Forsham v. Harris – Oral Argument – October 31, 1979

Media for Forsham v. Harris

Audio Transcription for Opinion Announcement – March 03, 1980 in Forsham v. Harris


Warren E. Burger:

We will hear arguments next in Forsham against Harris.

Mr. Sonnenreich, I think you may proceed whenever you are ready.

Michael R. Sonnenreich:

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

The issue before the Court, in our case, is whether certain scientific records that are presently located in a Maryland bank vault are federal records and are disclosable under the Federal Freedom of Information Act.

The committee for the Care of the Diabetic, of which petitioners are members, initiated a formal Freedom of Information Act request from the Department of Health, Education, and Welfare.

In 1974, the data resulting from a scientific study which has been called the UGDP Study.

The study involves the treatment and evaluation of the treatment of diabetes.

This request was denied administratively and we filed an action with the District Court in the District of Columbia.

Our request for relief was denied by Judge Corcoran in that Court — denied essentially on the grounds that the federal government did not posses the data and therefore the request for UGDP records in the Court’s judgment were not at agency records.

In the U.S. Court of Appeals, we again were denied in a 2:1 decision to release the data, but all three judges rejected the strict possession property test called upon by the lower court.

Judge Leventhal in the majority opinion set a standard where he defined agency records to include records created or obtained by an agency in the course of doing its work.

In elaborating on this test, he stated that agency records are not limited to the physical possession of an agency but it can include the records which are created and held by private entities where the agency is involved in the core planning or execution of the program.

Judge MacKinnon in his concurrence also stated that agency records can include those created by private entities and said that the determination of whether or not the agency records must be made will be done on a factual basis.

Judge Bazelon dissented and stated that the records created and held by private entities are agency records when the government has been significantly involved in the study in its records.

What is interesting —

Warren E. Burger:

What would be involvement, beyond finances or limited to finances and definition of the project?

Michael R. Sonnenreich:

No, Mr. Chief Justice, there are essentially — we have set it out in our brief at pages 53 through 55, but essentially there are four major factors that show government involvement.

The first is a question of funding.

We have demonstrated in the brief, and we are going to discuss this, that you not only have a funding over a period of more than 15 years, we not only have a funding where the exclusive funding agency is the federal government and the amount involved was in excess of $15 million, but in addition, the government also funded the planning grants — a two-year planning grant to initiate the study.

Another factor that was extremely important in terms of showing involvement is the fact that the monitoring and the supervision by the National Institutes of Health in the project itself, or we have demonstrated in our case, in our brief that some of that involvement is instructive to the Court.

For example, they established a Policy Planning Board and they put members of the National Institute of Health on that board, one of the senior officials being the Chairman of the board.

They were very actively involved in analyzing and evaluating the actual planning and policy of the study.

More important in respect to that second factor was that the Board met regularly at the National Institute of Health to review the actual progress of the study, which in and of itself was unusual.

More unusual still was that NIH, because of some of the questions that had been raised, went outside and contracted with an independent group of statisticians — the Biometric Society to actually go in and evaluate the raw data of the study.

This is not a usual activity on the part of the National Institute of Health or of HEW.

As a matter of fact, Dr. Donald Whedon in his memorandum, in his affidavit before the lower court so stated that that is, in fact, a very rare occurrence.

A third important factor to show the substantial Government involvement was the action — by the action of the Government in terms of regulatory action.

What is important is that while the National Institutes of Health were monitoring and supervising the study, another sub-agency of HEW — the Food and Drug Administration — began to take a series of very significant steps in reliance on this study.

In 1970, the FDA sent a series of drug bulletins to all physicians in the United States advising them of the findings of the study and warning them about its implications in the treatment of their diabetic patients.

These bulletins were then followed by the Food and Drug Administration making a proposed rule re-labeling the oral hypoglycemics by inserting a warning based on this UGDP study.

Michael R. Sonnenreich:

This proposal was first discussed by the FDA in 1972 and was finally proposed in 1975.

In 1977, in direct reliance, the Secretary of HEW proposed that one of the drugs in the UGDP study, Phenformin, be suspended from the marketplace.

Again, the Secretary relied heavily on the UGDP data and the UGDP study.

From all of these regulatory activities, ranging from official bulletins of the FDA that went directly to physicians to proposed rule-making to regulatory proceedings, the Government relied heavily on the UGDP as the justification for its actions.

