Reply Brief — O. John Benisek, et al., Appellants v. Linda H. Lamone, Administrator, Maryland State Board of Elections, et al.

Supreme Court briefMar 13, 2018

Ask Donna

What actually matters in this document.

Text

No. 17-333

In the Supreme Court of the United States

O. JOHN BENISEK, EDMUND CUEMAN,

JEREMIAH DEWOLF, CHARLES W. EYLER, JR.,

KAT O’CONNOR, ALONNIE L. ROPP,

and SHARON STRINE,

Appellants,

LINDA H. LAMONE, State Administrator of Elections,

and DAVID J. MCMANUS, JR., Chairman of the

Maryland State Board of Elections,

Appellees.

On Appeal from the United States District Court

for the District of Maryland

REPLY BRIEF

MICHAEL B. KIMBERLY

Counsel of Record

PAUL W. HUGHES

Mayer Brown LLP

1999 K Street, NW

Washington, DC 20006

(202) 263-3127

mkimberly@

mayerbrown.com

Counsel for Appellants

TABLE OF CONTENTS

Table of Authorities ....................................................... ii

Introduction ....................................................................1

Argument ........................................................................2

A. Article III courts are capable of detecting

constitutional burdens in cases like this..............2

B. The State concedes that the majority’s

change-every-election standard is wrong.............8

C. Plaintiffs’ claim neither outlaws politics in

redistricting nor favors the status quo ante ........9

D. Plaintiffs proved specific intent..........................12

E. Plaintiffs proved burden and causation .............15

F. The floodgates will remain closed ......................20

G. The Court cannot and should not affirm on

alternative grounds.............................................21

Conclusion.....................................................................23

TABLE OF AUTHORITIES

Cases

Allen v. State Bd. of Elections,

393 U.S. 544 (1969)................................................5

Anderson v. Celebrezze,

460 U.S. 780 (1983)............................................3, 6

Bartlett v. Strickland,

556 U.S. 1 (2009)....................................................7

Benisek v. Lamone,

320 F.R.D. 32 (D. Md. 2017) ................................19

Branch v. Smith,

538 U.S. 254 (2003)..............................................23

Bush v. Vera,

517 U.S. 952 (1996)..............................................13

Cook v. Gralike,

531 U.S. 510 (2001)........................................3, 5, 6

Cooper v. Harris,

137 S. Ct. 1455 (2017)..............................10, 13, 14

Crawford v. Marion Cty. Election Bd.,

553 U.S. 181 (2008)................................................8

Doe v. Reed,

561 U.S. 186 (2010)................................................4

Fortson v. Dorsey,

379 U.S. 433 (1965)................................................5

Gaffney v. Cummings,

412 U.S. 735 (1973)................................ 4, 5, 10, 11

Hartman v. Moore,

547 U.S. 250 (2006)..............................................18

LULAC v. Perry,

548 U.S. 399 (2006)................................................8

iii

Cases—continued

Manion v. American Airlines, Inc.,

395 F.3d 428 (D.C. Cir. 2004) ..............................22

McCreary Cty. v. ACLU,

545 U.S. 844 (2005)........................................13, 15

Mobile v. Bolden,

446 U.S. 55 (1980)..................................................7

Mt. Healthy City Sch. Dist.

Bd. of Educ. v. Doyle,

429 U.S. 274 (1977)............................ 17, 18, 19, 21

NAACP v. Alabama,

357 U.S. 449 (1958)................................................4

Parham v. Hughes,

441 U.S. 347 (1979)..............................................19

Parsi v. Daioleslam,

778 F.3d 116 (D.C. Cir. 2015) ..............................22

Rutan v. Republican Party,

497 U.S. 62 (1990)..............................................4, 7

Smith v. Doe,

538 U.S. 84 (2003)................................................13

Tashjian v. Republican Party,

479 U.S. 208 (1986)................................................4

Thornburg v. Gingles,

478 U.S. 30 (1986)..................................................7

United States v. Lopez,

514 U.S. 549 (1995)................................................7

United States v. O’Brien,

391 U.S. 367 (1968)..............................................14

Vieth v. Jubelirer,

541 U.S. 267 (2004)................................ 1, 2, 10, 11

Cases—continued

Walz v. Tax Comm’n,

397 U.S. 664 (1970)................................................7

Wisconsin Mem’l Park v. Commissioner,

255 F.2d 751 (7th Cir. 1958)..........................16, 17

Statutes

2 U.S.C. § 2a(c) ..........................................................23

Voting Rights Act § 2 ........................................6, 7, 21

INTRODUCTION

All nine Justices agreed in Vieth that partisan

gerrymandering can violate the Constitution. The

central question presented here is whether the Court

has the Article III authority to do anything about it.

