United States District Court, Middle District of Pennsylvania
April 8, 2002
RICHARD VIETH, NORMA JEAN VIETH, AND SUSAN FUREY, PLAINTIFFS
COMMONWEALTH OF PENNSYLVANIA, ET AL., DEFENDANTS
Before: Yohn, District Judge, Rambo, District Judge, Nygaard, Circuit
The opinion of the court was delivered by: Per Curiam:
OPINION AND ORDER OF THE COURT
In our previous order of February 22, 2002, we dismissed several claims
brought by the Plaintiffs which challenged the constitutionality of
Pennsylvania Senate Bill 1200 (Act 1), the congressional redistricting
plan enacted by the General Assembly and signed into law on January 7,
2002. On March 11-12, 2002, we held a hearing
on the Plaintiff's sole
remaining claim: Act 1 violates the constitutional principle of
"one-person, one vote." Upon conclusion of this hearing, we ordered the
parties to file proposed findings of fact and conclusions of law.
The 2000 Census reported that Pennsylvania's total population was
12,291,054 persons. As a result of this population figure, and in
accordance with Article I, Section 2 and the Fourteenth Amendment to the
United States Constitution, Pennsylvania's congressional delegation was
reduced from twenty-one to nineteen representatives in Congress.*fn1
The loss of two congressional seats made it necessary to increase the
size of the existing districts by approximately 100,000 persons.
Additionally, because of shifts in population, some congressional
districts were over-populated while some were well under the ideal
population. Therefore, significant changes were made to the existing
Pennsylvania Senate Bill 1200 was introduced by Senators Brightbill and
Lemmond on November 16, 2001. Senators Brightbill and Lemmond are
members of the Republican Party. A competing plan, Senate Bill 1241 was
introduced by Democratic state senators Mellow, O'Pake, Wagner, Musto,
Kasunic, Fumo and Stout. On December 10, 2001, the Senate considered
amendments to Senate Bill 1200. A Republican amendment was ultimately
agreed to. The version of Senate Bill 1200 that eventually passed the
Senate contained a population deviation of twenty-four persons.
After an amendment by the Pennsylvania House of Representatives, that
chamber passed Senate Bill 1200 on December 12, 2001. The version of
Senate Bill 1200 that passed the House had a total population deviation
of nineteen persons.*fn2 It also maintained two minority-majority
districts in the Philadelphia area, created one open seat in the
southeastern part of the Commonwealth and paired two Democratic
incumbents in the same district. One Democratic incumbent was paired
against one Republican incumbent.
The Senate refused to concur in the amendments to Senate Bill 1200
offered by the House of Representatives. A Conference Committee was
appointed and a plan was eventually devised that contained a nineteen
person deviation. This Conference Committee Report on Senate Bill 1200
was passed by the Senate on January 3, 2002 and by the House of
Representatives later the same day. It was signed into law on January
In its final form, Act 1 created nineteen congressional districts with
a total population deviation of nineteen. Although Act 1 did maintain
two "minority-majority" districts in Philadelphia, it created three
districts in which six incumbent congressman would have to run for
re-election against each other. Of these pairings, two of the districts
required Democratic congressmen to run against each other and one
district required an incumbent Democratic representative to run against
an incumbent Republican representative. These pairings of
congressman permitted the creation of an open congressional district in
the southeastern part of the Commonwealth. In contrast to its treatment
of Democratic incumbents, no Republican congressmen are forced to run
against each other.
Moreover, Act 1 splits eighty-four local governments, including
twenty-five counties, fifty-nine cities, boroughs or townships, as well
as forty-one wards. It also splits six voting precincts.
A. Population Deviation
The United States Constitution requires that each congressional
district in a state contain equal population. See Wesberry v. Sanders,
376 U.S. 1, 18, 84 S.Ct. 526 (1864) (holding that Art. I, § 2 of the
Constitution requires that "as nearly as is practicable one man's vote in
a congressional election is to be worth as much as another's."). The
Supreme Court has been exceedingly clear in requiring lower courts to
balance population among the districts with precision. See Karcher v.
