62%. The State argues that this is a conservative number because
Dr. Solanky assumed that the 167 voters from 1986 to 1993 whose race
was listed as “other” were ineligible, and had they been considered
eligible, this would have decreased the proportion of eligible AfricanAmericans. Dr. Solanky also calculated his figures setting the relevant time
period as 1980 through 1994, instead of this Court’s ordered time period of
1980 through March 1993. Had Dr. Solanky adhered to this time period,
the eligibility would have been slightly lower: 36.58% instead of 36.62%.
Finally, Dr. Solanky did not account for federal illiterates beyond 1985,
which the State claims would have further decreased the eligibility because
federal illiterates tended to be African-American.
Thus, the State argues that 36.62% is the correct baseline to assess whether any
racial disparity was significant.
In Woodfox’s post-hearing brief, he argues that Dr. Solanky’s baseline is
“fundamentally flawed,” and urges the Court to reject it. (Doc. 259 at 14).
Woodfox asserts that the State did not show the premises underlying Dr.
Solanky’s literacy calculation, and the State focused on only one factor that
affected the composition of the jury panel, literacy. Woodfox contends that the
data relied upon by Dr. Solanky is unreliable for several reasons.

9

 1980-85 illiteracy data: This data stems from a document with handwritten
tables, but the State failed to show how literacy was determined. The
current Registrar testified that this document was not hers, and Woodfox
posits that the tables were likely prepared by her predecessor, who was
the registrar during the Civil Rights era and allegedly was “responsible for
the egregious discrimination against blacks in the Parish.” (Doc. 259 at
15). In the State’s reply brief, the State asserts that this is speculation as to
who prepared the data and even if the previous registrar were indeed a
racist, it does not logically follow that he would register African-Americans
to vote in an effort to “artificially inflate the number of black illiterates.”
(Doc. 266 at 15).
 Illiteracy rates in West Feliciana Parish v. East Feliciana Parish: Woodfox
argues that Dr. Solanky testified that the illiteracy rate in West Feliciana
Parish was 9.1% in 1989, but in East Feliciana Parish it was 1.8%.
Woodfox contends that there is no reasonable explanation for this, other
than discrimination by the West Feliciana voter registrar, pointing out that
not only does the State not provide any explanation, but the State
explained that degrees of educational attainment were higher in West
Feliciana than in East Feliciana. In response, the State argues that Dr.
Solanky only testified from a chart that Woodfox provided and that
Woodfox never questioned the current registrar of voters, Bobbie Ross, at
the evidentiary hearing about literacy rate in East Feliciana Parish.
10

Moreover, the State argues that historical data suggests that East
Feliciana Parish underreported its illiterates because the National Center
for Education Statistics (NCES) estimated in 1989 that 16% of adults in
Louisiana were illiterate. However, the data that Woodfox objects to shows
that the illiteracy rate in 1989 in East Feliciana Parish was 1.8%. Moreover,
the State argues that both census and NCES data shows that educational
attainment was higher in West Feliciana Parish during the relevant time
period. The State asserts that Woodfox’s critique on the State’s literacy
estimate based on the 1989 East Feliciana literacy data is “plainly
misguided.” (Doc. 266 at 17).
 Federal illiterate voters: Woodfox points out that Dr. Solanky reduced the
number of literate African-American voters by assuming that voters listed
in the hand-written table with a “fed” notation were illiterate, even though
there was no evidence to support this. Woodfox further points out that
voters with a “fed” notation were not included in the tally of total illiterate
people. In response, the State argues that the West Feliciana Registrar
affirmed that these individuals were, in fact, illiterate. Moreover, even if
they were not illiterate and Dr. Solanky erred in assuming that they were,
this erroneous assumption would not have resulted in a meaningful
difference in Dr. Solanky’s results.
 Incompleteness of data: Woodfox argues that there is only reported data
for literacy by race for 5 of the 14 years that Dr. Solanky considered. For
11

