Evenwel v. Abbott – Oral Argument – December 08, 2015

Media for Evenwel v. Abbott

Audio Transcription for Opinion Announcement – April 04, 2016 in Evenwel v. Abbott


John G. Roberts, Jr.:

We’ll hear argument next this morning in Case 14-940, Evenwel v. Abbott. Mr. Consovoy.

William S. Consovoy:

Mr. Chief Justice, and may it please the Court: This appeal presents a fundamental question. That question is whether the one-person, one-vote rule affords eligible voters any reasonable protection.

We submit that the answer must be yes under this Court’s decisions, and as a consequence, Appellants have stated a claim under the the Equal Protection Clause. The districts at issue here, District 1 and District 4, have deviations as measured by eligible voters approaching 50 percent under any metric available to voters.

No decision of this Court has ever sustained vote dilution of that magnitude under a one-person, one-vote case.

Beginning with Gray, continuing to Wesberry, through Reynolds, and the Court’s many decisions since then, the issue has always been vote dilution.

Ruth Bader Ginsburg:

What about the many times the Court has said that the — the principle is equal representation of the population? And we have had now, for half a century, population — that the population is the — the legitimate standard.

We have never held to the contrary. So we have the States overwhelmingly for half a century using population as shown in the census, and now you’re saying they can’t do that anymore.

William S. Consovoy:

I can answer the question in — in three different ways. First, with respect to the phrase “equal representation for equal numbers of people.” That sentence originated in Wesberry.

But that’s only half the sentence.

There’s a dash, and it continues: “Therefore, for us to hold that within the States legislatures may draw the lines of congressional districts in such a way as to give some voters a greater voice in choosing a congressman than others would be unconstitutional.” And in fact, in every time the Court uses that phrase, which is the only one, I believe, my friends rely on, it is either within the same sentence or bracketed on one side or the other by protection of voters. Now, as to tradition, to — actually, to the word “population,” we don’t — we — we see that as asking the question, not answering it.

Burns explains that Reynolds used population without distinguishing. Burns itself reserved on the question.

Hadley confirms that Burns reserved on it.

And here we are today.

Ruth Bader Ginsburg:

I thought — I thought Burns said it approved a deviation from population but it took great pains to say, we’re not saying you could do that in every case.

Burns seems to be the only — the only case that you have, and Burns involved this really peculiar situation of Hawaii with a tremendous military temporary population.

William S. Consovoy:

I — I read Burns as reserving on it completely, to not choose one way or the other.

It certainly did say that you do not have to use the — the census to draw districts.

That supports our position.

It certainly says that you can protect eligible voters.

That supports our position. And — and further back to Your Honor’s question about tradition, if tradition were the rule, Baker would have come out the other way.

Before Baker, for centuries, geography was the basis.

And the Court said in Baker, as a matter of jurisdiction, and then in Reynolds as a matter of — of equal protection law, that tradition doesn’t trump the individual rights of a voter to be protected. And we don’t have to guess about that because we know from standing.

In all of those cases, standing was predicated on the right of the voter. It would be unusual if someone who couldn’t vote came to this Court and says — said my one — a child, for instance — my one-person, one-vote rights have been violated.

Sonia Sotomayor:

The problem is that — what you’re forgetting is the dual interest.

There is a voting interest, but there is also a representation interest, and it’s that which has led us to — to accept the total population base because States have to have some discretion to figure out who should be having the representational voice. Burns made it very clear that we were deferring to the State because it had a legitimate reason for its need. And — but Burns was in the 1960s, when we picked total population as a perfectly legitimate way because there’s a representational need at issue as well.

Not just voting.

A State has to be able to say — I think just as the Federal government did — we’re — the legislature is protecting not just voters; it’s protecting its citizens — or noncitizens.

The people who live there.

William S. Consovoy:

So if I can just clarify: It’s not really a representational interest that’s being claimed on the other side.

They — a non-voter will be — there’s 31 senate districts in Texas.

William S. Consovoy:

A non-voter will have one representative under our rule, and they will have their one representative under theirs.

It’s an access claim that’s being made.

That’s what the Garza opinion from the Ninth Circuit said. And it’s not even really an access claim. It’s a diminishing access claim.

That’s how far from voting the interest on the other side goes.

It is that, if we have districts that are overpopulated with non-voters, we will have diminishing access to our representative. We don’t deny that access is an interest, along with county lines, along with other interests that the State can take into consideration, and the 10 percent framework allows for that.

This is not a situation where we are here complaining about a deviation of 15 percent or 10.1 percent.

We’re complaining about a deviation of nearly 50 percent.

No interest such as diminishing access could ever overcome the individual right of a voter to an equal vote. And if you —

Elena Kagan:

Mr. Consovoy, of course it’s true that when we apportion House members, we use total population as the metric.

And the question that’s raised by your position is why it would be the case that the Constitution requires something with respect to one apportionment that it prohibits with respect to another.

William S. Consovoy:

Apportionment and intrastate districting are fundamentally different concerns. Apportionment at the time of Article I’s framing was focused on taxation issues, on giving States autonomy with respect to voter qualifications.

And there was a real concern.

That’s why it was a — the great compromise. What the Court held in Reynolds, as a matter of equal protection, is that that compromise does not justify this kind of injury. And we don’t have to guess about this either.

In — in Reynolds, Alabama came to the Court and said, “semi” we surrender.

How about a plan that mirrors the House precisely? Every county gets one representative, and the rest is done on a population basis.

Not only on a population basis.

The precise formula used for the House of Representatives.

Reynolds said no.

Sonia Sotomayor:

What’s interesting is in Reynolds is the reason they caved was because, constitutionally, the Arizona Constitution required total population.

It’s that fact that they deviated from their own Constitution that led them to court.

So it wasn’t a caving compelled by Federal law.

It was mostly a casing compelled by State law.

William S. Consovoy:

Well, the — not that this Court found the case, because it wouldn’t — couldn’t be before this Court on a State law ground.