It does. Plaintiffs’ First Amendment challenge to

Maryland’s single-district partisan gerrymander is

justiciable under the same standards that govern all

First Amendment challenges in cases involving

voting and elections. Thus, when a State deliberately

retaliates against its citizens for their past voting

behavior by diluting their votes so significantly that

it eviscerates their political opportunity—and the

burden would not have been imposed otherwise—the

targeted voters have a First Amendment claim, and

Article III courts have jurisdiction to resolve it. And

if any case fits that mold, this is it.

The State and its amici offer a mishmash of disconnected rejoinders. Yet almost none bear on the

issue of justiciability—that is, on whether the burdens of partisan gerrymandering are judicially

identifiable according to reasoned, manageable standards.

Nor are the State’s merits arguments persuasive.

Enforcement of the First Amendment in this case

will not “outlaw” all politics in redistricting; it will

forbid only deliberate efforts to dilute citizens’ votes

because of the way they voted in the past. This is a

difference of kind, not degree; most political considerations remain entirely permissible.

Nor do we argue that plaintiffs’ political success

before the redistricting entitles them to continued

political success after the redistricting. The point is

only that plaintiffs have a right not to be singled out

for disfavored treatment on the basis of their party

preference or successful past support of Republican

candidates for office.

For more than 200 years, our Nation has striven

to “demonstrat[e] to the world how democracy

works.” Vieth v. Jubelirer, 541 U.S. 267, 317 (2004)

(Kennedy, J., concurring). Partisan gerrymandering,

an admitted attempt to “rig[] elections,” is at war

with that aspiration. Ibid. It is also irreconcilable

with the First Amendment’s essential guarantee that

no State may punish its citizens for their political

beliefs.

ARGUMENT

A. Article III courts are capable of detecting

constitutional burdens in cases like this

1. The intended consequences of partisan gerrymandering are real, identifiable, and undisputed.

First and foremost, a gerrymander dilutes opposition

votes, with the purpose of dictating electoral outcomes in the targeted district. See Appellants’ Br. 1419, 40-41. This, in turn, depresses political support

for the opposition party. See Appellants’ Br. 19-20;

J.S. App. 71a (Niemeyer, J., dissenting).

Identifying these intended burdens entails a

common-sense assessment of a gerrymander’s practical effects. As we explained (Appellants’ Br. 37), it

requires an evaluation of how the map has affected

the distribution of voters throughout the district,

whether those changes have depressed voter engagement and turnout, and whether they have probably

altered (or probably will alter) the outcomes of any

elections. The evidentiary tools necessary for this

kind of assessment are neither complex nor elusive:

They include the tools that gerrymanderers themselves use to ensure their own success.

This kind of pragmatic assessment of evidence

would be at home in any Article III lawsuit. In the

election-law context, in particular, this Court has

recognized that election regulations can “burden[]

the availability of political opportunity” by, among

other things, making it “more difficult to recruit and

retain” supporters or secure “media publicity and

campaign contributions” and making voters “less

interested in the campaign.” Anderson v. Celebrezze,

460 U.S. 780, 792-793 (1983). Thus, according to the

Court’s cases, a law’s infliction of a particular “political disadvantage” that “disfavor[s] a [particular]

class of candidates” (Cook v. Gralike, 531 U.S. 510,

523, 525 (2001)) is an identifiable burden.1

2. The State asserts that gerrymandering does

not inflict any First Amendment burdens at all. As

the State sees it (Br. 29), “[a]ny member of a political

minority may still vote in elections, campaign for

preferred candidates, petition her representative,

and access constituent services,” even if she lives in

an openly gerrymandered district. Because partisan

gerrymandering does not “impact who is on the

ballot, who can cast a vote, or when votes may be

cast” (State Br. 30), gerrymandering does not impose

any cognizable burden on the right to vote. Accord

Freedom Partners Amicus Br. 8-10.

This line of reasoning ignores our theory. As the

district court explained, gerrymandering “indirectly

imping[es] on the direct rights of speech and association by retaliating against citizens for their exerThe State puzzlingly asserts (Br. 2 n.1) that nearly all factual

matters in this case are “not yet ripe for appellate review.” That

is wrong. The entire record was before the district court, and a

court abuses its discretion when it arbitrarily disregards or

clearly misconstrues probative evidence.

cise.” J.S. App. 101a (emphasis added). A claim

sounding in retaliation is distinct from a claim that

the State is directly restricting First Amendment

rights, which might be the case if the State forbade

Republicans from voting altogether. On the theory

presented here, plaintiffs must prove that the State

“penalized” Republicans for their past “exercise of

protected speech and association rights” (Rutan v.

Republican Party, 497 U.S. 62, 65, 77 n.9 (1990)), not

that it is directly and affirmatively restricting their

current or future exercise. That is exactly what we

have done.

Casting a ballot for one candidate over another is

perhaps the most fundamental and important expression of a citizen’s political beliefs and ideas. Cf.

Doe v. Reed, 561 U.S. 186, 195 (2010) (signing a referendum petition “expresses a particular viewpoint”

protected by the First Amendment). “[F]reedom to

engage in association for the advancement of beliefs

and ideas is an inseparable aspect of the ‘liberty’

assured by the Due Process Clause of the Fourteenth

Amendment, which embraces freedom of speech.”