the years that did not have data for literacy by race, Dr. Solanky made
assumptions, which Woodfox contends are unsupported. The State
objects, arguing that Dr. Solanky was able to estimate the number of
African-American illiterates for the years that did not break literacy
statistics down by race based on the overall figures that African-Americans
constituted the majority of illiterates.
 Incompleteness of model: Woodfox points out that Dr. Solanky’s model
only accounts for one factor in determining the composition of the grand
jury, literacy, when there were other available factors to consider. Such
factors include residency requirements, exemptions under state law, and
the fact that many people called for venire did not appear. The State
objects, noting that these other variables are not race-specific, but literacy
is race-specific.
Woodfox’s Proposed Baseline of 40.8%
Woodfox argues that the correct baseline must be calculated from the
“demographics of the people who were actually selected to serve on grand juries,
excluding the forepersons.” (Doc. 259 at 18). Woodfox asserts that between
1980 and 1993, there were 297 non-foreperson grand jurors in West Feliciana
Parish. Out of the 297 jurors, 277 of those jurors’ races were identified, and 113
of them, or 40.8%, were African-American. Therefore, Woodfox contends, this is
the proper baseline for determining whether African-Americans were
underrepresented as grand jury forepersons.
12

Woodfox points out that in Dr. Solanky’s testimony, Dr. Solanky noted that
non-foreperson grand jurors are randomly selected from the available grand jury
pool but that he did not use their demographics in determining the baseline. Dr.
Solanky explained that he did not use their demographics because race
information is missing for 20 of the non-foreperson grand jurors. However, as
Woodfox correctly notes, Dr. Solanky relied on more incomplete data in making
his calculations because he only had data for 5 of the 14 years that went into his
calculations. Woodfox contends that even though data information is missing for
20 of the grand jurors, data about the 277 identifiable grand jurors is the best
evidence available. Moreover, Woodfox points out that the Supreme Court of
Louisiana has stated that “common sense tells us that the group of grand jurors
who actually served is, by virtue of La. Code Crim. Proc. art 413(B), a randomlyselected sample or subset of the total grand jury venire.” Langley, 813 So. 2d at
360-61.
Turning to the State’s objection to Woodfox’s 40.8% baseline, the State
argues that it is misleading because it “fails to account for the effect of variation.”
(Doc. 258 at 15). The State urges the Court to consider that there is a difference
between a sample and a population, and Dr. Solanky’s estimate is based on the
population, whereas Woodfox’s expert, Dr. Marx, based his estimate on a
sample. The State contends that estimates based on samples “inherently have
some variation,” which is why polls always have a reported margin of error. There
are two ways to report the variation in an estimate based on samples: either (1)
13

the sampling error, or margin of error, or (2) confidence intervals. The State
argues that assuming the 277 out of 297 is a random sample, the 40.8%
estimate is subject to a sampling error of +/- 5.8%, meaning that the State’s
eligibility figure of 36.62% is within the margin of error. Additionally, the “95%
confidence interval for the target population of all seated non-foreperson jurors is
35% to 46.8%,” so the State’s 36.62% baseline is within the confidence interval.
The State argues that Dr. Marx’s failure to consider these variations undermines
the validity of his calculations.
Moreover, the State argues that comparing a sample to another sample
skews the results because of the inherent margin of error within the sample.
Because Dr. Marx compared a sample (5 out of 27 grand jury forepersons were
African-American) to another sample (113 out of 277 seated grand jury members
were African-American), the State argues that Woodfox’s results are
meaningless.
In response, Woodfox argues that the State’s arguments against the
40.8% are baseless. First, Woodfox objects to the State’s argument that 40.8% is
derived from a non-random sample. Woodfox asserts that this figure is a random
sample because the seated grand jurors do “reflect a random sample of a subset
of the population, i.e. registered voters in West Feliciana Parish who are qualified
to sit on a grand jury.” (Doc. 267 at 16). Woodfox asserts that this number
accounts for all of the variables that would affect the composition of the group,
such as who appeared or took an exemption.
14

Woodfox also objects to the State’s argument that Dr. Marx did not account
for the margin of error in his sample. First, Woodfox argues that the State
withheld this argument until after the hearing, not giving Woodfox the opportunity
to have his expert testify about the importance of variation or give Woodfox the
opportunity to examine Dr. Solanky about the margin of error associated with the
36.2% figure. Woodfox also asserts that it is incorrect to assume that Dr.
Solanky’s estimate has no variability and that it is incorrect to assume that any
number that is below 40.8% “and is within the margin of error rate . . . can just as
readily be relied on as baseline data.” (Id. at 17). Woodfox argues that the State
wants the Court to “rely on a number on the low end of the margin of error band,”
when the State has no reason for “such favorable treatment.” (Id. at 18).
Woodfox asserts that because the State has the evidentiary burden, the State
must show that “at least at a majority of the confidence interval, there is statistical
insignificance,” which the State cannot do. (Id.).
The Appropriate Baseline is 40.8%
After careful consideration of the parties’ arguments, the Court finds that
Woodfox’s baseline is appropriate. The Court finds that the State’s arguments
are similar to the ones considered, and rejected, in State v. Langley. In Langley,
the district court rejected the State’s argument that the defendant “should have
compared the percentage of black and female forepersons selected only to the
percentage of blacks and females in the population eligible to sit on the grand
jury, i.e., eliminating from the gross population figures those persons not eligible
15