But the most important of these effects is the fourth and final factor that I think is clear showing of reliance on the part of the Government, and that is when the Government itself through the FDA decided to audit the raw data.

Warren E. Burger:

Let me back up a little bit.

Is this fundamentally a statutory construction case for the Court?

Michael R. Sonnenreich:

To a certain extent, yes, Your Honor.

Warren E. Burger:

Well, what else is it besides a statutory construction case?

Michael R. Sonnenreich:

What has happened is that the lower courts have been evolving a series of standards because the statutory language does not speak directly to what constitutes agency records.

And what is before this Court, petitioners contend, is trying to amalgamate or define with more precision what it is, what type of test has to be utilized now in terms of determining what are agency records.

Warren E. Burger:

Then are you asking us to do anything except to construe a statute?

Michael R. Sonnenreich:

No, Your Honor.

We are asking you to define that test by construing the statute.

Warren E. Burger:

But you tell us very early in your brief that in this case, we cannot rely on the statutory language.

Michael R. Sonnenreich:

The statutory language in this particular case, Your Honor, is silent on what constitutes agency records.

What has happened is that the Courts have had to evolve a series of tests to determine what those actions are.

The statutory intent is very clear, and that is that the intent of the statute is to permit disclosure, and is to make the Government accountable for the activities and the actions that it performs.

The problem that we deal with now and the reason why we feel this case is so important is that in defining the threshold issue of what constitutes the agency records, we are also defining what is the accountability of the Government.

The problem that we have in defining, it is the only way you can examine and determine what are those agency records as to look directly at the involvement of the Government and then interpret what that action of the Government is in the determination of what is the agency record.

The statute, the only thin light or light that we get is the Attorney General’s memorandum opinion of 1967 which has some definition of what constitutes agency records in the sense that he makes two separate distinctions and they talk in terms of possession or control.

The problem is that the Courts have had a great deal of trouble in trying to determine possessory standards and apply possessory standards to the Freedom of Information Act, because they are involved in this balancing and this weighing of what it is that the Act really intends for the Government to do in terms of making it accountable, and on the other hand, trying to make certain that the uncertainties are not so great that the Act encompasses everything.

Could I ask you, is it your position that if a research contract expressly provided that the raw data would remain the property of the researcher, and that the Government would have no access to the raw data, would you still be making an argument that that data —

Michael R. Sonnenreich:

If the is completely privately owned and there is no access by the Government and there is no reliance by the Government.

Let us just say you had exact same facts in this case, everything happened except that the contract said that the data remains the property of the researcher and the Government will have no access.

That’s the only change in the entire case.

Would that settle the matter, as far as you are concerned?

Michael R. Sonnenreich:

No, Your Honor, I do not believe it would.

Well, why would it not?

Michael R. Sonnenreich:

Because what happens is that the Government has gone beyond that statement in reaching out to utilize the data, has reached out to substantially involve itself with the data process.

Well, there is access to everything that the researchers have reported to it, and it can take action based on what the researchers report to it.

But if this contract said, “You keep the basic data and we do not want to look at it, and we cannot look at it.”

Would that not settle the matter for you?

Michael R. Sonnenreich:

If that were the contract and there was no right to the data, the Government did not rely on it, then my answer would be yes, it would settle the matter for me.

That is not, of course, the case in point, nor is it that the case in point, in fact that the Government did have access, did exercise access, did audit the data, did put its hands on the data, and did rely on the data, both for regulatory action and for regulatory decision-making.

Warren E. Burger:

The Government has access to a great many things independent of contract.

I suppose your answer must be taken as meaning access which the Government could enforce in some way?

Michael R. Sonnenreich:

Or exercise.

Warren E. Burger:

Not access which a university might or a private research corporation might grant them as a matter of accommodation?

Michael R. Sonnenreich:


The question becomes that if they can have it.

If they have the right to exercise access to the data, that is one of the crucial points in determining whether or not there was substantial Government involvement.

Warren E. Burger:

Suppose one of the big drug companies working with millions of dollars in research should invite and permit Government’s people from the National Institutes of Health and elsewhere, Food and Drug Administration, to come in and monitor or at least observe all that they were doing.

Would that create a right of access, in your view?

Michael R. Sonnenreich:

I do not think that creates a right of access, Your Honor.