Tashjian v. Republican Party, 479 U.S. 208, 214

(1986) (quoting NAACP v. Alabama, 357 U.S. 449,

460 (1958)). And among the means of “furthering [a

party’s] political and organizational goals” protected

by the First Amendment is citizens’ “casting their

votes for some or all of the Party’s candidates.” Id. at

215.

It follows that a burden deliberately imposed upon particular citizens because of the views reflected

in their voting histories and party affiliations is a

burden upon protected speech. Partisan gerrymandering inflicts such a burden. As the Court recognized in Gaffney v. Cummings, 412 U.S. 735 (1973),

district lines may be drawn “to minimize or cancel

out the voting strength of racial or political elements

of the voting population.” Id. at 751 (quoting Fortson

v. Dorsey, 379 U.S. 433, 439 (1965)). Accord id. at

754 (lines may be drawn so that the “voting

strength” of particular “political groups” is “invidiously minimized”). Thus, the right to vote can be

burdened, not only “by an absolute prohibition on

casting a ballot,” but also “by a dilution of voting

power” because of past views expressed. Allen v.

State Bd. of Elections, 393 U.S. 544, 569 (1969).

In the words of the 12 States appearing as amici,

this burden is “obvious”: “If the party in power

considers voting history or party affiliation when

drawing a district line,” it does so to “disfavor the

other party and to ‘punish’ voters of the other party

by making it harder for them to win seats.” Michigan

Amicus Br. 9.

This case proves the point. The 2011 gerrymander inflicted vote dilution so significant that it

made it effectively impossible for Republicans to win

elections in the Sixth District, thus burdening

Republicans’ availability of political opportunity. See

J.S. App. 53a, 69a (Niemeyer, J., dissenting). Indeed,

the gerrymander was so successful that it has as a

practical matter “dictate[d] electoral outcomes”

(Gralike, 531 U.S. at 523) in the Sixth District since

2012, as was its goal: Whereas the chances of a

Republican victory were 99.7%-100% in 2010, the

chances of a Democratic victory were 92.5%-94.0% in

2012 onward. See 3JA826; 4JA887, 1107-1110, 1124.

The State’s own experts agree that the vote dilution

deliberately visited upon plaintiffs and their fellow

Republicans placed them at a clear political disadvantage, burdening their political opportunity be-

cause of the way they voted in the past. See 1JA255256; 3JA827.2

And the State’s contrary position proves too

much. By its rationale, Maryland law (even its

constitution) could provide that redistricting maps

must be drawn to ensure maximum dilution of Republican votes, subject only to the limits of the oneperson-one-vote rule and Section 2 of the Voting

Rights Act. Cf. Oral Arg. Tr. at 19-22, 26-28, Gill v.

Whitford, No. 16-1161. A provision of this kind would

not prevent citizens from “campaign[ing] for preferred candidates” or change “who is on the ballot, who

can cast a vote, or when votes may be cast.” State Br.

29-30. And at the conclusion of the election, all

citizens would be free to “petition [their] representative, and access constituent services.” Id. at 29.

Thus, in the State’s view, such a law would not be a

“tangible restriction impacting individuals’ ability to

cast their votes effectively” and would not violate the

First Amendment. Ibid. That cannot be correct.

The point is simple: If a State may not intentionally “burden[] the availability of political opportunity” using discriminatory filing deadlines (Anderson, 460 U.S. at 793); “dictate electoral outcomes”

using ballot notations (Gralike, 531 U.S. at 523); or

The State says (Br. 57-58) that Dr. McDonald’s vote dilution

analysis falls short because he neither demonstrated polarized

voting nor accounted for unaffiliated voters. That is wrong. See

3JA768-770 (Dr. McDonald finding that Republicans and

Democrats engage in highly polarized voting); 2JA526-527 (Dr.

McDonald explaining how he accounted for unaffiliated voters).

The State also repeatedly asserts (Br. 1-9, 23, 48, 56) that the

district was made “competitive” by the gerrymander. But saying

so does not make it so—and the evidence shows otherwise. See

Appellants’ Br. 12-20.

“interfere with [the] freedom to believe and associate” using patronage practices (Rutan, 497 U.S. at

76), neither may it intentionally do those things by

manipulating district lines. And if the burdens are

capable of rational judicial assessment in those other

contexts, they are equally so in this one.

3. The State complains (Br. 30-31) that the

district court’s vote dilution standard is “more easily

met” than the Gingles standard applicable to Section

2 claims. (Emphasis omitted). That ignores that the

1982 amendments to Section 2 “repudiated” the

“intent test” applicable to race discrimination claims

under Mobile v. Bolden, 446 U.S. 55 (1980). See

Thornburg v. Gingles, 478 U.S. 30, 43-45 (1986). It is

sensible than Section 2’s effects-only test would call

for a different analysis than a test that includes the

“inordinately difficult” burden of proving intent. Id.

at 44 (quoting Senate Report). For its part, this

Court has reserved the question of how “intentional

discrimination affects the Gingles analysis.” Bartlett

v. Strickland, 556 U.S. 1, 20 (2009) (plurality).