It could only be before this Court —

Sonia Sotomayor:


But what I’m saying is that we — we acceded on the ground that using total populations was permissible.

William S. Consovoy:

There’s no doubt that Reynolds thought, in that case, total population.

The Court in Reynolds thought it was permissible. Of course, Baker, the Tennessee constitution apportioned unqualified voters, and there was no suggestion in Baker that that was an additional problem with the Tennessee constitution. But the fundamental issue has always been the individual right.

And I think Gray is the best case to show that. Gray is about election for statewide offices, so it can’t be a representational issue. There’s going to be one governor.

That governor is — everyone is going to have the same access to that governor or not.

William S. Consovoy:

There are no districts. And yet Gray is the case that establishes this rule.

Gray is the case that says, voters are entitled to an equal vote.

You can’t marry up the representational interest that’s asserted on the other side with Gray.

It doesn’t — it doesn’t make any sense. We also know from — with respect to congressional districting, as late as 1969, in Kirkpatrick, the Court assumed, for purposes in that case, that Missouri could district at the congressional level on the basis of eligible voters.

So I don’t think it would be fair to say that this issue has somehow been clouded or decided by uses of the words “population” or using the census in prior cases.

Elena Kagan:

I’m — I’m sorry.

Did I just understand you to say that you think that the — the House apportionment rule is not clear?

William S. Consovoy:


The — that congressional districting intrastate at the Federal level as opposed to the State level. So if — if 1969 Kirkpatrick v. Chrysler says — because the United States has argued that not only is apportionment required at the Federal level but intrastate congressional districting.

And my point — they — and they rely solely on Westbrook for that — excuse me — Wesberry for that proposition. My point is in 1969 the Court flatly rejected Wesberry as having decided that issue.

There is no decision of the Court that resolves this question. It is — it is completely open. And the only way to make sense of the one-person, one-vote rule is to make it about eligible voters.

They are the ones who have standing.

They are the ones who can bring a claim.

They are the ones who are injured.

And not only is that our view and the case law’s view, that was — that was Congress’s view.

Ruth Bader Ginsburg:

Is it your view that what the Fourteenth Amendment means is that in all the years between — what was it? — 1868 and 1920, it was wrong for the States to include, for these purposes, women? They were not eligible voters.

William S. Consovoy:

Any — there is no question that was a problem.

It was an — it was an issue in the ’60s with minorities as well who were — who were disenfranchised.

The — the Court in Reynolds at the time was doing more than one thing at once.

Ruth Bader Ginsburg:

But you’re saying that that was wrong.

I mean, in your interpretation of the Fourteenth Amendment from 1868 till 1920, the State should not have been counting women for — for purposes of determining representation in the State legislature.

William S. Consovoy:

For purposes of the — of the Equal Protection Clause, the one-person, one-vote rule protects voters.

If disenfranchisement of women or minorities is an issue, those cases could have been brought.

Eventually, that issue was resolved by this country, as was minority representation. But the Warren court in Reynolds was accomplishing several things.

And equal weight for voters has to matter.

A noncitizen or any other disenfranchised person would not have the ability to bring a one-person, one-vote claim.

Stephen G. Breyer:

Yes, but here we have — I want to go back to Justice Kagan’s question.

And this is something that — it seems everyone is arguing this is as if this is an equal protection problem.

And certainly, the — Reynolds v. Sims does deal with equal protection.

And it did deal with instances in which voters and everything else were malapportioned.

Stephen G. Breyer:

So I don’t think the court really considered this. But if you step back from the Equal Protection Clause and say there are other parts of the Constitution that, in fact, are relevant here — maybe it’s the Republican Form of Government Clause.

But the words that Justice Kagan read are words about what kind of democracy people wanted.

And those words say if you look to other parts of the Constitution, such as those, or Republican Form of Government, that what we actually want is the kind of democracy where people, whether they choose to vote or whether they don’t choose to vote, are going to receive a proportionate representation in Congress. And if you take that as a constitutional principle, that shows an objective of some of the clauses of the Constitution.

Then you have to retreat from the idea that the Equal Protection Clause, as interpreted in Reynolds v. Sims, solves this case.

And indeed, it argues against you.

William S. Consovoy:

So two answers. One is to argue that this is justiciable on the other side as a Guarantee Clause claim, I think shows just how far the logic has to go to come up with something on the other side of the ledger here.

This Court has never — in fact, in Baker, the Court rejected the Guarantee Clause as a basis for hearing these cases. To turn around now, and turn 180 degrees, so that a — a somewhat abstract Guarantee Clause claim, that —

Elena Kagan:

Well, Mr. Consovoy —

Stephen G. Breyer:

I’m not making it.

I’m making — I’m pointing at her — she didn’t quote the Guarantee Clause.

Maybe I shouldn’t have thrown that in.

But it’s the same point.

William S. Consovoy:

If I could — but Reynolds, I think, does speak to this, because that exact theory would be the one that would have sustained the model that Alabama brought to the Court, that — that followed the House of Representatives, which does take account of these issues. And even if, Justice Breyer, even if you’re correct, that theory is correct, that doesn’t solve this case.

We have alleged in our complaint that Texas could have done much more to — to manage both representational equality, as it’s called, and voting equality to get both within 20 percent.

To say that we have not —

Sonia Sotomayor:

You had an expert say it, but you didn’t have an expert prove it.

He did not come in with a map that did that.

William S. Consovoy:

We did not come in with a map. One, we’re still at the motion-to-dismiss stage in this case, so our allegation, which is at paragraph 22 of the complaint, has to be taken as true. But second, the reason we didn’t come to a map is fundamental here as well.

We don’t want the Court or ourselves to write this map for Texas.

We want the Texas legislature to do its job. Texas, by State law, was precluded from taking voter equality —

Sonia Sotomayor:

How is — does practical possibility play into this discussion at all? I mean, the ACS, which you posit is the way that you can find who the eligible voters are, has been — has been — I think almost decisively been proven as being inadequate. It only measures cities with populations or places with populations over 65,000.