Daggett, 462 U.S. 725, 734, 103 S.Ct. 2653 (1983) ("there are no de
minimis population variations, which could practicably be avoided, but
which nonetheless meet the standard of Art. I, § 2 without
justification.'); Kirkpatrick v. Preisler, 394 U.S. 526, 531, 89 S.Ct.
1225 (1969) ("[T]he `as nearly as practicable' standard requires that the
State make a good-faith effort to achieve precise mathematical equality.
Unless population variances among congressional districts are shown to
have resulted despite such effort, the State must justify each variance,
no matter how small."). In a challenge to a congressional redistricting
plan, the plaintiff bears the burden of proving that the differences in
district-to-district population could have been reduced or eliminated
altogether by a "good-faith effort to draw districts of equal
population." Karcher, 462 U.S. at 730.
We find that the Plaintiffs have met their burden in this case.
First, it is undisputed that Act 1 has a deviation in population of
nineteen persons between the most populated and least populated
districts. Therefore, unless these deviations were unavoidable or
resulted despite a good faith effort to draw districts of equal
population, then the Defendants must prove a legitimate justification for
the deviations. The evidence conclusively demonstrates that this
population deviation was avoidable. Plaintiffs presented into evidence
"Alternative Plan 4" which had a minimum possible deviation —
districts that differ by only one person. Additionally, this plan
achieves zero deviation without splitting any precincts among
congressional districts and splits fewer municipalities than Act 1.
Witnesses presented by both parties testified that it would have been
relatively easy to take any plan or map and reach a point of zero
population deviation.*fn3 Indeed, the Defendants
themselves submitted a
map with zero population deviation. Thus, the nineteen person deviation
in Act 1 was avoidable.
Nor can the deviation contained in Act 1 be excused based upon the good
faith of the Defendants. Indeed, we find that the testimony of defense
witness Dr. John Memmi satisfies the Plaintiff's burden here. Dr. Memmi
testified that he was in charge of drawing up the map that became Act 1.
He further stated that he drew the map under the direct supervision of
the Republican leadership. He began manipulating the map in order to get
the deviations lower. However, once he reached the point of a
nineteen-person deviation, he was told to stop manipulating the map:
Q: Actually, what I would like to know is if there was a
number that anybody said you could stop once you
got to number nineteen or number 18?
A: Every move that was made was done under the
supervision of my supervisors. And at the time,
the team involved the House Republican Caucus
Redistricting crew, some members thereof, the
Senate Republican Caucus Redistricting team,
members thereof. And the Supervisors sat there
while Bill Shower and I — Bill is a gentleman
with the House Republican Caucus. They observed
every census block move that we made. And we
continued to make moves, and were monitoring the
overall range. I should say we all monitored it,
of course, but the supervisor's the one whose
monitoring counted. And when the districts were at
nineteen overall range, they said that is
Tr. Vol 3 at page 276. No other witnesses were called by the Defendants
to explain why Dr. Memmi was told to stop manipulating the map.
We find this testimony to be in clear violation of the Supreme Court's
express goal of getting as close as possible to zero deviation when
drawing congressional districts. See Karcher, 462 U.S. at 734, 103
S.Ct. at 2660. Therefore, we find that the Defendants did not put forth
a good-faith effort to draw districts of equal population.
It has been suggested that the deviation from absolute numerical
equality present in Act 1 is too trivial or minute to rise to a
constitutionally significant level. While it is true that the deviation
contained in Act 1 is small, Karcher specifically holds that "there are
no de minimis variations which could practically be avoided, but
nonetheless meet the standard of Art. I, § 2 without justification."
462 U.S. at 734, 103 S.Ct. 2660. Thus, we hold that the population
deviation in Act 1, even though relatively small, enables the Plaintiffs
to satisfy Karcher's first prong and shifts to the Defendants the burden
of proving justification.
Because the Plaintiffs have met their burden under Karcher, the
Defendants must now prove that this population deviation was "necessary
to achieve some legitimate goal." 462 U.S. at 731, 103 S.Ct. at 2658
(citing Kirkpatrick, 394 U.S. at 530, 89 S.Ct. at 1228; Swann v. Adams,
385 U.S. 440, 443-444, 87 S.Ct. 569, 571-72 (1967)). While Karcher did
not specify how the State might meet its burden in step two, it did give
several examples of legislative policies that might justify some variance
among the populations of the State's various congressional districts.