4. Some amici (but not Maryland) suggest that

these burdens are nonjusticiable because they cannot

be judged according to predetermined bright lines

established by statistical metrics. E.g., Wisconsin

Amicus Br. 12; Freedom Partners Amicus Br. 26-27.

But “as the branch whose distinctive duty it is to

declare ‘what the law is,’ [this Court is] often called

upon to resolve questions of constitutional law not

susceptible to the mechanical application of bright

and clear lines.” United States v. Lopez, 514 U.S.

549, 579 (1995) (Kennedy, J., concurring). “[I]t is an

essential part of adjudication to draw distinctions,

including fine ones, in the process of interpreting the

Constitution” one case at a time. Walz v. Tax

Comm’n, 397 U.S. 664, 679 (1970). Thus, the Court

has never before required a universal “‘litmus test’

that would neatly separate valid from invalid” burdens as a precondition to exercising Article III jurisdiction in election-law cases. Crawford v. Marion

Cty. Election Bd., 553 U.S. 181, 191 (2008) (plurality). It should not do so here.

B. The State concedes that the majority’s

change-every-election standard is wrong

We demonstrated (Appellants’ Br. 40-43, 53-54)

that the majority erred by inflexibly requiring plaintiffs to prove that the gerrymander flipped the Sixth

District in 2012, 2014, and 2016 and that it will

continue to control electoral outcomes in all future

elections. We showed, in particular, that the majority’s approach on this point is out of step with the

opinion denying the State’s motion to dismiss (e.g.,

J.S. App. 104a, 106a) and would deny relief in many

cases involving significant constitutional injuries.

The State now agrees. It affirmatively concedes

(Br. 30) that, no matter how one conceives plaintiffs’

burden in this case, the majority was wrong to

“[draw] the line between tangible and intangible

interests at the point where a seat changes hands.”

That is so, the State argues, because “[t]he circumstance that a group does not win elections does not

resolve the issue of vote dilution.” Ibid. (quoting

LULAC v. Perry, 548 U.S. 399, 428 (2006)).3

There is no dispute on this point. As we have

said repeatedly, vote dilution represents a constituThe State implies (Br. 18 n.11, 22, 47-48) that we invited this

error. But as we explained in the opposition to the motion to

affirm (at 1-2 & n.1), the sentence that the State cites for that

proposition has been taken entirely out of context and twisted

180 degrees.

tionally cognizable burden when it produces any

specifically intended, practical, adverse burden. See

Opp. to Mot. to Affirm 1-2 (collecting citations). If the

Court concludes that plaintiffs’ First Amendment

retaliation claim is justiciable, it should vacate the

order below and remand for reconsideration of the

motion for a preliminary injunction under the proper

standard for burden.

C. Plaintiffs’ claim neither outlaws politics in

redistricting nor favors the status quo ante

1. The State asserts (Br. 27) that this Court’s

cases call inflexibly for “a manageable standard for

courts to determine when partisanship ‘has gone too

far’” and accuse us of attempting to “eva[de]” that requirement. That is mistaken.

As we explained (Appellants’ Br. 33-34), the “how

much is too much” question arises only in challenges

under the Equal Protection Clause, which asks

whether mapdrawers have gone “too far” in their

efforts to gerrymander—as though a little gerrymandering in just one district is okay, but extreme statewide gerrymandering is not. Again, the First Amendment approach is different; the distinction it draws is

one of kind, not degree.

Thus, we cited (Appellants’ Br. 47-48) a range of

valid political considerations that may be taken into

account under the First Amendment—considerations

that do not require mapdrawers to single out citizens

for disfavored treatment on the basis of their past

protected conduct. These include avoiding contests

between incumbents, respecting municipal boundaries and communities of interest, and allocating

specific institutions to particular districts to preserve

affinity relationships. The State cites it own examples. See State Br. 13.

We also explained that, even as to consideration

of voter history and party affiliation, “‘[t]he inquiry is

not whether political classifications were used,’ but

‘whether political classifications were used [with the

purpose and effect of] burden[ing] a group’s representational rights.” Appellants’ Br. 48 (quoting Vieth,

541 U.S. at 315 (Kennedy, J., concurring)). Simply

put, the First Amendment would not require “extirpating politics” from an “essentially political” process. Gaffney, 412 U.S. at 754.

Maryland ignores our arguments on this score,

insisting without explanation (Br. 19) that application of the First Amendment here “threatens to

render any partisan motive fatal to redistricting.”

The 12 amici States take the same unreasoned

approach, asserting (Br. 10) that “[a]pplying the

First Amendment to partisan redistricting would

outlaw any consideration of politics,” but without

addressing our arguments.