Just on that ground alone, there are going to be districts that can’t rely on it.

William S. Consovoy:

I — I think — sorry.

Sonia Sotomayor:

It’s flawed on many levels.

We could — take my assumption.

It’s flawed. Does that practicality have any play in our decision?

William S. Consovoy:

I will take the assumption and then I will try to argue again against it, if you — you’ll allow me. But practicality, if the Court were to hold that — two different questions.

As the Court explained in Tennant just recently, it’s our burden to bring in evidence showing that total population census didn’t protect individual rights. We believe — at this stage, we have alleged it, so it has to be taken as true.

If we can’t prove it, then we have failed to meet our burden. It would be a different question if the Court held yes, the evidence you brought in is sufficient to show that your rights have been violated through the ACS data, but not sufficient to draw a new map.

Then I think the Court is in a very unusual place, where I think the answer is we’re back to Baker, which would then become — that would be the argument of the dissent in Baker, which is you have a violation but no judicially manageable way to solve it, so now we’re back to political question. But if I could get back to the fundamental premise, which is the ACS data, the ACS data — I think Your Honor was talking about the 1-year ACS data, but States for redistricting used the 5-year ACS data.

William S. Consovoy:

That matters — measures populations going down less than 3,000 people.

The only group it doesn’t have is individual block group data, and that data is rarely used for districting that we’re talking about here. Moreover, we know the ACS data is good because it’s used in Section 2 every day, and not just for a vague and general purpose.

Under Strickland, under Bartlett against Strickland, to bring a successful Section 2 claim, you have to show that you have a majority of the citizen voting-age population in your district to — to get through the first factor for Section 2. That means if there’s 50.1 percent minority eligible voters in your district, you can proceed; if it’s 49.9 percent, you cannot.

This data is used to determine that question. If it can do that in every circuit court in the country — and in this Court’s opinions in LULAC and Strickland supported using this data for that purpose — if it can do that, it can bring a deviation of 47 percent to somewhere between 10 and 20. And I don’t want to leave this abstract.

If you look at the Supplemental Appendix, the data is actually in there.

If you turn to page 5 of the Supplemental Appendix, there is a column called “CVAP” and it lists all of the CVAP numbers for every Texas Senate district.

I would point out that Texas asked for these numbers to draw this map.

They used CVAP to draw this map. If you pull those numbers and look at District 1, it has 557,000 people.

Right next to that is the plus or minus with numbers.

It says 6,784. That’s the margin of error.

That’s the margin of error for CVAP data. So if you took all of those margins of error and used them against our position at every turn — so for under-populated district, assume up; for overpopulated district, assume down — assume it at every turn against us, and you ran the numbers, it would move the deviation from 47 percent to 45 percent. This is not an issue about margin of error, about data.

This is not an issue about the availability of data.

This data is used if — by every demographer to draw statewide districts at every turn.

Elena Kagan:

Mr. Consovoy, could I go back to the question that Justice Breyer raised and can — stripped, if he’ll permit me, of the Guarantee Clause, because the Fourteenth Amendment is actually quite — you know, the framers of the Fourteenth Amendment explicitly considered this issue, and, you know, made a decision. So Senator Howard, who introduces the Amendment on behalf of the joint committee that drafts it, talks about these deliberations.

And he says the committee adopted numbers as the most just and satisfactory basis, and that’s the principle upon which the Constitution itself was originally framed, referring back to the original drafting.

And then he says numbers, not voters; numbers, not property; this is the theory of the Constitution. Now, this is the theory of the Constitution as to one thing, which is not the thing that you are talking about.

This is the theory of the Constitution as to House apportionment. But again, I’ll go back to this question. This is just a clear, explicit choice that was made about what it meant to — to have equal representation with respect to that area.

And how you go from that being mandated to it being prohibited in the State context is something that I still can’t quite work myself around.

William S. Consovoy:

Justice Harlan agreed with you.

He did.

Elena Kagan:

That’s a good person to be on the side of.

William S. Consovoy:

Yes. (Laughter.)

But his — his position was rejected 8-1 in Reynolds.

Because that exact argument was brought forth by Alabama.

They presented a plan that was not only somewhat generally modeled on an apportionment standard, it mirrored it exactly.

So I think there are reasons why that’s not correct as a — just a legal matter, because apportionment was concerned with many other things.

They wanted the States to have taxation basis.

They wanted — there was an issue with suffrage, for sure.

There was an issue with voter qualifications.

It was a complex, Federalism-based, sovereignty comprise that does not apply within a State. I can’t do any better, I apologize, than say Reynolds —

Elena Kagan:

I hear you as to that it does not apply.

I mean, I guess I can — I can understand.

I might not agree with, but I can understand the position that says that the requirement might not apply.

But you are suggesting that we go beyond that, and to say, not only does the requirement not apply, but that States have to do it the exact opposite way.

William S. Consovoy:

So, two answers. We take our cue on that from the right that is supposed to be protected.

It all follows from the right, and it starts with voting.

We start with the proposition that one person can’t be given two votes while their neighbor be given one vote, and from there the Court moved in Gray to say, well, you can’t do it by calling it weighting under some sort of electoral college model.

That’s the same thing. Then the third step was you can’t accomplish that same invidious voter discrimination by drawing lines.

Now if you accept all of that as true, that I can’t be given five votes and my neighbor be given one, then even if it follows from the apportionment model and you — and you defend it on that basis, if it causes that injury, I have a claim.

And to say that I don’t have a claim because a different constitutional provision protects a different right in a different way, I — we find, you know, not a satisfactory response beyond which Reynolds itself rejects — rejects the argument. And turning back to Section 2 for a moment, Congress agreed with our position.

Congress relied on Reynolds being a vote-dilution case to enact Section 5, not only in the Senate report that’s been so widely used in those cases, but in this Court’s opinions as well, in Perkins and in later cases.