to serve as grand jurors under La.Code Crim. Proc. art. 408.” Langley, 813 So.
2d at 364. On appeal, the Louisiana Supreme Court explained that:
This pre-qualification of venire members, according to
neutral criteria, does arguably suggest that eligible
population statistics might generally be more precise
than gross population statistics in calculating the degree
of under-representation of minorities in the grand jury
venire, for example. In this vein, our courts of appeal
have generally required that eligible population statistics
be presented to prove jury discrimination claims.
Id. at 369. For example, the court cited a Louisiana appellate court, State v.
Young, in which “the racial and gender composition of the general or grand jury
venire was not available.” Id. at 369. In State v. Young, 31.3% of the parish was
African-American and the parties stipulated that 6.6% of grand jury forepersons
were African-American. State v. Young, 569 So. 2d 570, 575 (La. Ct. App. 1990).
In finding that the defendant failed to make a prima facie showing, the court
explained that the defendant “failed to show the percentage of minority persons
in the general or grand jury venires, or the percentage of qualified minority
persons in the general population.” Id. at 576.
However, the Louisiana Supreme Court in Langley distinguished the
Appellate Court’s reasoning in Young, noting that Langley “presents a unique set
of facts.” Langley, 813 So. 2d at 369. In Langley, the defendant presented
evidence of “the racial and gender composition of the gross or general
population” for the relevant time period, as well as evidence of “the racial
composition of voter registration rolls . . . and the racial and gender composition
16

of the grand jurors who actually served” during the relevant time period. Id.
During the relevant time period, the percentage of African-Americans in the
parish “ranged from 21.7% to 23.2%” and the percentage of seated grand juror
African-Americans was 22.95. Id. at n. 17. The court explained that “common
sense tells us that the group of grand jurors who actually served is, by virtue of
La. Code Crim. Proc. art. 413(B), a randomly-selected sample or subset of the
total grand jury venire.” Id. at 369-70. Moreover, because the percentages of
African-Americans in the general population and the percentages of persons who
actually served were “all statistically nearly identical,” the court concluded that
“the percentages of those classes in either the eligible population or the total
grand jury venire for the relevant time period would not be significantly
statistically different.” Id. at 370.
Here, the Court finds that the 277 seated grand jurors was representative
of the entire eligible population. While the State argues that literacy should be
taken into account to reduce the proportion of eligible African-Americans in the
general population, common sense leads the Court to believe that literacy was a
factor in selecting the seated grand jury members. Illiterates are ineligible to
serve as grand jurors and thus, logic leads this Court to presume that all of the
seated grand jurors were literate. The process of selecting seated grand jurors
accounts for all of the possible variables that determine eligibility. Thus, 113 of
the African-American seated grand jurors were all eligible to be selected as
grand jury forepersons. It is the State’s burden to show that “the pattern of
17

underrepresentation proved . . . was the result of a ‘racially neutral selection
procedure.’” Guice, 722 F.2d at 280 (citing Alexander, 405 U.S. at 632). Here,
the State is attempting to show that any underrepresentation was the product of
chance and to do that, the State has altered the numbers to reduce the baseline
of eligible African-Americans. This argument might work if Woodfox were unable
to show how many seated grand jurors were African-American, and instead,
relied on general population figures. If Woodfox relied on general population
figures to show that the percentage of African-American grand jury forepersons
was under representative of the percentage of African-Americans in the parish as
a whole, then the State’s argument about determining who was actually eligible
to serve might be more persuasive.
However, this is not the case here. The State’s burden is to show that
there was some racially neutral reason as to why only 5 out of the 27 grand jury
forepersons were African-American. The judges did not select the grand jury
foreperson from the population as a whole. Rather, the judges selected the
foreperson from the eligible seated grand jurors who were before them. The
State’s argument about the appropriate baseline does not help the State meet its
burden.
Once the appropriate baseline is established, it is necessary to determine
whether the disparity is statistically significant. In other words, is the disparity so
stark that it cannot be explained by chance? In order to determine this,
hypothesis testing is utilized and the statistician formulates a null hypothesis. A
18