The question would follow, “And what did they do with what it is that they monitored, sought, and examined?”

If they then went ahead and relied on it, the question is yes, we would think that would be substantial Government involvement because they did have access –-

Warren E. Burger:

Is the keystone to a right of access the fact that the Government funded the enterprise?

Michael R. Sonnenreich:

And that they had the right to it, specifically.

My recollection is that the Fair Labor Standards Act authorizes the Government to inspect the wage and hour records of private employers if it thinks there may be a violation, and also requires the keeping of those records.

Do you think that a citizen could file a suit under the Freedom of Information Act requesting a private employer’s wage and hour records?

Michael R. Sonnenreich:

No, I do not, Your Honor.

I think that kind of data is a different type of data than the type of data we are looking at.

I think that the Court has to examine what it is that we are talking about when we talk about data.

The data that I believe you are addressing is a ministerial kind of data that is a collation and function of collection.

What we are involved in here is data the predicate of which Government action was taken because it forms a decision-making basis for conclusions, recommendations and findings, which is far different from the type of data that you are addressing in the Fair Labor –

Well, the information that my brother Rehnquist refers to might turn out to be the very basis for a regulation or a rule-making.

Michael R. Sonnenreich:

If in fact, Your Honor, the Government does then do rule-making on the basis of that, I would submit yes.

Warren E. Burger:

They could then go in, they could then require the Government to produce the wage records of private employers and make them public?

Michael R. Sonnenreich:

Unless it is exempted, yes.

Michael R. Sonnenreich:

The point that I did want to make in turn in talking about the fact situation in showing the Government involvement here was the fact that the audit is a very significant factor in this case and does to a certain extent make this case somewhat unique.

The reason, as I stated, was for that audit to determine the validity of the study and the integrity of the database.

This is in contrast to the Biometric’s audit which was contracted for by NIH.

The Food and Drug Administration decided to audit the data itself.

It physically went out to the coordinating center in Baltimore, selected data, examined the data, copied some parts of the data, and then took back limited portions of the data that they had seen back to Rockville, Maryland.

What happened was that after completing the audit, the Food and Drug Administration then published its findings as final findings in November of 1978.

It then went back and renewed its proposal of three years earlier, which was to require a warning be placed on all oral hypoglycemic drugs.

The central issue in this case we submit, is whether the raw data of the study funded by the Government, monitored by the Government, relied upon by the Government, and even audited by the Government, are Agency records under the Freedom of Information Act.

Looking at the fact basis of this case, petitioners contend that these are agency records.

Now, respondents in their argument to date have raised a series of issues which tend to blur what we perceive to be the central issue of this case.

I would like to just take a moment with the Court to raise some of the issues that respondent raises and answer them.

They are in basis three.

It has been suggested that the data being sought involved identifiable patient records.

Petitioners contend that this is not so.

The data that the petitioners are seeing or wishing to see and examine is the same data that the Food and Drug Administration, or for that matter the Biometric Society, saw or could have seen when they undertook their data, their audit of the data.

We are not looking to reveal patient identity and there is not a confidentiality issue, petitioners submit, in this case.

It was also suggested that it would be necessary to find the UGDP a Government agency in order to find the raw data agency records.

Again, we submit this is not so.

The agency that we are talking about is the Department of Health Education, and Welfare and its sub-agencies.

We are not contending that the UGDP is a Government agency.

Well, is it not that argument of the Government is just corollary to the basic argument that agency records, whatever else they are, must be in possession or control of a governmental agency.

And since UGDP is not a governmental agency, then these cannot be agency records, since these raw data are concededly in the possession or control of UGDP?

Michael R. Sonnenreich:

That is the argument that they advanced, but they advanced it principally on the issue of possession rather than on control.

Because in this particular case what has happened is that the Government has gone out on two occasions by contracting directly put their hands on the data, so the control issue of did the Government have control to reach out and examine the data –-

But does it now?

Did it have control at the time of your request, that is the issue, I think, in this.

Michael R. Sonnenreich:

Yes, but that is –

Not that it did ever have control.

Michael R. Sonnenreich:

And what we submit it did have control then and it does have control now.

Leaving that control to one side for just a moment, in the statement, the question presented in the Government’s brief, they recite that the records were generated, owned, and possessed by a private non-government group.