The Court has held that — and — and in Bolden, both the plurality and Justice Marshall’s dissent, said Reynolds is about vote dilution. And Section 2 — the same argument could be made, Justice Kagan, about Section 2.

Section 2 only counts eligible voters.

No one argues that we’re discriminating against non-voters by not taking them into consideration.

If the Court were to proceed with that kind of representational model, we would have one rule that minorities get the — excuse me — the benefit of under Section 2, and no protection for people who are not minority status under one-person, one-vote.

There is a — there is a real fundamental disconnect there.

Ruth Bader Ginsburg:

In — in your view, the States would have a choice between the citizen voting-age population or they could use the registered voters? Either one would be okay?

William S. Consovoy:

Well, we start with the proposition that Burns said.

It’s not the method by which you distribute legislators that count.

It’s the distribution of legislators that count. Therefore, as Burns explains, the State can truly use any metric that adequately and fairly distributes legislators.

We think registered voters is not ordinarily going to be the right one for two reasons: Gray says those who hold the one-person, one-vote right are those who meet the basic qualifications of voting.

So there, registered voters run into trouble. And then Burns essentially doubles down on that argument by saying it depends upon political activity. And where we’re drawing lines at essentially the beginning of the game, we shouldn’t make the right depend upon who ends up deciding to enter the fray and choose to vote. So we think the data that we principally rely on, the — the ACS measure of citizen voting-age population is ordinarily going to be the fairest and most accurate measure.

But that’s for the legislature to decide when it — when it reviews all this information. And the Texas legislature — I think it’s important to keep in mind when they drew this map, they did everything that we’re asking to be done here.

They took all of this data — the total census data, the citizenship data, the registered voter data, the — the precinct data — and they put it all into a computer. And they drew their districts. And they used our data to draw districts in this map.

They just used it to comply with Section 2 and when — and then closed their eyes and didn’t want to look at — to see what kind of deviations it caused for one-person, one-vote. All we’re asking the legislature to do is open its eyes.

Sonia Sotomayor:

Do you think they did that invidiously? Did they do it purposely?

William S. Consovoy:

Well, under one-person, one-vote, a deviation over 10 percent, as the Court just recently heard —

Sonia Sotomayor:

They knew that, and they intentionally decided to have deviations greater than 10 percent? That’s what you’re saying?

William S. Consovoy:

I don’t think we can know — we’ll ever know because they were handcuffed by State law.

There was an attorney general interpretation from 1981 that precluded Texas from considering voter eligibility.

William S. Consovoy:

It — so it’s really arbitrary —

Sonia Sotomayor:

That goes both back to my point that they decided that they wanted to make this a representational matter.

But do — so they were precluded — intentionally decide to exclude it?

William S. Consovoy:


Under the one-person, one-vote rule, a deviation exceeding 10 percent, we establish, as we — we argue as a matter of eligible voters, itself is prima facie evidence of invidious —

Sonia Sotomayor:

Well, we have — we have plenty of case law that says you can have deviations greater than 10 percent — Hawaii did — if you have a legitimate reason.

And so why would — the great representational need that Justice Kagan was talking about not be an adequate reason?

William S. Consovoy:

We think it is a — a reason to go over 10 percent.

That’s — we do not want the perfect to be the enemy of the good on this issue.

We understand that things need latitude.

We are asking for nothing more than to bring them within the 10 to 20 percent range that the Court has always held.

Sonia Sotomayor:

You’re — you’re now — you’re now saying 10 to 20 percent is okay instead of 10 percent when we use total population.

William S. Consovoy:

The — the Court has up — held up the 16.5 percent, and — and Mahan 20 percent as the outer limit.

We take our cues from those.

Stephen G. Breyer:

Suppose Texas said here, we want children to be represented? That’s all.

Children. See, they’re not voters.

So suppose — if we take children out of it, what’s the deviation?

William S. Consovoy:

We haven’t examined it.

We only examine on the basis of eligible voters.

But children are represented at the polls.

They’re represented at the polls by their parents.

If there are parents here who have been disenfranchised, they were disenfranchised by the State.

States like California and Texas and New York have —

Sonia Sotomayor:

Well, how about —

William S. Consovoy:

— have —

Sonia Sotomayor:

How about children who are citizens when their parents are not, which is fairly common in many areas?

William S. Consovoy:

And — and when — when they become eligible voters, they will move into this base. They are not counted for Section 2, and don’t — haven’t heard any argument that Section 2 discriminates against children. If I might, Mr. Chairman, reserve the balance.

John G. Roberts, Jr.:

Thank you, counsel. General Keller.

Scott A. Keller:

Thank you, Mr. Chief Justice, and may it please the Court: The only question the Court has to resolve here is whether the Equal Protection Clause requires every State to change its current practice and use voter population to reapportion.

The answer is no. Texas validly used Federal census data to equalize total populations States have done for decades. And the framers of the Equal Protection Clause accepted total population as a permissible apportionment base in Section 2 of the Fourteenth Amendment, as Justice Kagan said earlier.

John G. Roberts, Jr.:

Well, why don’t they use that under Section 2, then?

Scott A. Keller:

In Section 2 of —

John G. Roberts, Jr.:

The Voting Rights.

Scott A. Keller:

The Voting Rights Act?

John G. Roberts, Jr.:


Scott A. Keller:

Section 2 of the Voting Rights Act for tax voters — and our position, unlike the United States’s position, is that only voters are protected under the Voting Rights Act.

So in considering whether there is an opportunity to elect a candidate of one’s choice, only voters would count for that inquiry.

Indeed I —

John G. Roberts, Jr.:

Well, it is — it is called the one-person, one-vote.

That seems to be designed to protect voters.

Scott A. Keller:

It does protect voters, but there are multiple legitimate bases here on which a State can redistrict.

Electoral equality is one of them.

Representational equality is another. And if I can back out, what we’re dealing with here is the general Equal Protection Clause’s test which — it guards against invidious discrimination. The Court has noted before a mere disparate impact does not violate the Constitution.