Freedom Partners at least acknowledge our position on this point, but they say (Br. 6) that the Court

foreclosed any distinctions among political considerations in Cooper v. Harris, 137 S. Ct. 1455 (2017).

Not so. The Court in Cooper merely included “partisan advantage” in a catalogue of acceptable redistricting factors. Id. at 1464. That is entirely consistent with our point that most political considerations are acceptable, but deliberately targeting citizens on the basis of voter history and party affiliation is not.

2. The State asserts (Br. 32) that plaintiffs’ claim

rests on “the flawed premise that the preexisting

map was constitutionally preferable” and elevates to

preferred “constitutional status a prior district that

is itself the result of partisan considerations.”

That is unequivocally wrong. Those responsible

for Maryland’s 2011 redistricting were free under the

First Amendment to dismantle and redraw the Sixth

District however they liked, regardless of its pre2011 character—so long as it was for a lawful reason.

What they were not allowed to do is dismantle the

Sixth District for the purpose of diluting Republican

votes because of Republicans’ past successful support

of Roscoe Bartlett. See Vieth, 541 U.S. at 314 (Kennedy, J., concurring).

To be sure, the legality of the pre-redistricting

map may be implicated in some limited circumstances—but not in a way helpful to the State. For

example, the deliberate maintenance of a prior gerrymander would be every bit as unconstitutional as the

creation of a new one; deliberate inaction can violate

the First Amendment, too. See Appellants’ Br. 47.

Contra CLC Amicus Br. 23-24. In any event, the

point is that the status quo is not entitled to any

special weight under the First Amendment.4

The State is also wrong to say (Br. 43-44) that

the decision below is inconsistent with the outcome

in Gaffney. Accord CLC Amicus Br. 22-23. Although

it is not a question presented here, a State’s good

faith effort “fairly to allocate political power to the

parties in accordance with their voting strength”

may well pass strict scrutiny. Gaffney, 412 U.S. at

754. In other words, narrowly drawn efforts to undo

The Campaign Legal Center proposes (Br. 20-29) several

hypothetical scenarios in which the First Amendment supposedly fails to produce the results it prefers. But CLC’s

analysis is premised on the misconception that plaintiffs’ injury

inheres in Republican candidates’ electoral losses in the Sixth

District, rather than in the dilution of Republicans’ votes that

ensured those losses. It does not. See Appellants’ Br. 36-43.

past partisan gerrymandering are not likely to

violate the First Amendment. Of course, a State

wishing to undo a past unlawful gerrymander will

have several other tools at its disposal as well,

including using earlier, non-tainted maps as a

starting point or throwing out all prior maps and

simply starting from scratch.

D. Plaintiffs proved specific intent

The State asserts (Br. 37-41) that courts cannot

discern specific intent here because (1) the “secret

ballot” prevents government officials from knowing

how citizens have voted, and (2) it is impossible to

determine the intent of a legislature in any event.

Both contentions are meritless.

1. The State’s suggestion that it is ignorant of

voter history because of the “secret ballot” borders on

absurd. State Br. 39. The defendants “aggregat[ed]

address-level voter registration data, address-level

voter history, and official election results in Maryland” for use in redistricting. 3JA656. They also

made “available votes cast during early voting, on

election day, and by absentee or provisional ballot at

various levels of aggregation.” 3JA659.

According to this data, which was given to the

mapdrawers (1JA175; 3JA659), those responsible for

the gerrymander could sort Maryland’s nearly two

thousand precincts based on voter history. In Frederick County’s Precinct 02-017, for example—which

covers just a few blocks in downtown Frederick—

mapdrawers saw that there were 148 election-night

votes cast for the Republican incumbent and 314

votes cast for his Democratic challenger in 2010. This

precinct was kept in the Sixth District in 2011. But

in Precinct 24-001, just a few blocks away, the count

was 444 for the Republican incumbent and 339 for

his Democratic challenger. This precinct was moved

out of the Sixth District in 2011.5

Data like this, reflecting voter history in small

batches of citizens, is self-evidently sufficient to support a First Amendment retaliation claim. Indeed,

courts often adjudicate allegations that the State has

intentionally targeted groups of citizens rather than

specific individuals, as they do in all racial gerrymandering cases. See, e.g., Cooper, 137 S. Ct. at

1466; Bush v. Vera, 517 U.S. 952, 958 (1996) (plurality). “The fact that the State moved Republican

voters out of the Sixth District * * * based on

precinct-level data, and did not examine each voter’s

history” individually before taking “punitive action

does not make its action less culpable under the First

Amendment.” J.S. App. 67a (Niemeyer, J.).

We made these points in prior briefing. See Opp.

to Mot. to Affirm 9; Appellants’ Br. 14. The State

talks past them without rejoinder.

2. The State is also wrong that legislative intent

in indeterminable. See State Br. 37-41. Inquiry into

legislative intent is “common” and “makes up the

daily fare of every appellate court in the country.”