null hypothesis is an expected number against which the observed number can
be tested. David H. Kaye & David A. Freedman, Reference Guide on Statistics,
Reference Manual on Scientific Evidence, 249 (Federal Judicial Center 2011).
Here, the null hypothesis is 40.8% and the observed number is 5 out of 27, or
18.5%. Hypothesis testing compares the disparity between 40.8% and 18.5%,
and the result is the likelihood of obtaining a disparity this great, or greater. See
id. The likelihood, or probability, is called the P-value. If the P-value is small, then
this “discredits the null hypothesis,” and demonstrates that the difference is
statistically significant. Id. at 251. There are two P-value significance levels: 5%
and 1%. Id.
When testing the null hypothesis, statisticians will either use a one-tailed or
two-tailed test. The parties agree that for a two-tailed test, a P-value that is
greater than 5% is not significant. (Doc. 258 at 11 & Doc. 259 at 20). The parties
disagree about the significance level required for a one-tailed test. The State
argues that a P-value for a one-tailed test is weak evidence, and that a P-value
of .025 is more appropriate. (Doc. 258 at 11). See Reference Manual on
Scientific Evidence at 256 n. 110 (“One-tailed test at the 5% level are viewed as
weak evidence – no weaker standard is commonly used in the technical
literature.”) The parties also disagree about which test is more appropriate to
establish underrepresentation. Woodfox maintains that a one-tailed test
examines either underrepresentation or overrepresentation, whereas a two-tailed

19

test examines both underrepresentation and overrepresentation. (Doc. 259 at
22).
The Court does not find that it is necessary to determine which test is
appropriate, nor to determine which P-value (.025 or .05 ) is appropriate for a
one-tailed test. Using the baseline of 40.8%, the one-sided P-value is .0126 and
the two-sided P-value is .0185. (Doc. 259 at 20). Therefore, because the P-value
is below both of the thresholds (.025 and .05), the Court finds that regardless of
the test used, the disparity is significant.
V.
Because Woodfox has made a prima facie case, the State must prove
“objective, racially neutral criteria were used in the selection process.” Johnson v.
Puckett, 929 F.2d 1067, 1072 (5th Cir. 1991). Until 1999, Louisiana state law
mandated that the judge presiding over the grand jury would select the
foreperson. See La. Code Crim. Proc. art 413(B), amended by Acts 1999, No.
984. In West Feliciana Parish, there were two judges who appointed forepersons:
the Honorable William Kline and the late Honorable Wilson Ramshur, the judge
who appointed the grand jury foreperson who presided over Woodfox’s
indictment. Thus, it follows that the State must prove that the judges utilized
objective, racially neutral criteria to rebut Woodfox’s prima facie case.
The State asserts that judges in West Feliciana Parish used race neutral
criteria, such as employment, education, character, and independence, when
selecting forepersons. Judge Kline testified that he would often determine who he
20

thought would be a good foreperson and he would attempt to contact them
before the morning of the venire. (Tr. 2, 49). However, if Judge Kline did not
know someone who was in the venire, he “sought facts about the person, rather
than an opinion, as to whether he or she would be a good foreperson.” (Tr. 2,
50).
Judge Kline stated that character was one of the first factors he considered
when selecting a foreperson. (Tr. 2, 50). Judge Kline also stated that
communication skills, patience, independence, reputation and education were
also important factors. (Tr. 2, 51-52). He explained that while employment and
education were important, they were not the sole factors by which he selected a
foreperson. (Tr. 2, 52). Judge Kline looked for “basic education” and he looked
for employment because it “reflected some dependability.” (Tr. 2, 52). However,
he did not want to only appoint people “who had master’s degrees or advanced
degrees” because that would “eliminate consideration of a whole body of good
folks with good common sense.” (Tr. 2, 52). Judge Kline noted that he made a
conscious effort to include women and African-American citizens in order to be
inclusive and appointed an African-American female foreperson and a craftsman
who did not have as much education as others in the pool but was
“representative of the community.” (Tr. 2, 53-54).
Because Judge Ramshur passed away in 2006, the State presented
evidence of other officials who were familiar with the grand jury foreperson
selection process. See Johnson, 929 F.2d at 1073. The Honorable George H.
21