And so really the claim that’s being alleged here is one of invidious vote dilution. But Texas, by using total population, as States have done for decades, and no State today uses voter population, does not invidiously target groups to cancel out their voting power or reduce their ability to elect representatives of their choice. Rather, what Texas was doing was making the legitimate choice to use representational equality, which, as this Court’s cases have noted, is a legitimate interest that the State can count for in redistricting. What the State cannot do is submerge the population principle.

In other words, as Reynolds held, we cannot base apportionment on geography.

We have to take account of population.

And we have done that. There is no allegation here that our 8.04 percent deviation of total population would not satisfy the Court’s one-person, one-vote doctrine unless we are required to use voter population.

Samuel A. Alito, Jr.:

There are at least two arguments that could support your position.

One is that it’s one-person, one-vote, and what counts is giving each person an equal chance of affecting the outcome of the election.

But total population figures are a good enough proxy for eligible voters.

That’s one possible argument. And that’s — that’s what the census measures, and that’s close enough. Another argument is that representational equality is the real basis, and therefore that’s why you use population. So which argument are you making?

Scott A. Keller:

I don’t believe we’re making either of those arguments, Justice Alito. Total population is not permissible because it tracks voter population.

At the same time, while the Court doesn’t have to reach this question, representational equality is not the only basis on which a State can redistrict. It’s our position that we could choose a reliable measure of voting-eligible population without running afoul of the Equal Protection Clause’s guaranteed against invidious discrimination.

Samuel A. Alito, Jr.:

It seems to me that the two interests are not always consistent.

They can be in great conflict. You can have a situation if you — if you want to equalize population, you may have a situation where you cause great inequality in the — the chances of any — of voters affecting the outcome of the election.

On the other hand, if you choose eligible voters only, then you may have a situation where every person within two districts does not have an equal representation defined in some way in the legislature. I don’t think you can just say, well, it’s — you know, we serve both.

What do you do when they come into conflict?

Scott A. Keller:

I believe what this Court said in Burns is you allow the States to choose the theory of representation.

And indeed, the decision to include or exclude non-voters, Burns said, was left to the States, because part of what this Court’s doctrine has recognized is States need leeway, and that this is a core sovereign function.

It is part of the dignity of State sovereignty to be able to structure elections. And when a — when a State is choosing either representation or equality, when the two are intentioned, that’s not an illegitimate basis upon which to reapportion.

Anthony M. Kennedy:

Well, if the voter population is a permissible basis under the Constitution, I assume that’s because there is — is an ethical, a good government, a liberty interest in protecting these voters.

That’s a valid interest, correct?

Scott A. Keller:


Anthony M. Kennedy:

Well, if in a case like this where there is a 45 percent deviation, something of that order, then why isn’t Texas required at that point to recognize that these interests that are legitimate under the Constitution, which are voter based, should not be accommodated, and so that you should at least give some consideration to this disparity that you have among voters?

Scott A. Keller:

Well, first off, the court in Gaffney upheld the use of total population while recognizing that there could — there was, in New York at least, a different State, a 29 percent deviation in voter population.

Yet the court there said it was quite sure that a prima facie case in invidious discrimination had not been made out. And so while a State can, and legitimately does consider both representational equality and electoral equality, the Equal Protection Clause’s general language doesn’t mandate that either must take precedence over the other. So of course it would be legitimate for the State to look at that data.

At the same time, when we have Federal census data, which is the most robust data set available, it is not invidious for Texas to use that enumeration rather than a different data set when it reapportions.

And all we have under the census data is total population data.

Stephen G. Breyer:

What we have — and that’s why I think Justice Alito’s question is important — is a table on page 9 of the Blue Brief. Now, just looking at that table, by inspection, I don’t know whether the true — whether this is true or false.

So I thought the major difference between the two here is probably that some areas of — of the State — there are a lot of people who are working and they have children.

I mean, it can’t all be explained on the basis of illegal immigration or something.

It just can’t be, given those numbers.

I don’t think so. And if we accept the principle that it’s voter equality, we are saying that the family of two of certain age that has eight children or whatever is getting no representation for those other people or human beings.

And if we accept the opposite, we have to put up with inequality of — of power of voters.

You have to — you have to say the one or the other.

And you could take your position it’s up to the State. But I mean, that seems to me to be what’s actually behind the numbers that he’s — that — that are being quoted, but I’m not sure.

So I’d like your reaction.

Scott A. Keller:


Justice Breyer, I believe there’s a difference between diminishing access to representatives and actually having representation. The United States has said that if Texas or another State reapportioned on the basis of voter population, that non-voters would be invisible to the system. That’s not right.

They would still be represented. The issue is does State — does a State have to have the same amount of constituents per representative? And a State can do so.

It’s a legitimate —

Stephen G. Breyer:

That sounds an awful lot what they had in 1750 or something, where the British Parliament said, well, don’t worry, America, you’re represented by the people in England because after all, they represent everybody in the British Empire.

Scott A. Keller:

Which is —

Stephen G. Breyer:

I mean, that people are being represented through somebody else is a little — possible, but tough.

Scott A. Keller:

Well, for instance, a child in my congressional district would still be represented by that member of Congress.

So the issue is — really is diminishing access to the representative.

And while that’s a legitimate basis for a State to reapportion under, there is no Equal Protection principle that would elevate that as a rule of constitutional law that would say that the Sate of Texas invidiously discriminated.

Anthony M. Kennedy:

But why is one option exclusive of the other? Why can’t you have both? You have population equality and voter quality, both, especially when you have indicated that a voter-based apportionment is — is valid and serves important purposes.

And here it’s being completely — it’s being very substantially disregarded with this huge deviation. Why can’t you use both?

Scott A. Keller:

Well, first of all, there’s been no demonstrative plan that was submitted to the Texas legislature, which has a notice-and-comment procedure on this, or to the district court, that both of these could have been equalized within 10 percent. Indeed, their demographer didn’t specify the extent of the deviations.