Now, I gather you agree with the fact that it is a non-governmental group and that they are possessed and generated by that group.

What about ownership?

Do you agree that the records were owned by that group?

Michael R. Sonnenreich:

The problem with this kind of data is the definition of ownership.

Yes, they are owned by the UGDP group.

The fact that people can go and examine the data and take the data away and still have ownership remain with the UGDP is that we are transmitting information not dealing with tangible items, so what happens is yes, they have ownership of the data, but also yes, the government has a right of access to walk in, take the data, copy the data, analyze the data, and go back to Rockville, Maryland with it, leaving it in their possession, in their ownership, and then the Government has the ownership of what it did with the data.

Could just this one thing, if I may.

Are there any cases that you know of under the Freedom of Information Act where documents that were not owned by the agency involved were required to be produced

Michael R. Sonnenreich:

That were not owned?

That not owned by the government, yes.

Michael R. Sonnenreich:

They were not specifically owned?

No, I do not, Your Honor.

Of course, your point is that common-law notions, property notions of ownership are basically irrelevant?

Michael R. Sonnenreich:

Especially with respect to the generation of data and the transmission of scientific data.

William H. Rehnquist:

Well, how is your analogy of the data, of the use by the Government here any different than reading a magazine that somebody else owns or reading a book that somebody else owns?

Michael R. Sonnenreich:

Because what is involved here is, we have to step back a moment, Mr. Justice Rehnquist.

What we are talking about is the Government relied on a conclusion, and in order for the Government to deal with data, the Government has to analyze, evaluate it.

It is not just a question of reading data and copying it ministerially.

The data has to be organized in a way to prove or disprove conclusions.

This is why I was trying to draw the distinction between data which is merely collective, for example in SDC versus Matthews, which is put in the Medlar Survey at the National Library of Medicine.

And it is there and you can retrieve it and you can read it, as opposed to data which is action-oriented data in the sense that in order to understand it, in order to utilize it, you have to evaluate it, you have to analyze it.

And that is what the Government did, because the only way that you can test the conclusions of the Government is not to just compile a series of pieces of information.

What you have to do is analyze it to show that the integrity of the data, the veracity of the data, in fact meets those scientific judgments.

So, there is a fundamental problem in getting at that kind of data.

It is more than simply saying, “Well, it is there, and can I retrieve it or can I not retrieve it?”

I have to do more than retrieve it.

I have to use it.

And that is exactly what the Government did, and what the Government’s argument has always been is trying to separate utilizing in reliance on the study and utilizing in reliance on the raw data, which is the fundament of the study.

Mr. Sonnenreich, I want to be sure about one technical aspect.

The Freedom of Information Act does not define agency records?

Michael R. Sonnenreich:

That is correct, Your Honor.

And do you concede that the applicable definition is that in the Federal Records Act?

Michael R. Sonnenreich:

I am sorry; I did not hear that.

Do you concede that the applicable definition here is the one contained in the Federal Records Act?

Michael R. Sonnenreich:

No, I do not.

My feeling is that the definition is instructive, but it is not dispositive.

My feeling is that the purpose of the Federal Records Act is slightly different from the purpose of the Freedom of Information Act, and the disclosure requirements and the goal of the Freedom of Information Act could well and does well change what constitutes an agency record.

I do want to point out one other point that the Government contends, and that is that the Government contends that the test that we are seeking to have this Court adopt will in fact blur the so-called bright line of law defining agency records.

Now, this is not before the Court, but we will not argue and we do not argue whether or not the test, that test may exist as to who gets records under the Freedom of Information Act, but we do say that under this standard and the standards enunciated in the Courts below, it does not have applicability here.

Well, I thought if one thing was clear, it was that if anybody gets records under the FOIA, everybody does.


Michael R. Sonnenreich:

That is correct.

I mean, it abolished the normal concepts of standing, did it not?

Michael R. Sonnenreich:

That is correct.

Or any showing of need, or –

Michael R. Sonnenreich:

What we are involved in here and what the Court has to decide here is, where is the door to get into the Freedom of Information Act?

If you are right under the FOIA, then any citizen of the United States can equally get this information?

Michael R. Sonnenreich:

That is correct; that is absolutely correct, Your Honor.

What we are stating in this argument that Respondents made is that we do direct the Court that Ciba-Geigy versus.