Ware, Jr., who was the District Attorney in West Feliciana from 1985 through
1996, and Jesse Means, the Assistant District Attorney for the 20th Judicial
District Court from 1985 through 2006, testified about the grand jury forepersons
selection process. Ware testified that he would meet with the judges when they
were in the process of selecting forepersons and discuss the potential
forepersons. (Tr. 1, 160). He testified that the questions the judges asked
suggested
that they were seeking information about a community
leadership role, responsibility in the community,
background, whether or not this person was a gossip,
so to speak, because in state court the grand jury
proceedings are in secrecy, and so you don’t want
someone on the grand jury, and particularly the
foreman, discussing in the community at large what was
going on in the grand jury. Occasionally, the subject of
job, do you know where this person works, a question
about the person’s family.

(Tr. 1, 160).
Means testified that while Judge Ramshur never asked him specifically for
advice, on one occasion, Means advised him not to select a person that Means
believed “would not understand the case and would not maintain the requirement
of secrecy.” (Tr. 3, 41). However, Means did not give “specific advice for specific
people” because Judge Ramshur “didn’t need to be advised.”(Tr. 3, 45).
The State asserts that character, education, employment, independence,
and reputation are all race-neutral factors that the judges used, and these factors
have been approved by other courts when considering whether a state has
22

rebutted the prima facie case. In United States v. Perez-Hernandez, the Eleventh
Circuit approved of the guidelines used by the judges, which “generally consisted
of four separate factors: (1) occupation and work history; (2) leadership and
management experiences; (3) length of time in the community; and (4)
attentiveness during the jury empanelment.” United States v. Perez-Hernandez,
672 F.2d 1380, 1387 (11th Cir. 1982).
The State argues that the procedure for selecting a grand jury foreperson
was not random and was never designed to be random. (Doc. 258). The State
points out that it demonstrated that the race neutral criteria, such as education
and employment, were not distributed proportionally among the African-American
and Caucasian population. Because evidence showed that there was a large
disparity in education and employment levels, the State contends that this
explains the “higher percentage of whites” appointed because “whites were more
educated and employed in West Feliciana in the 1980s and early 1990s.” (Doc.
258 at 26). The State asserts that despite the disparity, evidence shows that
Judge Kline attempted to include African-Americans when selecting grand jury
forepersons.
Moreover, the State argues that the empirical data for the 27 forepersons
selected between 1980 and 1993 shows that Judges Kline and Ramshur “took
education and employment into consideration” and “acted affirmatively to include
blacks with less education.” (Doc. 258 at 29). 22 of the 27 forepersons selected
were Caucasian and 5 were African-American. By stipulation of the parties, 24 of
23

the selected forepersons had at least a high school diploma, and 3 did not. The 3
selected forepersons who did not have a high school diploma were AfricanAmerican. The State contends that the disparity between Caucasian and AfricanAmerican forepersons can be explained by the “fact that the judges were
applying race-neutral criteria – except to the extent the judges were affirmatively
attempting to include blacks among their selection.” (Doc. 258 at 30).
Woodfox asserts that the State’s argument that the judges in West
Feliciana relied on race neutral criteria is insufficient to dispel the inference of
discrimination as established by the prima facie case. (Doc. 267 at 20). Woodfox
suggests that the testimony of Judge Kline and Ware amount to “affirmations of
good faith in making individual selections,” which “are insufficient to dispel a
prima facie case of systematic exclusion.” Alexander, 405 U.S. at 632. Moreover,
Woodfox argues that the law requires that the State show “evidence that
objective, racially neutral criteria were used in the selection process,” and the
criteria that the State showed was not objective, but subjective. Johnson, 929
F.2d at 1072. For example, Woodfox contends that reputation, standing in the
community and character are subjective criteria and argues that these are
“unlikely to be applied race-neutrally.” (Doc. 267 at 22).
Woodfox further asserts that the State has not shown any evidence,
except for testimony by Judge Kline and Ware, that the judges applied the
selection criteria in a race neutral fashion.. Woodfox argues that there is no
evidence demonstrating that Judge Ramshur asked questions of the venire
24