Scott A. Keller:

Their demographer simply said, well, the deviations can be reduced. If the Court were to try to go down the road of requiring States to equalize within 10 percent of a deviation, both total and voter population, States would inevitably have to disregard many other traditional redistricting factors, like compactness, continuity, keeping communities together.

And that would be the opposite of what the Court has said that States have in this context, which is the leeway to structure their elections as part of the core function of their sovereignty.

Anthony M. Kennedy:

That sounds highly probable to me. Has anything been written on this, or any studies on this —

Scott A. Keller:

I — I — I don’t —

Anthony M. Kennedy:

— in — in the context of Texas.

Scott A. Keller:

I — I — I don’t believe so. We’re not aware of any.

And we’re also not aware that this would be practically feasible.

Indeed, if they had a plausible allegation that this was possible, we would have expected to see a demonstrative map at this phase in the litigation.

John G. Roberts, Jr.:

Do you have any idea how often this is a problem? I mean, it is a case that, of course, around the country, people use total population.

But it seems to me that there will be a lot of areas where, in terms of the actual numbers, it’s not going to make a difference. Do you have any idea?

Scott A. Keller:

I believe New York’s amicus brief suggests that in places such as California, Alaska, possibly New York, certainly New York City, the issue will absolutely come up.

However, even —

John G. Roberts, Jr.:

But only in — only in those particular handful of —

Scott A. Keller:

A few more examples —

John G. Roberts, Jr.:

And I’m not suggesting —

Scott A. Keller:

— Delaware, Maryland. By no means would this necessarily be a problem anywhere.

However, if there were a rule that a State had to consider voter population, that would change the nature of redistricting.

John G. Roberts, Jr.:

Well, what if it were you had the same minor or de minimis deviation allowed there? In other words, if you’re within — the deviation between total population and voter population was within — under 10 percent, does that take care of many of the areas where it’s a problem?

Scott A. Keller:

Well, Mr. Chief Justice, for the reasons I just suggested to Justice Kennedy, I believe that would be quite an onerous burden and change the nature of redistricting.

Could there possibly be a situation out there in which a plan might be able to get within a 10 percent deviation of total population and a 10 percent deviation of the five-year rolling average sampling in the American Community Survey? Maybe.

I’m not aware of any such scenario. And to back out to first principles, I don’t believe that would be a test of invidious discrimination.

That would be moving much further in the direction of a disparate, impact-like test that the Court has never fashioned to determine whether someone’s voting power is being canceled out, or is fenced out of the political process.

Sonia Sotomayor:

Could you explain why the ACS — your adversary says ACS is fine; it’s used in Section 2 and Section 5.

Why would it be inappropriate to use it in — in deciding the impact on an equal voting analysis?

Scott A. Keller:

Well, our position is that if the ACS data is reliable enough to hold the State liable under Section 2 of the Voting Rights Act, it would also be reliable enough to use in apportionment. Now there could be issues about the granularity of the data.

For instance, the five-year averages.

We do get at census-block level, which is about five to 600 to 3,000 people. The smaller you would get for district levels, depending on if you were at a city plan as opposed to our State Senate plan.

There may be issues where you couldn’t use the data to get within the 10 percent-deviation. But certainly in larger districts, like the Texas State Senate plan, you could use the five-year CVAP data, and you could do that to get within the 10 percent deviation.

Of course, we’re not constitutionally compelled to, because as the Court recognized in Burns, that is up to the States in choosing a legitimate population basis. If I could briefly address the United States’s argument on Section 2 of the Voting Rights Act.

We disagree on this point. Section 2 of the Voting Rights Act does not protect non-voters.

Scott A. Keller:

And earlier when we discussed Section 2, I’d like to return to that to cite to the Court the Persily amicus brief of page 26, because I think this cuts against the United States’s theory on Section 2. Persily brief says, “If the minority group has very low rates of citizenship, then the redistricting plan is not to blame for their lack of representation.

Rather, their lack of sufficient voters is.” So the United State’s suggestion that there could be packing or cracking claims of communities that have nothing to do with packing or cracking voting blocks, that is an incorrect interpretation of Section 2.

It is not consonant with the tax, and it would render Section 2 unconstitutional, as not congruent and proportional with the right to vote that is being protected. If the Court has no further questions, thank you, Mr. Chief Justice.

John G. Roberts, Jr.:

Thank you, General. Mr. Gershengorn.

Ian H. Gershengorn:

Mr. Chief Justice, and may it please the Court: Redistricting on the basis of total population, as Texas did here, vindicates the principle of equal representation for equal numbers of people that is at the heart of Reynolds and Wesberry.

We thus agree that Texas was not required to redistrict on the basis of some as-yet-undefined measure of voter population. However, we disagree that the Court should go on to decide that Texas is free in the future to redistrict on the basis of some measure of population — voter population if it so chooses. There are, in our view, at least four reasons why voter population cannot be required. First is the one mentioned by Justice Kagan. We think it would be a very odd interpretation to say that the Constitution forbids for State legislative redistricting what it requires for congressional redistricting. Second is the very long history of States — of States redistricting on a basis other than — other than voter population.

At the time of the framing of the Fourteenth Amendment, there were the vast majority of states redistricted on other-than-voter population. In the wake of the Fourteenth Amendment, Congress in the apportionment acts required districting on the basis of inhabitants.

And, of course, over the last 50 years, States have unanimously redistricted on the basis of total population, not voter population. Third is the — is — are the data problems, and they are real.

The ACS data has a number of limitations.

First of all, it is not constitutionally required, unlike the census.

It would be very odd, we think, for the Court to demand, as a constitutional standard, data that does not even have to be collected. Second, it does not measure what the — what the plaintiffs suggest is required.

It is not a measure of voter eligibility.

CVAP does not include — citizen-voting-age population data, for example, does not include felons.

It does not include overseas voters.

It does not include the mentally ill.

