Franklin v. Massachusetts – Oral Argument – April 21, 1992

Media for Franklin v. Massachusetts

Audio Transcription for Opinion Announcement – June 26, 1992 in Franklin v. Massachusetts


William H. Rehnquist:

We’ll hear argument this morning in No. 91-1502, Barbara Franklin, Secretary of Commerce v. Massachusetts.

Mr. Roberts.

John G. Roberts, Jr.:

Thank you, Mr. Chief Justice, and may it please the Court:

This case is here from the United States District Court for the District of Massachusetts.

That court ruled that the Secretary of Commerce could, consistent with the Constitution and the Census Act, include Federal personnel stationed abroad, primarily servicemen and servicewomen, in the State-by-State apportionment count on which the allocation of representatives is based.

The court went on, however, to hold that the Secretary’s method for doing so with respect to the 1990 census was arbitrary and capricious, in violation of the Administrative Procedure Act.

The Secretary relied primarily on home of record data in Defense Department files, as had been done in 1970.

Home of record is a serviceman’s actual home at time of enlistment and it is where the military will return him when he is discharged.

Based on–

David H. Souter:

Mr. Roberts, may I interrupt you at this point.

I think you have made the point clear, but it doesn’t seem to be clear throughout the briefs.

I take it, it is correct that when a person enlists, that person has no choice of which State to list his home of record.

It has got to be the State in which the person then legally resides?

John G. Roberts, Jr.:

–The instructions say you must list your permanent home or actual home at time of enlistment.

David H. Souter:

So that the suggestion that somebody could simply list a State whose tax laws he found attractive is not true?

John G. Roberts, Jr.:

That is not true.

Taxation is treated with a different concept, legal residence.

Someone can have a home State in one State and declare a legal residence for tax purposes elsewhere.

But the suggestion that home of record is subject to manipulation for tax purposes is not true.

I would refer the Court to the joint appendix, page 151 where that is clarified.

Now, based on its finding of a violation, the district court ordered the executive branch defendants not to count the nearly 1 million Federal personnel stationed abroad in the apportionment count at all.

It ordered that a new statement be sent to Congress specifying the number of representatives to which each State is entitled, and it ordered the clerk of the House to send to the governors new certificates telling them how many representatives their States were entitled to.

If allowed to stand, the decision below will shift a representative from the State of Washington to Massachusetts, only the second time in the 200-year history of the census that any such relief has been ordered.

The first time was last fall with the district court’s decision in the Montana case which this Court reversed last month.

The decision here should meet the same fate: first, because judicial review is unavailable under the APA; second, even if review is available, the Secretary’s decision to allocate personnel abroad to their home of record is not arbitrary and capricious; and third, even if there is a violation of the APA, the appropriate remedy is a remand to the Secretary, not judicial reapportionment of the House of Representatives.

Sandra Day O’Connor:

Mr. Roberts, as I understand the respondents in their briefs, they also make a claim that the Constitution itself is violated, and I suppose that would have to be resolved apart from any APA claim.

John G. Roberts, Jr.:

We believe, Your Honor, that the constitutional claims are also barred by the commitment of this matter to the Secretary’s discretion.

We recognize that this Court in Webster against Doe indicated that there may be a higher level that needs to be shown.

But in Webster, the constitutional claim was a personal right, extrinsic if you will, from the matter that Congress had insulated from judicial review.

Here, that is not the case.

John G. Roberts, Jr.:

The constitutional claim and the statutory claim are in large part indistinguishable.

Sandra Day O’Connor:

Well, what does the Constitution mean in requiring an enumeration and in referring to the whole number of persons?

What do you suppose that meaning is?

John G. Roberts, Jr.:

Well, assuming there is review, an enumeration is exactly what took place in this case.

The Secretary didn’t estimate the number of servicepeople abroad.

It didn’t take a sample and then extrapolate from that.

The Secretary counted them, 918,810, not 809 and not 811.

Now, the Secretary counted them from forms, but that is how the census is conducted by and large.

When you get a census form in the mail, it asks you to fill it out, or the head of the household, and state how many people are living at that home.

You list yourself, a spouse, three children, the form is returned.

The Census Bureau counts from that form, but that is still an actual enumeration.

And indeed, historically, it was considered sufficient, for example in the case of a prison, to ask the warden how many prisoners were there, or significantly, in the case of a military base, to rely on data provided by the commander.

And even if, in the normal case, if the Census Bureau has trouble reaching someone, they will go and ask the neighbors, who lives there, how many people?

It is still an actual enumeration.

It is not an estimate.

The other constitutional provision to which you referred, the original clause, the respective numbers of the States, the Fourteenth Amendment in each State, I think both sides agree that what the framers are referring to are inhabitants.

And the question is, is it rational to conclude, using the words of the first Census Act and the Census Clause, that servicemen abroad can still be considered as belonging to the respective numbers of their home States.

I think the answer to that is clear.

Madison knew what an inhabitant was.

He was involved in drafting the Census Act, and he said it includes someone who is absent for a considerable period on public business, certainly an apt description of our troops abroad.

It was tested in the case-of Minister Forsythe in 1824, our minister to Spain who was elected from Georgia to the House.

His election was challenged.

It said, you are not an inhabitant of Georgia.

You have been in Spain, and the House said, his absence abroad on public business did not deprive him of the status of an inhabitant, and the same should be true of a private serving overseas in Korea.

The review of both the constitutional claims and the administrative law claims is barred by the Census Act.

The APA says that if a matter is committed to agency discretion by law, there is no review.

Sandra Day O’Connor:

Well, what if the Secretary were to do some wholly arbitrary act like counting people who live in Massachusetts as part of the population of the State of Washington?

John G. Roberts, Jr.:

I think in any case in which the argument is that judicial review is precluded, you can come up with hypotheticals that seem to call for corrective action.

Sandra Day O’Connor:


John G. Roberts, Jr.:

But that is not enough to establish that the exception to review doesn’t apply or otherwise it would never apply.

John G. Roberts, Jr.:

For example, there is an exception to judicial review this Court recognized in Webster.

Now if the director of the CIA decided that every left-handed person was a threat to national security and dismissed them, that would be absurd, but it still would not be subject to review.

One point that is important to consider in discussing this is how likely is it that the absurd or extreme case will go uncorrected.

We think that this is a situation in which that is extremely unlikely.