assessing the members’ educational and employment background. (Doc. 259 at
9).
Woodfox points to Ware’s testimony concerning the questions asked of
him by Judge Ramshur. Ware testified that Judge Rahmshur would ask him more
frequently about potential forepersons in East Feliciana Parish, but “very rarely
for West Feliciana . . . because Judge Ramshur lived in West Feliciana . . . [and]
was much more familiar with the people in West Feliciana.” (Tr. 1, 158). Woodfox
argues that this factual scenario is similar to the one in Guice, where the judge
testified that in “the other two parishes in his jurisdiction he often made inquiries
as to the qualifications of the members of the grand jury venire.” Guice, 722 F.2d
at 278. However, the judge “[a]pparently felt no need to do so in Madison Parish,
where he resided and where he was often acquainted with members of the
venire.” Id. Woodfox argues that here, as in Guice, “[n]o evidence was presented
of any systematic attempt to obtain objective information about the qualifications
of venire members” in the Parish. Id. Woodfox contends that the State did not
show that Judge Ramshur used objective criteria, nor did the State present
testimony from anyone “to provide information on the actual operation of the
selection process.” Castaneda, 430 U.S. at 491.
VI.
The State must rebut the inference of discrimination by showing that “the
pattern of underrepresentation proved . . . was the result of a ‘racially neutral
selection procedure.’” Guice, 722 F.2d at 280 (citing Alexander, 405 U.S. at 632).
25

The prima facie case will not be rebutted if the State fails to put on any evidence
about the selection process. See Castaneda, 430 U.S. at 498 (noting that “the
State introduced practically no evidence” in their rebuttal). The prima facie case
will also not be rebutted if the State puts on evidence that amounts to good faith
affirmations, which are “insufficient to dispel a prima facie case of systematic
exclusion.” Alexander, 405 U.S. at 632. In Johnson v. Puckett, the Fifth Circuit
found that testimony which “merely indicates that the judges in Panola County
never stated or indicated to the circuit clerk that they selected grand jury foremen
based on their race” was insufficient to dispel the inference of discrimination.
Johnson, 929 F.2d at 1073. The court noted that “the testimony neither denies
the use of racial criteria nor advances any other objective non-discriminatory
criteria used by the judges.” Id.
Similarly, in Crandell v. Cain, the Western District of Louisiana found that
while the judge affirmed “that he did not intend to discriminate on the basis of
race,” there was “no evidence of objective criteria known in advance of the
selection. Crandell v. Cain, 421 F.Supp. 2d 928, 943 (W.D. La. 2004). Judge
Kitchens, the appointing judge, “would be provided a list of the names and
addresses of the 35 persons on the venire,” which did not include the venire
members’ races. Id. at 937. Judge Kitchens “would review the list to see if he
recognized someone who would make a good foreman.” Id. Because he was not
from the parish, he did not know many of the citizens and would review the list
with the Clerk of Court, Ms. Wilna Mabry, a “longtime political figure in Bossier
26

Parish [who] knew many of its citizens.” Id. Judge Kitchens testified that he
looked “for a ‘responsible person’ with a measure of leadership skills and
education.” Id. He also examined the person’s occupation. He obtained the
majority of this information from Ms. Mabry, and if neither she nor he knew a
member of the venire, that person would not be considered as a potential
foreperson. Id. After he selected a person that he thought would be a good
foreperson, he would meet with him or her in his office to ask if he or she could
serve. Id. The court found that the judge’s testimony amounted to an “affirmation
of good faith,” which was insufficient to rebut the presumption of discrimination.
Thus, prima facie cases are not rebutted when the testimony is lacking as to the
selection process or when the testimony is an “affirmation of good faith.”
In Guice v. Fortenberry, the Fifth Circuit noted that the testimony “revealed
that no objective criteria were used in his selection of grand jury foremen; rather,
he selected individuals, always white, who were known to him.” Guice, 722 F.2d
at 281. The testifying judge explained that he selected forepersons based on “his
personal knowledge of the qualifications of potential foremen.” Id. at 278. The
Fifth Circuit noted that while the judge would ask about the “qualifications of the
members of the grand jury venire” for two of the three parishes in his jurisdiction,
he did not ask questions about the venire in the parish in which he lived. Id. “No
evidence was presented of any systematic attempt to obtain objective information
about the qualifications of venire members in Madison Parish,” where the judge