That kind of data just does not exist. And third with respect to the data, picking up on Justice Sotomayor’s point, the data in the ACS does not exist at the level of granularity, accuracy, and timeliness needed to redistrict.

To be clear, the data level does not exist at the census block level, which is where districting happens.

It is not issued on a timely basis.

The census data comes out, for example, in April 2021.

The ACS data, the five-year average comes out in December, and it has data from 2016 going forward.

John G. Roberts, Jr.:

That’s often used for Section 2.

Ian H. Gershengorn:

Your Honor, it’s used for a very different purpose with respect to Section 2.

In Section 2 it is used as one factor among many to determine whether electoral opportunity has been — has been given.

So it’s used along with — with population data, voting data, turnout data, a whole variety of socioeconomic factors. That’s very different than this Court saying every State and thousands of local jurisdictions throughout the country have to use that data as the sole measure for redistricting. I’d like to then pick up on Justice Kennedy’s question.

Samuel A. Alito, Jr.:

Well, can I ask you this? Who has standing to bring a Reynolds v. Sims claim? Is it anybody who is counted in the census?

Ian H. Gershengorn:

So Your Honor, that’s a question this Court noted and reserved in Baker v. Carr in footnote 23.

And it’s a question that this Court has never had to resolve in the context of Wesberry, which of course has the exact same rule. We think that nothing — not much turns on it because, as a practical measure, you can get a voter. You can always find a voter in the district.

But let me explain why we don’t think it, sort of, is dispositive here, and this goes to a number of those issues —

Samuel A. Alito, Jr.:

Are you going to tell me who has —

Ian H. Gershengorn:

— we’ve heard this morning.

Samuel A. Alito, Jr.:

Are you going to tell me who has standing or not?

Ian H. Gershengorn:

Yes, Your Honor.

We think that it is — we think that the — the better understanding is that a non-voter would have standing. But I — here’s why I don’t think it matters: Because you can view our position as through either lens, through a representational lens in which what’s happening is that the Reynolds v. Sims right is a way to ensure that all persons covered by the Equal Protection Clause who can’t — even those who cannot cast a ballot still have a voice in representational —

Samuel A. Alito, Jr.:

That includes everybody who is counted in the census.

Ian H. Gershengorn:

Yes, Your Honor.

Samuel A. Alito, Jr.:

It includes —

Ian H. Gershengorn:

But let me say —

Samuel A. Alito, Jr.:

It includes aliens.

It includes prisoners.

Ian H. Gershengorn:

And let me —

Samuel A. Alito, Jr.:

It includes undocumented aliens.

Ian H. Gershengorn:

But let me explain why I don’t think it’s necessary.

Samuel A. Alito, Jr.:

But does it include all those groups?

Ian H. Gershengorn:

I’m sorry, Your Honor?

Samuel A. Alito, Jr.:

Does it include members of all of those groups?

Ian H. Gershengorn:

So we think it might, but we don’t think that you have to agree with that to rule in our way.

Because we do think that the right at Reynolds is also viewed, as we said on page 14 of our brief — and we think this is important — as a voter right.

The way to think about this, as Reynolds did, was to view this as a right — consistent with the way Reynolds thought about it, was to say that, when you have twice the representatives in — twice the — the inhabitants in a district, you get half the vote. What Reynolds said, picking up on Plaintiff’s counsel’s position, was that of course it would violate the Constitution to count somebody’s vote as two or five or ten times.

But then what it said in the next sentence: “Of course, the effect of State legislative districting schemes which give the same number of representatives to unequal number of constituents is identical.” That is exactly the point we’re making here. And if I could pick up, then, on Justice Kennedy’s and the Chief Justice’s point about why can’t you do both. The reason is very much — and we agree with General Keller that the problem with doing both is that it — it largely eliminates a State’s flexibility to deal with the traditional redistricting factors.

What you are forced to do is take a large, for example, Anglo population in one part of the State that has high citizen rates and pair it with the situation where it has — with — with populations that have low citizenship rates in another part of the State. Or to take an example from the amicus briefs, Manhattan has 9 percent children.

Brooklyn has 30 percent.

If you have to do both, what you’re doing is pairing people from the — from part of Manhattan and — and pairing them with part of — of voters in Brooklyn.

What ends up happening is to do both at the level of 10 percent is to eliminate a State’s ability to take into account things like political subdivisions, to take into account compactness, and all of the other things that this Court has said is critical in redistricting. As to the Chief Justice’s question about whether this is a big deal or not, or whether it’s — “big deal” isn’t the right word — whether it would have a large practical effect — I would assume it’s a big deal.

That’s why we’re here. (Laughter.)

— as to whether it would have a large practical effect.

I think the answer to that is yes. What we’re talking about is not just 50 States but thousands of jurisdictions around the country, local jurisdictions, none of whom use voter population as a measure for redistricting. What the amicus briefs show is this is not just a situation in which things are affected — States are affected where there are citizenship differences between citizens and noncitizens, but that children actually are a critical part of it.

It’s not just that Manhattan is 9 percent and Brooklyn is 30 percent children.

In Texas, the counties range, the amicus briefs suggest, from 9 percent in some counties to 35 percent in other counties.

In Alaska the difference between rural and urban is 20 percent children in some and 30 — 37 percent in another. This is an issue that is going to affect States and local jurisdictions throughout the country. And local jurisdictions, to be clear, don’t have the data at the level and — at the level in which this Court would now be requiring as a constitutional matter. Now, I’d like to pick up on one other point that Plaintiff’s counsel raised, which is that, in his view, it’s quite unclear as to what Wesberry actually holds.

We think that that is really a fundamental misreading of Wesberry. What Wesberry said was — the whole point of Wesberry was that the — the method of apportioning or allocating representatives to the States had to be the same as the method for allocating within districts in a State.

Ian H. Gershengorn:

That was the reason that — when what Wesberry said was that the great compromise had to be reflected into — into redistricting. That principle in Wesberry was exactly the principle that then the Court adopted in Reynolds, what — what the Court said in Reynolds.