McCreary Cty. v. ACLU, 545 U.S. 844, 861 (2005). It

“is a key element of a good deal of constitutional

doctrine,” as with respect to Establishment Clause

(ibid.) and Ex Post Facto Clause cases (Smith v. Doe,

538 U.S. 84, 92-93 (2003)).

So, too, is it a key element in racial gerrymandering cases. As the Court just recently confirmed in

Vote-history data is at elections.state.md.us/elections/2010/election_data/index.html and archived at perma.cc/W3VPBRXZ. Frederick County’s 2010 precinct map is archived at

perma.cc/W83G-B86W.

Cooper, “[t]he plaintiff may make the required showing * * * of legislative intent” through either “direct

evidence” or “circumstantial evidence of a district’s

shape and demographics, or a mix of both.” 137 S. Ct.

at 1464. The Court tellingly did not express any

concern for the supposed unworkability of evaluating

legislative intent in redistricting cases. See also

NAACP Amicus Br. 15-17.

The Court’s decision in United States v. O’Brien,

391 U.S. 367 (1968), which is not a First Amendment

retaliation case, does not suggest otherwise. Indeed,

the Court there reaffirmed that laws with “inevitable

unconstitutional effect[s],” such as “the redrawing of

municipal boundaries” to disenfranchise voters “for

no reason other than” their race, are properly subject

to challenge. Id. at 385. It rejected the First Amendment claim in that case because draft-card burning is

not “inevitably or necessarily expressive,” and Congress had alternative, legitimate grounds for the challenged law. Id. at 385-386. Moreover, this case concerns the core workings of our democratic processes;

the Court’s resolution of plaintiffs’ claim in this

special and narrow context need not predetermine

the permissibility of claims challenging legislation in

other contexts. Cf. State Br. 35-36 (citing cases).

Proving specific retaliatory intent in cases like

this does not pose insurmountable practical problems

either. Here, the record abounds with evidence (and

it is frankly admitted by those who were responsible

for the redistricting plan) that the goal was to dilute

Republican votes to ensure “the election of another

Democrat” in the Sixth District. 1JA44. This goal

was confirmed time and time and time again. See,

e.g., 3JA661-665 (¶¶ 40-51) (stipulated public statements); 1JA107-108 (testimony of Eric Hawkins);

1JA236 (testimony of Sen. Robert Garagiola).

3. The State asserts (Br. 40-41, 49, 51-52) that

intent is not provable in this case because of the

later-occurring veto referendum. Not so.

The constitutional violation was complete at the

time the State enacted the map; that the public later

rejected a referendum challenging that action has no

logical bearing on the intent underlying the legislature’s original action. See McCreary, 545 U.S. at

866. That is especially so with respect to a public

referendum; constitutional rights are counter-majoritarian precisely because they protect political

minorities from the majority’s overreach.

Besides, the referendum proves little: The ballot

question was drafted so ambiguously that the State’s

expert, Dr. Lichtman, could not explain what it

meant. Dkt. 177-49, at 174:22-175:10.6 It would be

difficult to read much into voters’ “approval” of text

so opaque that not even a redistricting expert can

understand it.

E. Plaintiffs proved burden and causation

The State says (Br. 21, 41-42) that the causation

element of the test “significantly complicates the

problem of judicial manageability” because it turns

on “speculation” about a “hypothetical state of affairs.” That both misunderstands the First Amendment framework and confuses the burden and

causation elements of the claim.

1. The question of burden requires a comparison

of the status quo post with the status quo ante: Compared with the state of affairs before 2011, have

The ballot asked whether voters were “for” or “against” the

following text: “Establishes the boundaries for the state’s eight

United States Congressional districts based on recent census

figures as required by the United States Constitution.” Ibid.

plaintiffs been placed at an identifiable political disadvantage by the gerrymandered map? The inquiry

does not require the district court to conjure counterfactual maps or to imagine how elections might have

turned out under them. See supra, pp. 9-11. Accord

Wisconsin Amicus Br. 3 (plaintiffs’ First Amendment

claim “would not force courts to adjudicate the

legality of maps based upon conjectural, statewide

hypothetical states of affairs”).

The majority below worried (J.S. App. 26a-28a)

that the reversal of Republicans’ political prospects

in the Sixth District after 2011 may not have been

attributable to the vote dilution visited by the complete reconfiguration of the district’s lines, and

therefore that the gerrymander’s dilution of Republican votes may not have represented a burden at all.

But as we explained in our principal brief (at 55),

that concern depends on the implausible speculation

that between 2010 and 2012, tens of thousands of

historical Republican voters spontaneously abandoned their party and began supporting Democrats

in each election thereafter.

That speculation “overlooks the obvious.” J.S.

App. 34a (Niemeyer, J., dissenting). The opposite

conclusion—that the inversion of the political

complexion of the district after 2011 is attributable

to the gerrymander—depends not on speculation, but

on rational, well-supported inference. “The difference

between speculation and inference lies in the substantiality of the evidence constituting the premise.”