27

lived. Id. The judge explained that “when we have a Black Judge, they’re gonna
know who’s the most qualified amongst the Blacks to be a foreman.” Id. at 281.
Prima facie cases are rebutted when the State puts forth affirmative
evidence showing that objective and racially-neutral criteria are used. For
example, in United States v. Breland, the Northern District of Georgia found that
the testimony of eleven district judges indicated that objective and racially-neutral
criteria were used. United States v. Breland, 522 F.Supp. 468, 480 (N.D. Ga.
1981). The judges testified that they “generally relied solely on information
provided in the juror questionnaires and made no independent investigation of
the qualifications of persons drawn as grand jurors.” Id. at 471.The judges
considered age, education, occupation, and leadership experience. Id. at 471-74.
The court explained that “the predominant criteria of occupation, education, and
age, were standards not only objective but clearly relevant to the specific tasks
required by Rule 6(c), F.R.Crim.P., to be performed by forepersons and deputies
and as visualized by the judges.” Id. at 480. The court further explained that
“most important, the substantial evidence is that the stated criteria were in fact
faithfully applied.” Id. The record indicated that the grand jurors who were
appointed as forepersons “actually possessed the qualities which the judges
were seeking, i.e., they had in large part responsible employment positions
and/or significant education, and were of mature age.” Id. The court noted that
the defendants in the case did not “challenge the sufficiency of the qualifications
of the persons appointed but maintain only that others on each grand jury might
28

have been no less qualified.” Id. Finally, the court rejected the argument that
“judges were required to elicit additional information from, or about, randomlydrawn grand jurors to seek out for appointment blacks and women who might be
‘qualified’ as forepersons but whose jury questionnaires did not affirmatively
indicate such ‘qualification.’” Id. (explaining that the Constitution “did not require
the appointing judges to accord preferential treatment to any group but imposed
upon them no more than the obligation to use nondiscriminatory, neutral criteria
in selecting persons deemed adequate to perform the role of foreperson.” Id.).
Similarly, in United States v. Perez-Hernandez, the Eleventh Circuit found
that the State had rebutted the prima facie case. Perez-Hernandez, 672 F.2d at
1388. Eight district judges testified about their selection processes, and while
each judge testified “that he acted independently,” the judges “employed similar
guidelines in making a selection.” Id. at 1387. The guidelines typically consisted
of these four factors: “(1) occupation and work history; (2) leadership and
management experiences; (3) length of time in the community; and (4)
attentiveness during the jury empanelment.” Id. The court explained that these
criteria were not “arbitrary and unrelated [to the duties of a grand jury
foreperson].” Id. However, the court did note that if the record had shown “that
these judges abused their discretion by selecting foremen without regard to the
stated criteria or by excluding equally qualified blacks and women, then we would
conclude that the presumption established by the prima facie test was not
rebutted.” Id. at 1388. “This kind of evidence would establish that the gender and
29

racially neutral selection procedures were not in fact applied in a neutral
manner.” Id.
VI.
Here, the Court finds that the State has failed to rebut the prima facie case.
The record indicates that the only information that the judges received prior to
selecting the panel was a list that contained the names, addresses, and in later
years, the telephone numbers of the members of the venire. Pet’r. Ex. 13. Judge
Kline testified that “we normally would have in mind somebody . . . to be foreman
of the grand jury.” (Tr. 2, 49). Judge Kline explained that he would contact the
person that he had in mind ahead of time. If he did not know anybody who was in
the list of the venire, he would attempt to seek facts about that person. (Tr. 2,
50). However, like the venire list in Crandall, there was no evidence of objective
criteria that was available to the judges in advance. The record does not indicate
that the judges had education or employment data prior to making the decision.
While Judge Kline testified that he did seek out facts about people that he did not
know, he stated that normally, he would select the foreperson in advance and
contact him or her. “That’s how that process worked.” (Tr. 2, 49).
The State has argued extensively that the empirical data shows that the
judges did take education and employment into consideration when making their
selections. However, the evidence shows that the only objective criteria that was
known in advance to the judges was a name and an address. There is no
evidence that education and employment data was actually provided to the
30