Matthews, Forsham versus Califano below, SDC versus Matthew, all went beyond discussing the possessory test and all went towards trying to evaluate Federal involvement, Government involvement, in the use or in the dealings with those records, and they are all talking about private entities which are possessing the records.

William H. Rehnquist:

None of those are cases from this Court, are they?

Michael R. Sonnenreich:

No, Mr. Justice Rehnquist.

All are cases below.

But the cases below do indicate, Your Honors, that the concept of agency records do not lend themselves to precise definition, and they are moving away from this technical property test toward a much more functional approach of examining the factual basis of the area, of the particular case, to determine whether or not the Government involvement is such that would render those agency records.

All agree you do not have to be an agency in order to have agency records.

The question is, functionally, what triggers the test and makes them agency records, letting people in the Freedom of Information door?

William H. Rehnquist:

Tell me again, why do your people want these records?

Michael R. Sonnenreich:

The reason that the petitioners are looking for the records is because of this enormous scientific controversy that has now raged for nine years, and what we are trying to do quite honestly is evaluate the records against the published conclusions of the UGDP to determine whether in fact they are true, and whether or not they are not.

In fact the data supports the conclusions made.

William H. Rehnquist:

But if your people can get them under the FOIA, presumably a drunk can walk in off the street and get them.

Michael R. Sonnenreich:

But that is true with any —

That is true.

Michael R. Sonnenreich:

— Freedom of Information situation.

So, the inquiry as to why your people want them is totally irrelevant under the statute?

Michael R. Sonnenreich:

That is correct.

I agree with that.

I do want to raise one —

Except that I am curious. [Laughter]

Michael R. Sonnenreich:

The reason we are here is very simply that this has affected the impact on the treatment of diabetes, and it has affected five million diabetics.

Well, certainly your claim is not that of a drunk off the street.

You claim a legitimate scientific reason for seeing it.

Michael R. Sonnenreich:

The Committee of the Care of the Diabetic are reputable diabetologists and physicians and their patients, and the only reason that this committee was formed was because of the scientific controversy surrounding the UGDP, was to try and get and evaluate this data to determine in fact whether or not it was scientifically correct.

One unusual factor in this case that should be pointed out is the test of scientific credibility is always to make the data available whenever there is scientific controversy.

The peer review process is the way in which the conclusions are tested by examining the data.

What makes this case unique is that we are not getting the data, and what is unique is that researchers that were involved in the program have been denied the data.

Foreign scientists have been denied the data.

That is the uniqueness of this case.

For reasons that we cannot understand, the normal peer process is not working.

Warren E. Burger:

But if the Government had not funded it, if the Ford Foundation did not funded it, there would not be any case at all, would there?

Michael R. Sonnenreich:

No, if the Ford Foundation had funded it and the Government had had access to the data, and the Government had substantially involved itself in that process and relied on it, the request would be made under Freedom of Information for that data.

Warren E. Burger:

Well, by saying the request would be made, you are telling us that you think the Freedom of Information Act —

Michael R. Sonnenreich:

Would apply.

Warren E. Burger:

— covered it?

Michael R. Sonnenreich:

Yes, Your Honor.

Warren E. Burger:

Then how does then the Government funding is not a major factor?

Michael R. Sonnenreich:

It is a factor.

It is one of many factors that the that Courts must look at in defining what is significant Government involvement in a particular fashion.

Warren E. Burger:

In this case if there were no Government funding at all, would you still be here?

Michael R. Sonnenreich:

If we had the right of access and the Government relied on the data, yes, Your Honor, we would.

The only point that I would like to leave the Court with is that there is a problem that the Court must address, and that is, if the test is very narrowly construed from the possession standpoint, one of the problems that arises is the whole question of Government accountability.

Michael R. Sonnenreich:

The Freedom of Information Act was intended to grant public access to documents.

It was not intended to be used tactically to circumvent disclosure, and the narrower the test, the greater the likelihood of disclosure.

We also concede that the broader the test, the more uncertainty creeps in, in trying to define what is it significant Government involvement.

The problem before this Court on a larger scale is to define that test so that the Freedom of Information Act itself is not subverted, and that there is not a withholding by using various means to get around definitions of agency records so that legitimate reasons where the public has a right to access to that data.

Thank you, Your Honor.

Warren E. Burger:

Mr. Geller.