Congress pays attention to the conduct of the census very carefully, too carefully I think according to my brother, and they certainly pay attention to the reapportionment of the House of Representatives.

This is a situation where the political branches can be expected to correct for extreme cases of the sort you hypothesized.

Anthony M. Kennedy:

–But under your theory of the case, it is not necessary to do so, but what do you think the agency action would be here?

John G. Roberts, Jr.:

Well, I am not sure.

The only conceivable final agency action is the Secretary’s report to the President of the tabulation of population by States, which certainly doesn’t provide a suitable case for judicial review.

All that is required is a document stating Massachusetts has 15 million people–

Anthony M. Kennedy:

And in your view, is the President authorized to correct that?

John G. Roberts, Jr.:

–Absolutely; he has the obligation to take care that the law be faithfully executed.

He could decide, he could be, say I have always believed that the usual residence rule should never be departed from and I disagree.

Send me some statistics without apportioning servicemen abroad.

He has that obligation.

Anthony M. Kennedy:

So then I guess the next possibility is the President’s transmittal of the data to Congress as the agency act?

John G. Roberts, Jr.:

Yes, except that the President is not an agency subject to suit under the APA, and I don’t understand Massachusetts to challenge that submission here.

The next step would be Congress, and that is how the process has traditionally operated.

The President submits his statement.

Congress has a period in which it can consider that statement and take legislative action if it wishes.

It did not decide to take any action in this case.

The action they took years ago in the Census Act was to specify that the conduct of the census, the Secretary will conduct the census in such form and content as he may determine, pursuant to the constitutional grant to Congress to conduct the census in such manner as they shall by law direct.

Antonin Scalia:

Mr. Roberts, I know the President is not covered by the Administrative Procedure Act, but do you really think that that means that there is no review whatever of Presidential actions?

I mean, can the President do anything at all and you say it is not reviewable because the APA doesn’t cover him?

John G. Roberts, Jr.:

Well, he–

Antonin Scalia:

I agree, he is not reviewable under the APA, but might he not be reviewable in some other fashion?

John G. Roberts, Jr.:

–Theoretically, in a purely ministerial task, by mandamus, but this is not a purely ministerial task, and there is no suggestion that there is any other basis for suit here, other than the APA.

So his part in this process, which is key, the Secretary’s statement doesn’t entitle a State to anything.

It is the President’s statement by law which entitles a State to that number of representatives unless changed by Congress.

Antonin Scalia:

When the APA refers to agency action, don’t you think it means action binding upon private individuals somehow?

Antonin Scalia:

I always thought that is what it meant.

Can it be the filing of a report with somebody else, even the filing of a report by the President with Congress, does that constitute agency action?

John G. Roberts, Jr.:

Not in the sense, for example, as the Court discussed in the Luhan case, it does have to have an effect.

The Secretary’s report of population doesn’t affect legal rights of any individuals or of the States.

It is the President’s statement that gives the States an entitlement to a certain number of representatives, and yet that is insulated from review under the APA.

Now the district–

John Paul Stevens:

I am a little puzzled by it; how far can the President depart from what the Secretary’s report would show?

Could he just say, I think I will give Massachusetts three extra seats without any factual basis or anything, just I think that is a good State and they ought to have three more seats.

John G. Roberts, Jr.:

–The law directs him to apply, of course, a particular mathematical formula to the population figures he receives, but I don’t think there is a limit on his exercise of authority to direct the Secretary of Commerce to conduct the census in a particular manner.

It would be unlawful, maybe not subject to judicial review, but unlawful just to say, these are the figures, they are right, but I am going to submit a different statement.

But he can certainly direct the Secretary in the conduct of the census.

John Paul Stevens:

But would he have to remand it in effect to the Secretary or could he say, well, I have had somebody over at the FBI making some checks for me and they tell me there are really more people in Massachusetts, so I am going to give them extra seats.

John G. Roberts, Jr.:

I think under the law he is supposed to base his calculation on the figures submitted by the Secretary.

John Paul Stevens:

But what if he didn’t, what if he did what I said?

John G. Roberts, Jr.:

Well, again, the manner of review is established in the statute.

His statement goes to Congress where it stays until Congress by subsequent statute changes the entitlement in his statement, that is spelled out in the law, and that indicates that that is the mechanism for review, not judicial review of any of the thousands upon thousands of judgments that went into the underlying census figures.

Now the district–

Anthony M. Kennedy:

If by law the President’s action were final, would that be reviewable?

John G. Roberts, Jr.:

–No, it would not because he is not an agency subject to review under the APA.

Anthony M. Kennedy:

But the President can’t seize the steel mills or withhold tapes either, but his actions are subject to review in those instances.

John G. Roberts, Jr.:

But not pursuant to the Administrative Procedures Act, pursuant to another–

Anthony M. Kennedy:

Well, but your position, he is not subject to review at all.

John G. Roberts, Jr.:

–I think that is right, because the statutory scheme which indicates his statement is to go before Congress–

Anthony M. Kennedy:

But my hypothetical was that his action was conclusive.

John G. Roberts, Jr.:

–It is conclusive unless reviewed by Congress, and my submission is that by providing for a special mechanism for congressional review, and the historical fact that that is how Congress has reviewed it, indicates an intent to preclude review of that interim step in the transmission process.

Now, the district court in this case relied heavily on the usual residence concept for its law to apply, the concept that people should be counted for purposes of the census where they live and sleep most of the time as of census day.

Historically, there have been exceptions to that principle time and time again.

Boarding school students are not counted where they live and sleep most of the time, they are counted at their family home.

The same was true for college students from 1900 to 1940–

Antonin Scalia:

Before you get… just one more thing about this series from the census to the President to the Congress.

Antonin Scalia:

You say if Congress doesn’t like what the President gives them, the Congress doesn’t have to accept it.

Is there any way that Congress can not accept it except by passing a new law?

John G. Roberts, Jr.:

–No, the requirement is for them to pass a new law.

Antonin Scalia:

Both houses, plus the concurrence of the President?

John G. Roberts, Jr.:


Now that is a way, of course, that Congress can change anything, by passing a law.

Antonin Scalia:


John G. Roberts, Jr.:

However, they have specifically provided in the census and in the Apportionment Act for a mechanism, and it is a historical mechanism, for this to lie in Congress and to be addressed xx a particular period so that they can take action.

Antonin Scalia:

But no single house of Congress can just look at this stuff from the President and say this is junk, and we are not going to proceed on this basis.