It was Wesberry that clearly established the fundamental principle of representative government in this country as one of equal representation for equal numbers of people, without regard to race, sex, economic status, or place of residence within the State. So it is precisely that same principle from Wesberry which looked at the — looked at the — looked at the framing and looked at the discussion of — of calculation of representatives at the framing, which looked at the drafting of the Fourteenth Amendment, and took that history, and then translated that —

Samuel A. Alito, Jr.:

Isn’t your argument that voters are — are irrelevant?

Ian H. Gershengorn:

So Your Honor, I don’t think our argument is that voters are irrelevant.

And first of all — so a couple of points on that. First, of course, the question here is, when Texas has chosen to use total population, is that permissible? And we think that clearly is. Second, we don’t think voters are irrelevant for the reasons that I’ve said.

The — what Reynolds did was — and the Reynolds line of cases was to use total population to vindicate the voters’ right.

It is a voters’ right — because Reynolds understood that, when you have twice the inhabitants in the district, you have half the — half the voice before your representative.

Samuel A. Alito, Jr.:

What would you say about the extreme case — I mean, your time is going to going to expire — an extreme case.

And maybe this would never come up, but what if it did? Suppose you have a district — you have a rural district, and suppose it’s a State where the — the total number of — the total population per district is — is fairly small.

You have a rural district with a huge prison and very few other inhabitants.

So you — and you have a neighboring district that has no prison. So in one district, you have that 10 percent of the population are eligible voters; and the other district, 90 percent of the — the population are eligible voters.

That would be okay?

Ian H. Gershengorn:

So Your Honor, two points in response. First, this Court has recognized — and we don’t dispute — that census data isn’t the sole data. A State can — and this Court approved it in Mahan — make adjustments to census data to more accurately capture actual residents in the State.

We think that’s what has been happening in Hawaii, and that’s what Mahan said. Remember in Mahan, what the State had done — what Virginia had done was count all of the Navy personnel as home ported, which is what the census had done.

And the Court said you have to make an adjustment to that. Thank you, Your Honor.

John G. Roberts, Jr.:

Thank you, counsel. Mr. Consovoy, you have four minutes remaining.

William S. Consovoy:

Thank you. Justice Breyer, to your question about representation of children.

If that were the principle of Reynolds against Sims, in a statewide election, a State could give five votes to a family of five and one vote to an individual. That would —

Stephen G. Breyer:

I’m just thinking that I’d like to know, before knowing whether this is mandatory or not, your position.

I’d like to know an awful lot more than I know about who these people are who are being represented on the representational theory, and who are not being represented on the voter theory.

William S. Consovoy:

In each —

Stephen G. Breyer:

Now I don’t know who they are from the briefs; and therefore, it’s pretty tough for me to — to tell.

William S. Consovoy:

The data shows that it’s a mix of noncitizens, children, all the categories; disenfranchised felons.

It’s a — it’s a mix.

There are children involved, of course. But — but our point is more fundamental. If —

Stephen G. Breyer:

Illegal immigrants.

William S. Consovoy:


But who have not — who the State has chosen not to allow to vote.

The State can solve this problem themselves.

These States can enfranchise these people and give them the vote. The States come here to say we do not want them to vote, but we want them to count for districting.

William S. Consovoy:

That should be rejected by this Court. Second —

Sonia Sotomayor:

That’s not quite accurate.

For — for most States, too many, they disenfranchise prisoners, except for those who come from that locale, which is quite rational.

Most States disenfranchise the mentally ill.

So how are they — who else are they going to disenfranchise?

William S. Consovoy:

I’m not suggesting — we’re not suggesting we should choose for the State who they allow to vote.

We are — we are arguing that we should not allow the States to come to this Court and argue that they should get the benefit of them counting when they make the choice, that is their right, to disenfranchise them. You cannot disconnect this rule from voting and allow it to stand up.

The whole thing collapses. Wesberry has the famous sentence now that says all other rights are illusory if the right to vote is taken away. That’s — the authors of that sentence would be surprised to learn that the one-person, one-vote rule has literally nothing to do with voting; that you could have a system that crowds, in 31 Senate districts, all eligible voters but 30 — 30 of them into one, and give each other person their own district.

That plan would be sustainable, absent some evidence of racial or political discrimination. The State comes in to say we know we can’t do it, but we’ll never try.

That’s not how one-person, one-vote works.

The State-by-State law forced themselves not to try this by saying they weren’t allowed to.

If they were told by this Court that they could at least — to your point, Justice Kennedy — do both, they would go back to the drawing board and try. If they failed, they may win that case.

We suspect, and we have alleged, so it must be taken as true, that they can do both. And Justice Kennedy, it will not be traditional interest like districting, or county lines, or anything like that will be — that will inhibit them. It is political and racial gerrymandering that they want to do, and that our rule, and especially a rule balancing both, will stop them.

And we don’t know that abstractly; we know that from the case the Court heard just before us. The actual deviations in Arizona — the hypothetical case — they are claiming it’s an 8 percent deviation.

On page 26 of their own jurisdictional statement, they concede that the CVAP deviations are 54 percent.

And in District 8, the district mostly at issue, is underpopulated by 22 percent. If Arizona had to go back to the drawing board with the Districting Commission and accommodate at least voter, but at least — or maybe both, there would be no opportunity to engage in the political and racial gerrymandering that has come to dominate the redistricting process.

That would not involve the Court in those issues anymore.

It would be solved legislatively, as they should. Section 2 does not work without this understanding.

As Justice Scalia pointed out in his dissent in Chisom v. Roemer, there is nothing to measure against if one-person, one-vote doesn’t protect voters. It’s the baseline. How do you know if minority vote dilution has occurred unless you have a baseline to measure against? The baseline is equal voting power of voters absent discrimination.

It completely unravels.

John G. Roberts, Jr.:

Thank you, counsel. The case is submitted.