Kenneth Steven Geller:

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

This case concerns the definition of agency records under the Freedom of Information Act.

In the Government’s view, agency records are records relating to the affairs of a Federal agency that are in the possession and control of that agency.

As we have explained at some length in our brief, we believe that this definition is first of all supported by the language of the FOIA which requires all Federal agencies to “make available agency records,” and which prohibits agencies from “withholding” such records.

Giving this language its ordinary commonsense meaning, we contend that an agency can neither make available nor withhold what it does not possess.

Moreover, this definition also finds support in the legislative history of the Freedom of Information Act.

Legislative debates show quite clearly that what Congress was principally concerned about in passing the FOIA and what was thought to be the principal defect under the predecessor of the FOIA, Section 3(c) of the Administrative Procedure Act, was that Federal agencies had refused to divulge public documents that were in their unquestioned possession and control.

There is not a shred of evidence in the voluminous congressional hearings or committee reports or floor debates to indicate that Congress intended the Act to reach documents in the hands of private parties that happened to do business with the Federal Government.

Well, you say in your brief that when you talk about possession and control, you concede that a government agency could not avoid its obligation under this statute by transferring records to a warehouse or something like that.

What about the equivalent of a fraudulent conveyance, where in fact neither had possession nor control because it had legally conveyed possession and control to John Smith?

Kenneth Steven Geller:

Well, to begin with, we construe possession to mean constructive possession.

If the agency —

No, it had absolutely given up possession and control.

Kenneth Steven Geller:

Well, the FOIA does not require Government agencies to hold on to any particular records.

There are other Federal statutes that may well do that, such as the Federal Records Act and the Records Disposal Act, and it may well be that the agency or the Attorney General could force to return those records.

But we think that the FOIA only requires Government agencies to turn over to the public on request records agencies that are in their possession and control when the request comes in.

Potter Stewart:

So, an agency could in fact make a fraudulent conveyance, anticipating FOIA requests for information that was embarrassing to it, and it could convey all those records to John Smith, so far as the FOIA goes?

Kenneth Steven Geller:

So far as the FOIA is concerned, Mr. Justice Stewart, the Federal agency could destroy the records, if they —

Potter Stewart:

Well, but just stick to my question.

Kenneth Steven Geller:

That is correct.

Warren E. Burger:

What powers would the Attorney General of the United States have in the hypothetical suggestion?

Kenneth Steven Geller:

Well, the Federal Records Act requires the Government agencies to maintain record systems that indicate the work of the particular agency.

If something falls within the Federal Records Act, that is, if an agency determines that something is a Federal record, then that agency cannot dispose off that record without the consent of the Administrator of GSA, who has delegated that authority to the archivist.

Now, if the agency in violation of the Federal Records Act were to dispose, transfer certain records, then the Records Disposal Act would allow the Attorney General to bring a suit for the return of the records.

Warren E. Burger:

Well, an employee of the Post Office Department — well, that is private now, but let us say of the Defense Department, sells a truck or an airplane that is in his possession, certainly the Government could get it back. Is there any doubt about that?

Kenneth Steven Geller:

Not if he did not have authority to dispose off it.

I assume that that is correct.

Warren E. Burger:

He just went out and sold it — that sometimes happens, as we know.

Kenneth Steven Geller:

As the Government is the true owner, it could bring a replevin action, I assume, and the Federal Records Act, the Records Disposal Act provide a similar cause of action for the Government when Federal records are unlawfully disposed off.

Do you think, Mr. Geller, that where the records were generated has any bearing at all on the definition of what are agency records?

In this case, the records, the alleged records were not generated within the agency.

In the case to be argued next, the records, if they were records, were generated within the agency.

Do you suppose that has any effect upon the definition?

Kenneth Steven Geller:

For FOIA purposes, I would think not.

I would think that what is relevant for FOIA purposes is whether at the time the FOIA —

Possession and control are the only two tests, in your submission?

Kenneth Steven Geller:

That is correct, although for other purposes, as I said, Federal Records Act —

Well, I understand that but is it not FOIA –

Kenneth Steven Geller:


For FOIA purposes, I think that is right.

Mr. Geller, let me ask the question I asked of your opponent. Do you accept the Federal records definition of agency records as applicable to the FOIA?

Kenneth Steven Geller:

Well, we believe that the agency records definition under the FOIA includes everything that is a Federal record under the Federal Records Act if it is in the Government’s possession and control.