John G. Roberts, Jr.:

No, it has to be, as the statute says, a subsequent statute that has to be enacted.

I was indicating the numerous exceptions to the usual residence rule.

Six censuses, from 1870 to 1920, seafarers, not matter how long they were absent, counted at the land home.

The military in 1900 counted at their family home even if stationed as the instruction said, at home or abroad.

Members of Congress are counted at their home district, even though they live and sleep most of the time in the District of Columbia.

From 1910 to 1940, the instructions said, if there is an American citizen in your family abroad, he should be counted at your family home, and it said, it does not matter how long the absence may continue.

In other words, the usual residence concept cannot provide law to apply in considering this case, because this is a departure and modification of the usual residence rule, as has been historically done in a variety of circumstances including, of course, most prominently in 1970 when again, Federal personnel abroad were apportioned among the States based on home of record.

Now turning to the merits, the district court concluded that the Secretary’s decision to apportion the personnel overseas was not arbitrary and capricious in itself, and it was not.

The record indicates the reasons for the Secretary’s action.

These personnel are away from home temporarily and involuntarily in the service of their country.

They still consider themselves usual residents of the United States, although they are overseas.

The Secretary noted the bipartisan congressional support for apportioning the overseas population, and in turn, that support noted these personnel vote in the United States, they pay taxes in the United States.

Now, the Secretary did note in his decision memorandum that the Defense Department was going to undertake a survey, an actual survey of its personnel overseas, but contrary to the view of Massachusetts, that was not a condition of his decision.

If the Court would look at the joint appendix on page 122, it spells out exactly what the condition is.

This is the decision memorandum approved by the Secretary.

It says: I recommend that you count them, provided that there is timely receipt of acceptable data from the Defense Department.

Now, when the Defense Department cancelled its survey, that question became pertinent.

Is there acceptable data that can be provided?

And the Secretary determined that home of record, supplemented and improved, did provide acceptable data.

David H. Souter:

Mr. Roberts, is your argument undercut by the colloquy that Massachusetts sets out on pages 12 and 13 of its brief, which as I read it indicates that counsel for the Government, counsel for your side are saying that the only so-called justification to remove the decision to count based on home of record from the realm of the arbitrary and capricious is the mere fact that Mr. Mossbacher had made a commitment to Congress to count them.

Which seems to say that regardless of the… we are not claiming that it is the substantial accuracy of the home of record data that rescue us; it is simply the principle that we ought to keep our word.

John G. Roberts, Jr.:

I don’t think that undercuts our position.

First of course, representations or statements of counsel are not part of the administrative record, and what needs to be reviewed is the administrative record.

But secondly, all that the counsel was saying is that the Secretary had made a commitment.

He had made a decision, the commitment referred to is simply the decision in the administrative record.

Yes, we are going to apportion the overseas personnel if, if there is acceptable data.

So the commitment that the counsel in the trial was referring to is no different than that in the decision memorandum in the record.

Now, is home of record acceptable data?

The Secretary, who has the discretion to conduct the census in the form and content he may determine, determined that it was.

The technical staff of the Census Bureau said it comes closest to usual residence.

It is the last place that a serviceman or woman chooses to reside.

The Congressional Research Service, whose study was in the record, indicated that it is closest to a serviceman’s own concept of his usual residence, an important point, since much of the census is self-selecting.

Now, in the course of reaching that decision, concerns were raised about the data.

This is not a sanitized, pre-cooked administrative record.

It is not an administrative record at all in the usual sense of a discrete proceeding leading to a decision.

It is part of the Census Bureau’s business over the past years, but every one of the concerns that Massachusetts raises, every one of the concerns cited by the district court was addressed and answered in the record.

Let’s take them: First, the Department of Defense testified in 1989 that none of its data met a reasonable test of validity for apportionment purposes.

Answer: joint appendix, page 161, the data will be supplemented and improved by methodologies devised by the technical experts who designed the overseas enumeration.

Second, there was a gap in home of record.

Up to 10 percent of servicemen did not have a home of record.

Answer: joint appendix, page 158 and 162, the Defense Manpower Data Center would use its automated records to complete, fill in the gap, to the extent that it will be negligible.

It went from 7 percent to 0.2 percent.

The Census Bureau itself said that the data was of unknown reliability because it could be used to choose a low or no income tax State.

Answer: joint appendix, page 151, it is legal residence that determines where you are taxed, not home of record.

The Pentagon opined that last duty station would be closest to the concept of home of record.

Answer: joint appendix, page 162, the Census Bureau disagreed.

It noted, for example, if your last duty station was the Pentagon, you would show up as living in D.C., and you may have lived in Virginia or Maryland.

And finally, the point that we hear most frequently, the Census Bureau itself said that it would be, quote, arbitrary, end quote, to rely on the Defense Department data to apportion Federal personnel overseas.

Answer: joint appendix, page 210, that same study relying on legal opinions in 1949 and 1969 noted that the decision whether to include Federal personnel overseas was discretionary.

So they were obviously not using arbitrary in an APA sense, but simply in the general sense that any categorical determination is arbitrary.

Now the question is not… Massachusetts may not agree with these answers, but the question is not whether they were right or wrong.

John G. Roberts, Jr.:

The question under this Court’s precedence, assuming there is judicial review, is whether the relevant concerns were considered and if the Secretary committed a clear error of judgment.

The record establishes that the concerns were addressed, were considered, and it seems clear that there was not a clear error of judgment in allocating the personnel overseas to their home of record.

If there are no further questions, I would like to reserve the remainder of my time.

Antonin Scalia:

You have a lot of time.

I have one more question.

When you say that the President’s action is… is it that the President’s actions are not reviewable or that you simply cannot bring a suit directly against the President?

Suppose I am in jail by direct order of the President.

You don’t mean that I can’t get habeas corpus simply because the action of the President is not reviewable, do you?

John G. Roberts, Jr.:

No, you can file a suit for habeas corpus against the prison warden.

Antonin Scalia:

Against the prison warden.

The action would have to be against the warden, but the President’s action would be reviewable.

John G. Roberts, Jr.:

Would be reviewable to the extent it is the warden’s action, yes.

But the only point I am making is that there is no action against the President under the Administrative Procedure Act, that is the District of Columbia court of appeals, the Federal court of appeals held recently, and as I said, I don’t understand Massachusetts to challenge that.