Wisconsin Mem’l Park v. Commissioner, 255 F.2d

751, 752 (7th Cir. 1958). Here, the premise underlying the inference is that the DPI and PVI are

generally accurate and reliable measures of a party’s

electoral prospects because voters are relatively

predictable from election to election—a fact sup-

ported in the record (1JA151-154; 4JA1107), uncontested by the State, and confirmed by the State’s own

expert (1JA259, 266-267). What is more, reasonable

inferences do not require certitude: “[I]n law we

speak in terms of the probability and likelihood that

the premises buttress the conclusions,” not certainty.

Wisconsin Mem’l Park, 255 F.2d at 752.

Against this background, the majority’s concern

about mass defections from the Republican party

independent of the gerrymander was not reasonably

founded. Using the same tools that gerrymanderers

themselves use—metrics like the DPI and PVI—we

showed (Appellants’ Br. 17-19) that it was the purposeful swapping of tens of thousands of Republican

voters with tens of thousands of Democratic voters

that eviscerated Republicans’ political opportunity in

the district after 2011.7

2. The causation element asks a different question: whether the vote dilution visited by the gerrymandered map would have come to pass even absent

the mapdrawers’ intent to burden voters on the basis

of their voting histories and political-party affiliations. See Mt. Healthy City Sch. Dist. Bd. of Educ. v.

Doyle, 429 U.S. 274, 286-287 (1977). The mode of

analysis on that point is identical to the analysis in

any First Amendment retaliation case—it requires

the State to prove that it would have drafted and

adopted a map that cracked the Republican majority

in the Sixth District independent of the specific

intent to dilute Republican votes there.

Contrary to the State’s assertion (Br. 49), therefore, “the

official action” (dilution of votes) was also “the asserted injury”

(infliction of concrete political disadvantage), just as in other

First Amendment retaliation cases.

The State insists (Br. 49-55) that disproving butfor causation was not its burden to bear because the

Mt. Healthy burden-shifting framework should not

apply here. Its arguments are unpersuasive.

The State says first (Br. 51-52) that Mt. Healthy

should not apply to partisan gerrymandering claims

because there is “particularly attenuated causation”

between the decision to dilute votes and the actual

burden imposed. That is not what the evidence

shows. Rank-and-file lawmakers were not involved

in the drafting or approval of the map; rather,

Democratic leaders explicitly directed mapdrawers

the “to go for the Sixth” (1JA44) and thereafter

presented the map as a fait accompli, jamming it

through both houses and past the governor’s desk in

under three days (3JA660 (¶ 34)). There was no

intervening deliberation by others—no independent

check, like the prosecutor in Hartman v. Moore, 547

U.S. 250 (2006).

Hartman ultimately reflected a concern to protect government actors against claims that are easy

to allege and difficult to disprove. In this case, the

inverse obtains: Plausible partisan gerrymandering

claims will be difficult to allege and—when there are

genuine alternative explanations—easy for the State

to disprove.

The State also asserts (Br. 55) that “[t]he named

defendants, who merely administer elections, are in

no better position than the plaintiffs to ‘know[]’” of

the possible alternative explanations for the map.

The State raised that same meritless argument as

part of its sustained refusal to cooperate in discovery;

Chief Judge Bredar rightly rejected it. See Benisek v.

Lamone, 320 F.R.D. 32, 36 (D. Md. 2017) (holding

that the named defendants were able, and therefore

required, “to obtain documents and other information from the various state agencies and actors”

involved in the redistricting).

Finally, the State cites an Equal Protection

Clause case (Br. 53 (citing Parham v. Hughes, 441

U.S. 347, 351 (1979)) for the proposition that its laws

are entitled to a “presumption of validity.” But this is

a First Amendment case, and the Mt. Healthy

burden-shifting framework comes into play only

assuming that the plaintiffs have shown a specific

intent to burden citizens on the basis of their

protected expression. When a statute is unconstitutionally discriminatory like this, any presumption of

validity is overcome. Ibid.

3. Regardless who bears the burden, causation is

shown here. The State relies principally on its

insubstantial “I-270” explanation for the decision to

reconfigure the Sixth District. For support on this

point, it cites (State Br. 7) a Baltimore Sun article

(2JA486) and a state environmental impact study

(4JA1052), but without any evidence that anyone

actually had or considered either one of them; and

“public testimony” concerning the “I-270 corridor”

(2JA403-423), but without acknowledging that each

person whose testimony is cited is a high-ranking

Democratic Party insider with a vested interest in

cracking the Sixth District. Cf. Dkt. 177-49, at

126:15-127:8 (Dr. Lichtman testifying concerning his

experience with “sham explanation[s]” in racial

gerrymandering cases).

The State also offers a series of unexplained

citations to testimony in this case. State Br. 7-8. But

the testimony comprises only a few artfully stated

descriptions of the effects of the gerrymander, not its

goals. As we explained in our principal brief (at 21),

all of the witnesses who were asked directly about

the I-270 corridor flatly denied ever actually considering it during the mapdrawing process. The

State does not deny this.