Specifically, the Court stated that "making districts compact, respecting
municipal boundaries, preserving the cores of prior districts and
avoiding contests between incumbent Representatives" may legitimate minor
population deviations among congressional districts. The Court also made
clear that the burden borne by the State varies inversely with the
magnitude of the population deviation. Id. at 741, 103 S.Ct. 2664. That
is, the greater the deviation, the more compelling the government's
justification must be.
Defendants contend that the population variances in Act 1 are justified
by a desire to avoid splitting voting precincts.*fn4
Both parties agree
that splitting precincts creates additional costs and work for county
election officials in acquiring voting machines, in customizing ballots,
in training precincts officials, in registering voters and in counting
ballots. Moreover, we recognize that split precincts increase the
potential for voter disorientation and candidate confusion. Therefore, to
the extent that such justification is genuine, we acknowledge that the
desire to avoid splitting precincts is a legitimate state interest which
could justify a nineteen person deviation.
We find, however, that the Defendants' arguments on this point are a
mere pretext. Specifically, the evidence has demonstrated that it is
possible to draw a congressional district map with zero deviation and no
precinct splits. Plaintiffs' witness, Robert Priest, drew exactly such a
map using the same tools as Dr. Memmi. However, unlike Dr. Memmi, Mr.
Priest was free to draw the map by manipulating entire precincts from
district-to-district in order to achieve absolute population equality.
As we have previously pointed out, Dr. Memmi testified that he was more
constrained by his superiors. According to that testimony, Dr. Memmi
initially formulated three congressional districts of ideal population by
manipulating entire precincts. However, at that point, he was instructed
to stop using whole precincts and to begin using census blocks to lower
the deviation. Because every voting precinct contains several census
blocks, manipulation of the census blocks leads almost inextricably to
splitting precincts. We also note that none of the congressional
districts that have split precincts have an ideal population.
Dr. Memmi was equally restrained in his efforts to bring Act 1 down to
a zero deviation. That is, his superiors in the Republican caucus
instructed him that he was only to manipulate census blocks and not
precincts. This wholesale reliance on the census block method — the
lone cause of the precinct splits — is what Defendants now employ
to justify the deviations in their map. The logical inconsistency is so
deep that it causes us to pause and consider the sincerity of such
Defendants truly wanted to avoid splitting precincts, they
would have done so by enacting a zero deviation map that did not split
Moreover, it is worth noting that of the maps presented at trial, Act 1
is that which least comports with the neutral legislative policies that
the Karcher Court stated would justify a Congressional redistricting plan
with some deviations. Act 1 is the plan which contains the least compact
districts. (Pls. Ex. 12 at 7.) Act 1 splits the most counties
(twenty-five) and municipalities (fifty-nine cities, townships, or
boroughs).*fn5 (Id.) With the exception of Act 1's twin, the
zero-deviation version, Act 1 even splits the most precincts. (Id.) To
the extent that Act 1 retains the cores of prior districts, it does so
only for districts containing Republican incumbents. (See id. at 8.)
However, it is on Karcher's final endorsed neutral criteria — the
avoidance of contests between incumbents — that Act 1 fails most
miserably. Although Pennsylvania's loss of two congressional seats
requires the pairing of only two sets of incumbents in the 2002
elections, Act 1 pits six incumbents against each other: two pairs of
Democrats, and one Republican against one Democrat (the latter in a
district that heavily favors Republican candidates). At the same time
Act 1 pits more incumbents than necessary against one another, it also
creates a new district, District 6, in which no incumbent resides. In
the face of such evidence, it is clear that Karcher's neutral criteria
were not high on the priority list in enacting Act 1.
The dissent suggests that our analysis should pay deference to the
policies that Act 1 reflects. However, there is no legal authority to
support the corollary assertion that the only relevant zero-deviation
maps with no precinct splits are those which comport with the general
contours of Act 1.