And since there is no review of the President’s action under the APA, there is no review of his action in this case, and since his action is indispensable to harm to Massachusetts, the case is not justiciable.

Thank you.

William H. Rehnquist:

Very well, Mr. Roberts.

Mr. Golann, we will hear from you.

Dwight Golann:

Mr. Chief Justice, and may it please the Court:

The Secretary has argued that he has virtually unbridled discretion to conduct the decennial census without review by any court.

We disagree.

The apportionment census is carried out pursuant to specific constitutional and statutory mandates.

The Constitution and the statutes which mirror its language require three things of the apportionment census.

First, it requires an actual enumeration.

That is, the Secretary must actually count the population on a State-by-State basis.

Second, it must be a decennial census.

The Constitution requires that the census be repeated every 10 years, so the Congress has a current, up-to-date statement of the distribution of the population among the States on which to base apportionment decisions, and finally, as I think the parties agree, the apportionment census must be based solely upon the number of inhabitants living in each State.

The possibility of considering citizenship, voting population and other criteria was explicitly considered but rejected.

All of these criteria… actual enumeration, decenniality, inhabitancy… must be met for an apportionment census to comply with the Constitution.

William H. Rehnquist:

Well, do you think the word actual in the Constitution changes the meaning of enumeration and gives that noun a meaning it wouldn’t otherwise have if the word actual weren’t in front of it?

Dwight Golann:

We believe so, Your Honor.

Dwight Golann:

We believe that enumeration has been historically understood and applied consistently except for overseas employees in 1970 and 1990 to require, as the Seventh Circuit recently observed, a head count of the population, not a statistical sampling or estimation.

I would note–

William H. Rehnquist:

Well, but if you take the language leading up to it in Article 1, section 2, and it talks about representatives and direct taxes shall be apportioned among the several States which may be included within this Union according to the respective numbers which shall be determined… determined at… and this is the actual enumeration… it seems to me you can use the word… read the word actual as simply indicating when it actually comes to enumerating rather than modifying the word enumeration the way you say it.

Dwight Golann:

–I believe, Your Honor, that was required, that there be an enumeration within 3 years and then each term of 10 years afterwards, and that it would be an actual enumeration seems to have been understood by the contemporaneous interpretation of the framers, who in fact mandated that there be a head count, and dis–

William H. Rehnquist:

How did they mandate that there be a head count?

Dwight Golann:

–Through the first Census Act of 1790 passed by the First Congress in March 1990, which was implemented by a head count, actual census-taking of the entire population.

William H. Rehnquist:

And you say because that was the First Congress they probably reflected what the Constitution intended.

Dwight Golann:

We believe so, Your Honor, and we note that that was continued decade after decade.

It has been a head count, as the courts of appeal have noted, ever since then except 1970 as to overseas employees and 1990 as to overseas employees.

Antonin Scalia:

In what respect is this not a head count?

This isn’t an estimate.

They are using actual–

Dwight Golann:

Well, we believe, Your Honor, that it is an estimate as to where people are residing, which is the point of an apportionment census.

Antonin Scalia:

–It’s where they said… they said they resided on the basis of the last information we had.

Dwight Golann:

Your Honor, we believe that the requirement of a decennial census requires an update every 10 years.

We believe, and I might answer Justice Souter’s question next, that the Defense Department, the stipulations, and the court’s opinion, all say that that… this home of record data might be a place where the employee had not resided for many years and in some cases had not resided at all, so it was not an accurate statement.

The entire rest of the United States population in 1990 was asked, who resides at this address?

Who lives here most of the time?

That question was never asked of overseas Federal employees.

They were not allowed the answer that question.

The parties have stipulated that if they did answer it, they would have answered that their usual residence is their overseas post, circulation–

David H. Souter:

Well, on that theory, then, I cannot be counted as a resident of New Hampshire if they happen to take the census during the term of this Court.

Dwight Golann:

–Well, Your Honor, from the first days… in fact–

David H. Souter:

Do you take it that far?

I mean, unless somebody is sort of literally there and they can see–

Dwight Golann:

–No, Your Honor.

David H. Souter:


How far do you go from literalism, then?

Dwight Golann:

Someone should be counted at their usual place of abode, was what the First Congress said, and if they are temporarily absent they should be counted at the place they usually reside.

Most of the population and the particular overseas employees we have stipulated resided at their overseas post.

Dwight Golann:

There is a subset of the population that splits their time between two residences.

Seafarers, for example, have a home ashore and a berth aboard ship.

College students are another example.

They’re saying, as the First Congress said, you must be apportioned at your usual residence, doesn’t answer the question of which of two places that you live is your usual residence.

That requires line-drawing.

I would emphasize there was no line-drawing in this case.

William H. Rehnquist:

Well, what exactly was it… what is the actual language used by the First Congress on which you’re relying?

Dwight Golann:

Your Honor, I believe that is set forth in the joint appendix.

Let me find the quotation.

William H. Rehnquist:

Is that the grey… the grey volume.

Dwight Golann:

No, Your Honor, in the joint appendix–

William H. Rehnquist:

The brown volume.

Dwight Golann:

–I believe there is a direct quotation.

Actually, Your Honor, it is also quoted in the lower court’s opinion in the jurisdictional statement at 59(a)… jurisdictional statement appendix… that every person whose usual place of abode shall be in any family on aforesaid first Monday in August next shall be returned of such family, and below it says every person occasionally absent, for example on business with the Court, Mr. Justice Souter, at the time of the enumeration shall be returned as belonging to that place in which he usually resides in the United States.

So usual residence, the parties have stipulated, has been the guiding principle now for 200 years, and usual residence we have stipulated is the place where a person is living and sleeping most of the time.

That was in fact line 1 of the census instructions for 1990, as they appear in the joint appendix.

Antonin Scalia:

Yes, but that all depends upon what you mean by occasionally absent.


Dwight Golann:

That is true, Your Honor.

Antonin Scalia:

If occasionally absent includes absence of years, then most of the time means most of the time when he’s not absent, and that could mean many years ago, couldn’t it?

Dwight Golann:

Yes, Your Honor.

In the age of sail or horse, journeys were, of course, much longer than they are today.

The criteria that has been consistently applied is, if someone is away on a trip and they haven’t established a dwelling any place else, then even if they’re gone for a long time, they have one dwelling that is at their home.

On the other hand, as we have stipulated, overseas employees had a dwelling, a place where they usually resided, and it was their overseas post.