judges prior to making their selection. The parties stipulated to the State’s Exhibit
4, which lists the education and employment data of the selected grand jury
forepersons. However, this information was compiled based on affidavits and
transcripts, not based on what the judges had in front of them. For example, in
United States v. Breland, the judges made their selections based on information
contained in juror questionnaires. Breland, 522 F.Supp. at 471. While the
questionnaires did list the venire members’ race and sex, the questionnaires also
listed race-neutral criteria, such as education and employment. The judges
testified that they looked to race-neutral criteria on the questionnaire to make
their decisions. Here, there is no evidence that there were any forms or
questionnaires that outlined race-neutral criteria.
The State argued that “West Feliciana is not Orleans Parish or a federal
district encompassing a large portion of the state – it is a small rural parish where
people know most everyone else.” (Doc. 266 at 10) The State posited that
“Judges Kline and Ramshur were well positioned to select qualified forepersons
of any race based on their personal knowledge about their neighbors.” (Doc. 266
at 11). While it may be correct to say that West Feliciana Parish is a small, rural
Parish where everyone knows everyone else, this is insufficient to rebut the
prima facie case. In Crandall, while the appointing judge did not live in the Parish
and did not know many of its citizens, his clerk of court did know many of the
citizens and advised him about the members’ education, leadership skills, and
occupation. Crandall, 421 F. Supp.2d at 937. However, the court, while finding
31

that his “affirmation that he did not intend to discriminate” was “credible,” it was
not enough to rebut the presumption of discrimination.
Here, the Court finds that while Judges Ramshur and Kline may very well
have known the majority of the West Feliciana Parish citizens, much like how Ms.
Mabry knew the majority of the citizens in Crandall, this does not relieve the
State of its obligation to put forth evidence that objective, race-neutral criteria
was used. The evidence and testimony put forth by the State is an affirmation of
good faith, which the jurisprudence has made clear is insufficient to rebut a prima
facie case.
VI.
As a final matter, the Court will address the State’s arguments and
concerns regarding the three stages of evidence presented in this case as well
as evidence presented concerning the individual selections of Judge Ramshur.
The State objects to this Court’s May 11, 2012 order, in which this Court ordered
that the evidentiary hearing would consist of three stages. First, Woodfox would
present his prima facie case, then the State would rebut by showing evidence of
race neutrality, and finally, Woodfox would have the opportunity to rebut the
State’s showing. (Doc. 229). The State also objects to this Court’s order
permitting Woodfox to present evidence concerning Judge Ramshur’s individual
selections.
The State argues that Castaneda mandates a two-stage framework and
that this Court has improperly introduced a third-step by permitting Woodfox to
32

present evidence rebutting the State’s rebuttal argument. This Court’s previous
order explained that this third stage “comports with general notions of burdenshifting motive analysis.” (Doc. 229 at 1) (citing Miller-El v. Dretke, 545 U.S. 231,
239-40). The State objects, noting that Miller-El was a burden-shifting analysis
for examining the discriminatory use of peremptory challenges under Batson v.
Kentuckyk, 476 U.S. 79 (1986), and is not applicable here. The State argues that
unlike Batson, Castaneda examines the actions of the judges who appoint grand
jury forepersons over a “significant period of time.” Castaneda, 430 U.S. at 494.
(Doc. 258). Additionally, Castaneda does not require the State to show a
plausible reason for its actions. Instead, the State is required to show the court
that “racially neutral selection criteria and procedure have produced the
monochromatic result.” Id.
The State further argues that it is inappropriate to consider any evidence
that is specific to Judge Ramshur. The State asserts that Castaneda focuses on
whether or not the procedure was allegedly tainted by discrimination, and not
whether any particular judge had discriminatory intent. The State notes that in
rebuttal arguments, testimony from other judges other than the actual appointing
judge is often used to explain the procedures, suggesting that the selection
practices of both Judges Kline and Ramshur are equally important.
Woodfox argues that the State is confusing the distinct stages of the
litigation. Woodfox points to Guice, in which the Fifth Circuit stated that “it has
been our task to determine whether in fact the petitioners were indicted by a
33

grand jury whose foreman was chosen in a discriminatory manner,” and the Fifth
Circuit did consider the selection procedures of the actual appointing judge.
Guice, 722 F.2d at 281-82.
The Court does not find it necessary to revisit these arguments. The Court
explained in a previous order that “[i]f and when the State’s production rebuts the
prima facie case, the presumption of discrimination appearing from the statistical
prima facie case disappears, and at that point Woodfox is entitled to introduce
evidence to meet his ultimate burden of persuading the Court as to Judge
Ramshur’s discriminatory intent in making foreperson selections.” (Doc. 237 at
6). Because the Court finds that the State failed to rebut the prima facie case,
any further evidence is immaterial and will not be considered.

VII.
Accordingly, Woodfox’s habeas relief is GRANTED.

Signed in Baton Rouge, Louisiana on February 26th, 2013.





JAMES J. BRADY, DISTRICT JUDGE
MIDDLE DISTRICT OF LOUISIANA

34