The State also points to Governor O’Malley’s

decision that the First District should not cross the

Chesapeake Bay. State Br. 56. (citing 1JA44). But

Dr. McDonald showed that it would have been possible (indeed, easy) to draw a map that accomplished

that goal—and, indeed, respected every single other

political consideration reflected the 2011 map—without diluting Republican votes in the Sixth District.

3JA786. He did so by altering the lines between the

Sixth and Eighth Districts only. Ibid. We have

referred to this alternative map on numerous prior

occasions, but the State has not once bothered to

acknowledge it, let alone respond.

F. The floodgates will remain closed

Wisconsin (but not Maryland) worries that

holding Maryland’s 2011 gerrymander to violate the

First Amendment will “promptly spawn expensive

and uncertain litigation” because it “will be trivially

easy for plaintiffs to scrounge up an expert or two to

testify to a more-than-de-minimis impact.” Amicus

Br. of Wisconsin 2-3. Cf. Texas Amicus Br. 14, Gill v.

Whitford, No. 16-1161 (affirming claim will “invite[]

massive judicial intervention”).

That is wrong for two reasons.

First, the standard advocated here and adopted

below “contains several important limitations that

help ensure that courts will not needlessly intervene

in what is quintessentially a political process.” J.S.

App. 105a. Among other things, it requires proof of

specific intent; mere awareness of consequences is

insufficient. Id. at 105a-106a. No matter the charac-

ter or magnitude of the burden asserted, there can be

no relief if the burden was not purposely imposed in

disapproval of voters’ protected conduct. And even

with respect to less substantial burdens that are

specifically intended (scenarios that are candidly

hard to imagine), the Mt. Healthy standard will

protect States that would have drawn a map with

the same effect either way. And come 2020, States

presumably will abide a decision of this Court holding that legislatures may not deliberately burden

voters because of their voting histories and politicalparty affiliations.

Second, in the absence of a standalone cause of

action against partisan gerrymandering, citizens targeted by partisan gerrymanders are funneling their

challenges through other causes of action, including

racial gerrymandering claims under the Equal Protection Clause and racial vote dilution claims under

Section 2 of the Voting Rights Act. See Appellants’

Br. 45-46. This rerouting of partisan gerrymandering

claims places counterproductive emphasis on race in

redistricting and distorts and undermines clear and

consistent enforcement of constitutional law. See

generally Law Professors’ Amicus Br. 6-30, Gill v.

Whitford, No. 16-1161. It also means that approval of

plaintiffs’ claim is unlikely to encourage new litigation; it will, instead, lead litigants to continue to

bring challenges, but under a more accurate and apt

legal framework.

G. The Court cannot and should not affirm on

alternative grounds

Citing cases that do not involve the abuse-of-discretion standard (Br. 55), the State insists that the

Court can affirm on any ground apparent in the

record. That is incorrect.

“[W]ith respect to a matter committed to the

district court’s discretion, [an appellate court] cannot

invoke an alternative basis to affirm unless [it] can

say as a matter of law that it would have been an

abuse of discretion for the trial court to rule otherwise.” Manion v. American Airlines, Inc., 395 F.3d

428, 431 (D.C. Cir. 2004). This generally means that

a reviewing court will not affirm “on any basis other

than that articulated by the district court.” Parsi v.

Daioleslam, 778 F.3d 116, 126 (D.C. Cir. 2015). We

made this point in the opposition to the motion to

affirm (at 7), but the State again talks past our argument without response.

In any event, the State’s alternative grounds are

insubstantial. For the most part, the State rehashes

its merits arguments (Br. 55-59), insisting that we

failed to prove intent, causation, and burden—issues

we have already addressed.

The State also cites our supposedly “extraordinary delay,” faulting us for not filing for a preliminary

injunction earlier and complaining that there is now

inadequate time for relief. State Br. 59-60. But as we

explained in the reply in support of the motion to

expedite (at 2-3), seeking injunctive relief was not

possible with respect to earlier elections because of

the convoluted procedural history of the case. And a

preliminary injunction became necessary with

respect to the 2018 election only as a result of the

State’s own dogged refusal to cooperate in discovery.

Id. at 3.

In any event, the question of delay and whether

there is now adequate time for relief is a discretionary one for the district court in the first instance.

It is assuredly still possible to correct the 2011 gerrymander in time for the 2018 elections—pursuant to

2 U.S.C. § 2a(c) if by no other means. See Branch v.

Smith, 538 U.S. 254, 275 (2003) (plurality) (describing 2 U.S.C. § 2a(c) as “a last-resort remedy to be

applied when, on the eve of a congressional election,

no constitutional redistricting plan exists and there

is no time for either the State’s legislature or the

courts to develop one”).

CONCLUSION

The Court should vacate the order below and

remand for further proceedings.

Respectfully submitted.