That is why, for example, the domestic militar, are counted where they are.

Antonin Scalia:

Why is their overseas post their dwelling any more than the Spanish Ambassador’s residence was his dwelling?

I don’t understand why that’s the dwelling.

Dwight Golann:

First, Your Honor–

Antonin Scalia:

They’re assigned there, just as the Spanish Ambassador is assigned there.

They have to be there, and when they’re out of the Army they expect to go back to where they reside… to New Hampshire or wherever.

Dwight Golann:

–First, Your Honor, because that is where they are living and sleeping most of the time.

Secondly because we have stipulated that that is their dwelling.

As to ambassadors, there is a special term of art.

The actual ambassador residence is deemed to be territory of the United States just as the French Ambassador’s residence here is deemed to be territory of France.

Antonin Scalia:

Oh, so all we have to do is call U.S. bases abroad… we can deem them to be territory of the United States and you’ll be happy.

Dwight Golann:

No, Your Honor, because I believe that would violate the guiding principle of usual residence.

Antonin Scalia:

Well, I mean that’s a fiction.

Surely it has nothing to do with the meaning of the Constitution.

Also, we grant certiorari in cases to review undecided questions of law.

We don’t regard ourselves as bound by any sort of a stipulation that could affect a question of law.

Dwight Golann:

Well, Your Honor, I would suggest that the Court has said, for example, in the Universal Camera case that it may rely on the findings of the district court, and the district court in this case may find–

William H. Rehnquist:

Yes, but you have constantly talked about stipulations.

If you’re talking about a finding of the district court which we would not upset unless it’s clearly erroneous, that may be something different.

Dwight Golann:

–Excuse me, Mr. Chief Justice, the reason that I referred to it is because it is the facts which the Justice Department and the Massachusetts Attorney General agreed would provide a record for the Court to make a decision, and so we have relied on them.

Some of those stipulations differ, for example, from some of the statements in the briefs, and so we would ask the Court to at least review the stipulations which we submitted is the basis for a decision in this case… we jointly submitted.

Let me–

John Paul Stevens:

May I ask you one other question about the 1790 act?

It doesn’t just say the place in which he usually resides.

It’s the place in which he usually resides in the United States.

Dwight Golann:

–That’s correct, Your Honor.

John Paul Stevens:

Are you saying that language applies even though he doesn’t usually reside in the United States?

Dwight Golann:

Your Honor, I would suggest that if someone is absent on a journey, as I am in Washington today, and if it were the census date today… my usual residence is in Massachusetts–

John Paul Stevens:


Dwight Golann:

–But if I am living abroad, I’ve rented out my home or sold my home and I have–

John Paul Stevens:

I understand your argument, but I just don’t think it is supported by the language, usually resides in the United States.

Dwight Golann:

Well, Your Honor, I believe that this was designed for people who are occasionally absent on a journey and that is the way it has been interpreted for 200 years.

That is the way, for example, the Census Bureau interpreted in 1990 in its official instructions where it had said that you should count people that are usually living in the residence but not people that are temporarily on a journey, on a vacation, in the hospital, on a business trip.

That has been the consistent interpretation of this usually… occasionally absent language.

The issue is whether or not they establish a dwelling at the place where they travel on their journey and give up their dwelling where they were living before.

–Let me note that there is also statutory law to apply in this case for purposes of reviewability.

Dwight Golann:

Constitutional language is mirrored in statutes.

The Apportionment Act particularly requires that there be a census, it be carried out on a decennial basis, and that quote… it provide, quote, a statement showing the whole number of persons in each State, end quote, mirroring the language of the 14th Amendment.

Now, I should note that the lower court never actually ruled on the constitutional issue or the statutory issue because it found the Secretary’s action arbitrary and capricious.

The final decree contains no finding on a ruling as to constitutionality.

In 1990, the entire resident population of the United States, including the domestic military, I might add, was apportioned based on questionnaires asking them where do you live most of the time?

Overseas Federal employees were not apportioned on a basis consistent with the entire rest of the United States population.

Instead, they were apportioned to States based on this home of record file data.

As noted, the Census Bureau concluded in 1987 when formulating official rules for the census that using file data would be, quote, arbitrary, end quote, and the Department of Defense said in 1989 to Congress that neither home of record nor any of its other file data, quote, meets a reasonable test of validity for apportionment purposes.

The problems with home of record which were identified by the lower court are also what create the constitutional violation in this case.

Home of record is the State listed, for whatever motive, by an employee when he or she first enters the military.

The Department of Defense prohibits the employee from ever changing their home of record while they remain in service, regardless of how their family circumstances, personal circumstances may change.

The data may be years or decades old, much more than 10 years old.

The Department of Defense testified to Congress in August 1, 1989, that the home of record may be a place where the employee lived many years ago, and in some cases a place where the employee never lived at all, and that was the testimony of the Department of Defense.

Mr. Roberts is correct as to the rule or the question that was asked, but the answers, as the Department of Defense testified, varied quite a bit, and people were apportioned where they had never lived.

When the Secretary decided to change historic practice and apportion–

John Paul Stevens:

Excuse me, and you say that’s not constitutional?

That is not constitutional–

Dwight Golann:

–That’s correct, Your Honor.

John Paul Stevens:

–Because I’ve never lived there.

Suppose I’m a college student and my parents moved their residence while I am in college, and right after college I go abroad for several years, and I list my residence as my parents’ new residence.

Is that unconstitutional?

Could I not be allocated to that new residence, although I’ve never lived there?

Dwight Golann:

I understand, Your Honor.

The test of inhabitancy is where you live and sleep most of the time.

Where that student in your hypothetical had never lived or slept at that place… might intend to reside there in the future but had not ever resided there and was not residing there on the census date, so he could not be apportioned to that State under the test set forth by the Census Bureau.

I might note that when the Secretary did decide for the first… second time to apportion overseas employees back to States, his sole factual basis for doing so was the Department of Defense promised an actual enumeration, but the Department of Defense cancelled that promise.

At that point, the lower court said the Secretary had a legal obligation to at least consider an important alternative.

The alternative… that requirement, rather, the court said, comes from this Court’s decision in Motor Vehicle Manufacturers, both the majority and concurring opinions.

The primary important alternative that the Court said the Secretary ought to have considered was returning to the historic practice, which was to count overseas employees as Americans, but not apportion them back to States where they were not living.