In summation, it has been conclusively proven that it is possible to
draw a congressional district map with zero population deviation amongst
districts without splitting any precincts. Therefore, Defendants cannot
rely on a general desire to avoid splitting precincts as a legitimate
justification for a map that splits six such precincts. Put another
way, if it is possible to achieve a zero deviation map with no precinct
splits by manipulating whole precincts, the actions of the Defendants in
switching to census block data call into question the legitimacy of their
proffered justifications. We find, therefore, that the Defendants have
failed to provide any legitimate justification for the population
deviations contained in Act 1. As such, we enjoin the implementation of
Having determined that Act 1 is unconstitutional, we now address the
question of the appropriate remedy. The Supreme Court has cautioned that
"reapportionment is primarily a matter for legislative consideration and
determination, and that judicial relief becomes appropriate only when a
legislature fails to reapportion according to federal constitutional
requisites in a timely fashion after having had an adequate opportunity
to do so." White v. Weiser, 412 U.S. 783 (1973). We recognize that the
Pennsylvania General Assembly has primary jurisdiction over congressional
redistricting. Furthermore, no evidence was presented that would permit
us to conclude that the General Assembly would be unable to enact a new
constitutional redistricting plan in time for the 2002 elections.
not a case where the respective legislative bodies are hopelessly
deadlocked and the likelihood is that they will be able to enact a
constitutionally sound plan.
Therefore, the Defendants are ordered to submit a constitutionally
sound redistricting plan that has been enacted into law by the
Commonwealth of Pennsylvania to this court within three weeks of the date
of this opinion.
An appropriate order will follow.
O R D E R
In accordance with the preceding opinion, IT IS HEREBY ORDERED THAT:
(1) Plaintiffs' request for declaratory judgment is GRANTED. Act 1 is
hereby declared unconstitutional as a violation of Article I, § of
the United States Constitution;
(2) Plaintiffs' request for injunctive relief is GRANTED IN PART as
(A) To the extent that Plaintiffs request that the
court enjoin Act 1's implementation, Plaintiffs'
request is GRANTED. Defendants are permanently
enjoined from implementing Act 1;
(B) To the extent that Plaintiffs request that the
court devise its own redistricting plan for use in
the 2002 congressional elections, Plaintiffs'
request is deferred pending compliance by the
Pennsylvania General Assembly with this order; and
(3) The Pennsylvania General Assembly shall, within three weeks of the
date of this order, prepare, enact and submit for review and final
approval by this Court, a congressional redistricting plan in conformity
with this opinion.
This Court shall retain jurisdiction pending issuance of a final order in
IT IS SO ORDERED.
I agree with majority that the plaintiffs have met the first prong of
the Karcher test and that the appropriate remedy, if Act 1 is
unconstitutional, is to give the Commonwealth of Pennsylvania a
reasonable period of time to enact a constitutionally sound redistricting
I respectfully dissent, however, from Section II.B. of the majority
opinion and conclude that defendants have met their burden, although just
barely, to establish a justification for the nineteen person maximum
deviation, and that Act 1 therefore passes constitutional muster.
In Karcher v. Daggett, 462 U.S. 725 (1983), the Court stated that the
plaintiffs' "success in proving that the Feldman Plan was not the product
of a good faith effort to achieve population equality means only that the
burden shifted to the state to prove that the population deviations in
its plan were necessary to achieve some legitimate state objective."
(emphasis added). 462 U.S. at 740. Thus, the focus of the second prong
of the Karcher test is the state's justification for the population
deviations in Act 1.*fn1
The defendants have pointed to a litany of state interests behind Act
1; however, since the focus of our inquiry in the justification phase of
the Karcher analysis is the
population deviation, the only justification
submitted which deals with that issue is the policy objective of limiting
voter precinct splits. Defendants contend that Act 1 could be
"zeroed-out" by using Modification One but that this would entail
splitting twenty-six voting precincts rather than the split of six voting
precincts in Act 1.