The Secretary did not consider any alternatives.

Dwight Golann:

As indicated, he treated the issue as solely one of choosing among three kinds of file data, all of which had been condemned by both the Census Bureau and the Department of Defense within the last year or two.

The lower court properly held–

John Paul Stevens:

Are you saying that the three… what I would call three alternatives.

They’re not three alternatives.

Dwight Golann:

–Well, they are three alternatives, Your Honor, but they miss one very important alternative, which is not to use file data at all.

John Paul Stevens:

So the duty under the seat belt case is not to consider an alternative but to consider all alternatives.

Dwight Golann:

Or at least all the important alternatives, and the obvious other alternative was returning to the historic practice, which was, as I say, to count them as Americans… in fact, apportion them the same way as the domestic military, where they were living on the date of the census and not where they had lived years or decades before.

The lower court concluded that this complete and unexplained failure to consider an important alternative violated the standard of Motor Vehicle Manufacturers and was arbitrary and capricious.

I should note that this Court should not be concerned with some of the undesirable results that have been suggested in the Secretary’s brief.

For example, that every State will file suit in every decennial census if… or nearly all the States if this decision were upheld.

The key to this case is that Washington received the 435th and last seat awarded in the apportionment.

Massachusetts would have received the 436th seat.

As a result, Massachusetts, but not other States, was able to show causation and standing resulting from this reversal of historic practice.

In any particular decade, there might be a handful of States close enough to qualifying for that last seat awarded to be able to make out standing to sue.

We would also argue that–

William H. Rehnquist:

Mr. Golann, how about any number of other entities such as counties or school districts which may depend on census data for Federal funding?

I mean, I don’t think it would have to just be a controversy over the apportionment seats.

Dwight Golann:

–Mr. Chief Justice, we would emphasize that this case involves only the constitutional apportionment census.

It is true that the Secretary conducts many other censuses, for example for Federal funds.

He does not apply the apportionment standards to those censuses, nor need he.

For example, in 1970, when overseas employees were apportioned back for State apportionment purposes, their numbers were not included in the Federal figures used to allocate aid to municipalities.

We see no reason why the Secretary would have to do that for the nonconstitutional, entirely discretionary–

William H. Rehnquist:

Can we reasonably differentiate, though, the kind of census you’re talking about from the other kinds of census that these other groups justifiably rely on?

Dwight Golann:

–I believe so, Your Honor, and the reason is that the Apportionment Act, which contains the three standards I’ve articulated, only applies to the apportionment census.

It does not apply to all these other discretionary censuses.

In fact, some of them say they should be conducted without regard to the apportionment census.

William H. Rehnquist:

You’re saying that nothing but the apportionment of the House of Representatives depends on the Apportionment Act census?

Dwight Golann:

That, and votes in the Electoral College, Your Honor.

Those are the two purposes of the apportionment census.

In 1850–

William H. Rehnquist:

It’s not used, then, for any other purposes, for allocating any sort of funds at all?

Dwight Golann:

–The Secretary might have discretion if he wishes to use it.

I note that in the population totals for States, published in 1990, there are different numbers.

There is a resident population, and then there is a total population which includes overseas employees, and they are different populations for the same State, and the Secretary, at least in 1970, did not include overseas employees in the resident populations used to allocate municipal aid.

That was what was at issue, Your Honor, in Borough of Bethel Park v. Stans, and the Third Circuit held he need not, so we don’t see any connection between the Constitutional census and all of the other surveys that the Secretary has conducted and as to which he enjoys very wide discretion, perhaps unreviewable discretion.

John Paul Stevens:

May I ask one other question?

You said that only Massachusetts has standing because it’s the 436th seat and it’ll take a seat from Washington, but why couldn’t the error, if it be an error, affect five or six States?

Dwight Golann:

It is conceivable that there might be some sort of earthquake in the census.

It would have to be an illegal earthquake.

It would have to be exceedingly widespread in order to affect States that are many, many places away from qualifying for the last seat.

Under the very discretionary–

John Paul Stevens:

Why do you say that?

I don’t think the figures are that clear.

Maybe they were on this particular allocation, but it would seem to me it would be entirely possible for three or four States to be affected by an adjustment of the kind you think might be necessary.

Dwight Golann:

–The universe might be, I would suggest, perhaps five or six or half-dozen.

For example, Montana was the 441st, I believe, but they had to advocate a formula that had never been used in the history of the United States in order to make out a case of standing, find a formula that would produce entitlement to that seat.

John Paul Stevens:

Well, they had two formulas that would have worked.

Dwight Golann:

Well, I don’t believe either one had ever been used, and that illustrates the problems when you are very far away from qualifying for the last seat.

The legal theory has to be increasingly unusual.

We simply don’t see any reasonable likelihood that a state could make out a case of standing.

Antonin Scalia:

Mr. Golann, suppose I disagree with you that somebody residing abroad can’t be counted so long as he intends to return to the United States and intends to return to a particular place.

Assume that I disagree with that, but assume also that I am concerned that this data is pretty stale, that they said they were going to return to this place 20 years ago.

It may be pretty bad data, but it may be better than any other data that we have.

I mean, why is it arbitrary to say, it’s all we got?

It’s certainly going to be more accurate than not counting these people at all.

It may be wrong two times out of five, but if you don’t count them at all, you’re going to be wrong more than that.

Dwight Golann:

Your Honor, I would have two concerns.

One is that the Constitution specifically requires an updating every 10 years.

Under the theory that the Secretary applied to overseas employees, he could cancel the entire 1990 census, say we’ll use 1980 data, it’s better than nothing, we’ll adjust it 5 percent, and that will be our apportionment count.

That would clearly not be a decennial census, and that is essentially what happened for overseas employees in this case.

Antonin Scalia:

Well, it’s… you mean there are no estimates made in the census at all?

The census-takers goes and they lay their hands on shoulder by shoulder?

Don’t they ask people, how many people in the house, things of that sort?

Dwight Golann:

They do, Your Honor, and they ask very specific questions, and they check the forms against each other, as you will find in the official forms.

It’s really quite precise.

They do not do an overcount or undercount or a statistical adjustment.

At least, no court has every ordered the Secretary to do so.

It is a head count.

It is done not by touching shoulders but by–

Antonin Scalia:

How long does the process take?

Dwight Golann:

–I believe it occurs over a few days in April of 1990.

At least, that’s what the instructions say in the joint appendix.

This data, I might note, was not… the selection wasn’t made until July of 1990, and our concern, for example, if there were no reviewability, is that a future Secretary could sit down, having received the census results, knowing what they show, choosing among three forms of file data that he or she would know would throw a seat to one State or to another State and might succumb to partisan political concerns to move a seat, and under the Government’s theory, as I understand it, that would not be reviewable by any court, and that is very troubling.

I should note that, Your Honor, one other concern raised in the briefs were that Americans would be thrown out of the census counts, and I’d like to emphasize that these Americans were counted as citizens.

They were counted in the same way that the domestic… or should have been counted, would be counted under the court’s order in the same way domestic military were counted… that is, at the base where they are serving, and not apportioned back to where they lived before they joined military service.

They would be apportioned in the same way as residents of the District of Columbia, of Guam, and other dependencies.

Sandra Day O’Connor:

Mr. Golann, would Massachusetts count for purposes of allowing them to vote military personnel with a home of record in Massachusetts?

Dwight Golann:

The standards, Your Honor… they might or might not.

The standards under the Absentee Voting Act are inconsistent with the standards the Secretary used in this case.

That is, the standards in the Absentee Voting Act are… there are various standards.

One is domicile, and so it is quite possible, and we think undoubted, that some people were apportioned to States different than the States in which they had voted and–

Sandra Day O’Connor:

But also, I suppose, true that some people with a home of record in Massachusetts would be allowed to vote there–

Dwight Golann:

–Yes, Your Honor.

Sandra Day O’Connor:

–Even though Massachusetts is here saying you can’t count them for apportionment purposes.

Dwight Golann:

That, Your Honor, was decided by this Court in Carrington v. Rash.

In that case, the State of Texas said you are inflexibly allocated to the State you indicate when you enter military service.

Sergeant Carrington was told that because he had indicated a certain residence 17 years before, he couldn’t vote in Texas even though his domicile was there.

This Court held that allocating inflexibly people to States based on, I think it was home of record in that case, was irrational and violated the equal protection clause.

That does not directly control this case, but we find it a useful analogy, because it indicates the problems with this data and the problems with an inflexible application of stale data, in that case 17-year-old data.

Your Honor, I would note again that this issue of decenniality, as we call it, is very important.

Dwight Golann:

The framers made a conscious decision to have a census recount every 10 years.

They saw the example of the English parliament, which had so-called rotten boroughs where there was not a recount, stale data, and the population moved greatly, and they wanted an up-to-date, in effect a snapshot of the American population as of April 1st.

Home of record doesn’t give that snapshot.

As we say, it shows at most where people lived years before, and in some cases where they didn’t live at all.

It violates the core requirement of a decennial census.

Byron R. White:

Well, it would be rare that military people have that kind of staleness.

Dwight Golann:

Your Honor–

Byron R. White:

Military people ordinarily don’t stay overseas forever.

Dwight Golann:

–No, Your Honor, but they move around, and in this voluntary Army–

Byron R. White:

Well, you think a very high percentage of military people stay out of the United States for 15 years?

Dwight Golann:

–No, but they move from one–

Byron R. White:

All right.

They may move around overseas, but how long do you think they… on the average they would be overseas out of this country?

Dwight Golann:

–Your Honor, I don’t know, and I’d have to refer to personal experience, which I don’t wish to.

Byron R. White:

Well, when they come back they’re going to be counted where they live.

Dwight Golann:

That’s right.

If they are–

Byron R. White:

They won’t be any longer any problem about them, but when they’re overseas for 4 years, what’s wrong with the home of record?

Dwight Golann:

–The problem with home of record is–

Byron R. White:

Certainly it won’t be staleness, will it?

Dwight Golann:

–Well, Your Honor, the record–

Byron R. White:

Well, will it, or not?

Dwight Golann:

–Yes, it will, Your Honor.

Byron R. White:

What, 4 years stale, is that it?

Dwight Golann:

No, because it’s a record… it’s not where they lived just before they left the United States.

Byron R. White:

Well, that isn’t staleness.

That isn’t the problem about staleness.

Dwight Golann:

Your Honor, I believe–

That’s just plain inaccuracy.

–It is where they lived when they first entered the military, and if they entered the military 10 or 15 years before, it’s where they lived 10 or 15 years before, not the last place they lived in the United States before going overseas, not the place they’re going to go back to, and when they return, Your Honor, again the–

Byron R. White:

You think a military person who enters and has a home of record, and he might have established a domicile while he’s in the service in some other State–

Dwight Golann:

–That’s correct, Your Honor.

Byron R. White:

–And he would still be counted if he went overseas as being in the State of his home of record.

Dwight Golann:

Indeed, he would have to be, because he could not change the home of record until he broke military service, left the military for at least one day and then after being out of the military, if he reenlisted he could name again, but otherwise not.

Byron R. White:

So he might have been counted a couple of times in some other State while he was in the military in this country.

Dwight Golann:

If he were in this–

Byron R. White:

He might have been counted in a couple of–

Dwight Golann:


Byron R. White:

–States other than his home of record.

Dwight Golann:

Your Honor, if he were in the domestic military–

Byron R. White:


Dwight Golann:

–That’s correct.

If a person from Massachusetts enlisted in the military is sent to Fort Benning, he will be counted in Georgia as an inhabitant of Georgia.

If he’s sent to Wiesbaden, Germany under this ruling he’s counted to his home of record.

If it’s Massachusetts, it’s Massachusetts.

If he decided he wanted to move to Hawaii when he left the service, it would be Hawaii, whatever he listed, and he could not change it.

I should… I will simply state in conclusion that there are really two ways to look at this case.

The Secretary argues that this is a mere choice of data, that if the Court accepts this characterization it should find for the Secretary.

We believe this case arises with constitutional and statutory claims that involve issues of enumeration, decenniality, and inhabitancy, none of which were satisfied by the use of stale and outdated file data.

We also believe, as the lower court found, that the Secretary refused to consider important alternatives, such as returning to historic practice.

These troops should have been counted the same way the domestic military were counted.

They should have been counted where they were, where they were living, and not in States which they weren’t residing.

If there are no further questions, Mr. Chief Justice, we submit.

Antonin Scalia:

What if I think they should have been counted where they intended to reside when they came back to the United States?

Then what should the Secretary have done?

Dwight Golann:

He should not have used home of record data.