Both parties agree that splitting precincts creates additional costs
and work for county election officials in acquiring voting machines, in
customizing ballots, in training precinct officials, in registering
voters and in counting ballots. In addition, it increases the potential
for voter confusion and candidate confusion. Thus, the desire to minimize
the splitting of voting precincts is a legitimate justification,
particularly when the overall population deviation is only nineteen.
The evidence with reference to this justification is, however,
minimal. Dr. John Memmi testified that he was the cartographer who worked
on the final draft of what became Act 1. When he reached the population
deviation of nineteen he was told by his superiors that he could stop
work. Inexplicably, neither party asked Dr. Memmi who told him this or
called as a witness the person who told him to stop work in order to
explore the basis for the decision. As a result, we are left to draw
inferences from the evidence presented.
Defendants ask us to draw the inference that the effort to equalize
population was halted because of a desire not to split any more voting
precincts. There are 9,427 voting precincts in Pennsylvania but there are
322,424 census blocks. Act 1 as enacted contained six voting precinct
splits. (Defendants' exhibit 92). Subsequent to the enactment of Act 1
Dr. Memmi made further adjustments of census blocks in order to produce a
plan with a zero population deviation. Using census blocks inevitably
leads to more voter precinct splits and the resulting plan, designated as
Modification One contains twenty-six voting precinct splits. (Defendants'
This inference is also supported by the testimony of Dr. Memmi
concerning the conversation when he was working on the last draft of what
became Act 1 and told that he could stop work.
Q: Did they tell you why they thought that was
A: They did not specifically say John, Bill, these are
the reasons why we are stopping at 19. I don't mean
to be curt.
One of the prevailing themes throughout the exercise
was to minimize the number of precinct incursions. We
were at six. I worked with individuals who understand
the election process. They understand the
administration at the county level.
I do not profess to have any more than at best a
superficial understanding of that. But the concern
was to minimize the number of precinct incursions so
that election administration would be as smooth as it
R. at 320-21.
Finally, I note that neither party has, even with the benefit of
hindsight, submitted a plan following the general contours of Act 1 which
does not involve the splitting of more than six voting precincts. Thus,
I find that the inference defendants suggest, that the justification for
Act 1 is the reduced number of voting precinct splits, is reasonable.
Plaintiffs, on the other hand, ask us to infer that the Act 1 plan was
not "zeroed out" because of partisan considerations. Plaintiffs produced
a map that was "zeroed out" without splitting any voting precincts,
Alternative Four. However, this plan does not follow the general
contours of Act 1. The Court has oftentimes reminded us of the necessity
not to override state goals, policies and preferences any more than is
necessary. See Abrams v. Johnson, 521 U.S. 74, 101 (1997); Upham v.
Seaman, 456 U.S. 37, 41-42 (1982); White v. Weiser, 412 U.S. 783, 795
There is no evidence to support a conclusion that Alternative Four
encompasses the many other state goals, policies and preferences inherent
in Act 1 which resulted in its enactment. In addition, plaintiffs have
suggested no connection between the nineteen person deviation in Act 1
and the alleged partisanship of the plan. Indeed, any such suggestion is
contradicted by Dr. Memmi's specific testimony that when he was trying to
reduce the plan to the minimal population deviation possible on December
31, 2001, nobody expressed any concern that continuing to trade whole
precincts might change the political characteristics of some of the
districts and that during the entire procedure there was no commentary of
a political nature. R. at 322. Moreover, to the extent that the
plaintiffs are arguing partisan gerrymandering, the Karcher Court noted
that "beyond requiring states to justify population deviations with
explicit, precise reasons, which might be expected to have some
inhibitory effect, Kirkpatrick does little to prevent what is known as
gerrymandering. . . . Kirkpatrick's object, achieving population
equality, is far less ambitious than what would be required to address
gerrymandering on a constitutional level." Karcher, 462 U.S. at 734 n.
6. Thus, the inference which the plaintiffs ask us to draw is not
supported by the evidence, by reason, or by law.
I conclude, therefore, that the nineteen person deviation was justified
by the "legitimate state objective" of avoiding further splitting of
voter precincts and that there has been no proof that the population
deviation of nineteen persons bore any relationship to the alleged
partisanship of Act 1. Thus, Act 1 is not unconstitutional.