Antonin Scalia:

I know that.

Dwight Golann:

He should have enumerated–

Antonin Scalia:

I know you believe that.

What do you believe he should have done?

Dwight Golann:

–He should have conducted an actual enumeration and asked the question Your Honor posed, where do you plan to live when you come back?

Antonin Scalia:

What if there was no time to do that?

Dwight Golann:

If there is no time now, in that case they should not be counted… or not apportioned back to States at all.

Antonin Scalia:

At all.

Why is that preferable?

Dwight Golann:

That is preferable because the decennial census is a snapshot, and a snapshot of the population on April 1990, and that didn’t happen.

Antonin Scalia:

Why is that snapshot closer to… in focus than the snapshot of using stale data but at least some data?

Dwight Golann:

Because for one thing it involves the problem… and this was the reason for some other variations from usual residence of double counting.

People would be counted one place in 1990, they’d be counted a different place in 1992, and that would be one problem.

The other problem, I believe, Your Honor, would be that as a practical matter it couldn’t happen.

You will note in the record that when the Secretary did conduct a minor enumeration of 60,000 civilian employees, only 20 percent of them ever returned their questionnaires.

Now that they have left the military in Your Honor’s hypothetical, I suspect the response rate will be so low that it will simply not be an enumeration.

David H. Souter:

So that no data except that gathered on essentially an account made every 10 years will ever suffice on your theory.

Dwight Golann:

That is correct, Your Honor.

That is, to us, the meaning of the decennial census.

William H. Rehnquist:

Thank you, Mr. Golann.

Mr. Roberts, you have 5 minutes remaining.

John G. Roberts, Jr.:

Thank you, Mr. Chief Justice.

One reason the Court should be hesitant to conclude that judicial review is available is the consequences.

My brother says it’s not going to be a big problem, it’s always just going to be between number 435 and number 436.

That’s not true at all.

If your State is number 440 on the list and you look, and you see if the undercount litigation comes out one way, you move up to 438, and then if litigation about how foreign personnel should be counted moves you up to 436, and then you need to look very carefully at how many boarding schools you have in your State to see if there are enough of those, that’s a departure from the usual residence rule to bring you up to 435.

You have to file that lawsuit even though another State the same day is going to file a lawsuit against the Secretary for not counting American citizens employed by private companies abroad, and that’ll lower you down again, and all of this litigation is going to be going on while the States are redistricting, while candidates are considering whether to run, while primaries are taking place.

The point is that the conduct of the census and the reapportionment is not suited to a case-by-case judicial review of each one of the thousands upon thousands of judgment calls that go into the census.

That’s particularly true when you consider that it’s done in a limited time and on a limited budget.

The district court looked at one decision… how to allocate Federal personnel overseas… and concluded that an actual survey would have led to more accurate data.

That may or may not be true.

The record is unclear on that.

But even assuming it’s true, it would have led to more accurate data only at a great increase in cost.

The Secretary determined that the additional accuracy, if any, was not worth the cost.

John G. Roberts, Jr.:

There’s no standard–

Byron R. White:

Mr. Roberts, do you think we have to vote this method of enumeration up or down to decide this case?

John G. Roberts, Jr.:

–Absolutely not.

I think the Court should conclude that the decision of the Secretary is committed to his discretion by law and therefore not subject to judicial review.

Byron R. White:

What if we don’t agree with that?

Couldn’t we say… couldn’t we decide this case by saying that well, it may be all right, but the Secretary has not done a very good job of telling us why, and then we would just send it back.

John G. Roberts, Jr.:

Well, if you agree, if that is your view, you should still reverse because this matter was not remanded to the Secretary.

Byron R. White:

Yes, all right.

All right.

I know the United States position is that it was the wrong remedy–

John G. Roberts, Jr.:


Byron R. White:

–And it should have been remanded, but we could remand without deciding whether this home of record approach is correct.

John G. Roberts, Jr.:

You could remand to the Secretary, yes, of course, but that is not what the district court did, and if you disagree with us–

Byron R. White:

I understand that.

I take it you wouldn’t be satisfied if we just did that.

You’d rather have some other decision than just–

John G. Roberts, Jr.:

–Yes, we wouldn’t consider that a resounding victory.


We believe that the matter is not subject to judicial review.

The point I was making on the cost is there’s no standard… there are no standards for a judge to decide, is this additional accuracy worth the cost, particularly since the additional funds that would be devoted would be taken from some other enumeration task, say an effort to increase the count of the homeless or migrant workers.

The district court’s in no position to decide whether the increase in accuracy of the overseas count is more important than the decrease in the accuracy of the count of migrant workers.

Now, Massachusetts has no trouble treating these servicemen and women overseas as its own when it comes to taxing them, and they don’t say you have to have a residence here before we can tax you, and it does allow them to vote.

They also should be permitted to be counted for purposes of the census.

Antonin Scalia:

–I think Mr. Golann is saying inaccuracy in not counting is okay.

If you don’t want to spend money, you don’t count.

So long as you don’t put any noses in there that you haven’t counted, it’s okay.

Do you know of any other instances where the Secretary saves money by being a little inaccurate, by counting noses that he shouldn’t count?

John G. Roberts, Jr.:

I would say in every instance of the census.

We miss people.

We miss people when we count and go house by house.

Antonin Scalia:

Well, that’s fine, but those are noses you haven’t counted.

That’s okay.

John G. Roberts, Jr.:

When we give you the census form and it says list the people in your house, I mean, just as on the answered home of record, maybe people lie.

They say well, I’ve got four children.

Or maybe they list and they’ve got four children including one who’s off at college, even though the rules say that person shouldn’t be counted.

There is inaccuracy inherent in a census of 250 million people.

The Secretary does his best… her best to eliminate the inaccuracy, but is it a better thing… if there are, let’s say an inaccuracy of 4 percent on the home of record data, is it better to exclude it so you don’t bring in that 4 percent error while you’re excluding 96 percent who should be counted?

John Paul Stevens:

The one thing about the home of record, some of that data is more than 10 years old, and how do you respond to the requirement of a census every 10 years, his argument there?

John G. Roberts, Jr.:

The home of record data is current data.

It is the judgment about the… what approximates the usual residence of that person today.

It may have been entered on the records in a prior year, but it is current data.

Thank you, Your Honor.

William H. Rehnquist:

Thank you, Mr. Roberts.

The case is submitted.