Now, there is a question whether something that is not a Federal record under the Federal Records Act but is instead non-record materials has to be turned over under the FOIA if it’s in the Government’s possession and control.

We discussed that issue in Footnote 38 of our brief in the Kissinger case, to be argued next, and I think that is an unsettled question.

But we think at the very least the definition of agency records in the FOIA includes everything that is a Federal record under the Federal Records Act, if it is in the Government’s possession at the time the FOIA request comes in.

Now, as I was saying, the legislative history includes nothing to support petitioner’s theory of agency records, and we think it is unlikely that Congress would have intended that result without some discussion of what Judge Leventhal correctly termed the, “awesome implications” that such a result would have.

Indeed, it seems entirely reasonable to conclude that in referring to agency records when enacting the FOIA, Congress was aware of and, as Mr. Justice Blackmun just mentioned, meant to rely at least in part on the definition of Federal records in the Federal Records Act and the Records Disposal Act.

This definition refers to documentary materials that are made or received by an agency in connection with the transaction of public business, and that are preserved or appropriate for preservation by the agency.

And once again, we think that this definition plainly suggests that only documents in an agency’s possession and control were meant to be included within the disclosure requirements of the FOIA.

Finally, we believe that the test we have proposed which focuses on actual or constructive possession has the benefit of being easy to apply in the various different and unpredictable situations in which this issue is likely to arise.

Now, as applied in this case, this definition of agency records yields a clear result.

The UGDP raw data that petitioners have sought are not now and have never been in the possession of the Department of Health, Education, and Welfare or any other Federal agency.

As the District Court found, the data had at all times since their creation been in the sole possession of the UGDP, which is a private research group.

In light of these undisputed findings, we think that the District Court and the Court of Appeals correctly held that the data are not agency records within 552(A)3 and (A)4 of Title 5, and that they need not be produced to Petitioners under the FOIA.

Warren E. Burger:

Did the Government have made copies of all these records, under their right of access?

Kenneth Steven Geller:

Well, as to the right of access, petitioners make it seem as if the NIH or the FDA have an unlimited and unrestricted right of access.

As I hope to discuss a little later, this is untrue.

The right of access, we believe and in the affidavit of Dr. Whedon in the record supports, is only a right of access for purposes of the grant; it is not an unlimited right of access.

The FDA has exercised that right of access, as petitioners pointed out, and in conducting its audit, some of the raw data of the UGDP project did come into FDA’s possession.

It was brought back from Baltimore to the NIH headquarters in Rockville.

All of that data has been turned over to petitioners.

So, we are not talking here about any data that ever came into the Government’s possession and control, because that’s all been made available to the Petitioners.

So, your answer, Mr. Chief Justice, would be yes, if the grant allowed it and the access regulations approved of it, the Government could make copies, and at that point they may well be agency records.

It is not an issue in this case, because every document of that sort that was actually copied and brought back into the NIH’s possession was made available to petitioners.

Now, the problem that we see with petitioner’s contrary theory of the agency records provisions of the Freedom of Information Act is that there is really no theory at all.

Petitioner’s submission, although obviously designed to convince the Court that they should be given access to the UGDP data at issue in this case, is based on no principle standards for determining when documents in general are agency records.

Therefore, offers this Court virtually no guidance in deciding when any of the literally billions of documents in the hands of private groups that deal with or receive money from the Federal Government should be considered agency records within the meaning of the Act.

Let me explain what I mean.

Petitioners have never claimed that the UGDP itself is a Federal agency under the FOIA and any such claim would be totally untenable.

Agency is a defined term under the act and the UGDP, which is an organization of private and State University medical research groups working at private and State institutions clearly does not satisfy the statutory definition of agency.

What petitioners have instead contended in this litigation is that the Department of Health, Education, and Welfare is the relevant agency under the FOIA, and then they have attempted to argue that the raw data in the hands of the UGDP are agency records of HEW by emphasizing every conceivable connection between the data in any branch or agency of HEW.

They specifically rely on three main factors.

First, they claim that the National Institute of Arthritis, Metabolism, and Digestive Disorders, which is a part of the Public Health Service within HEW was significantly involved in the design, implementation and funding of the UGDP study that led to the raw data.

Second, they claim that NIAMDD enjoys an absolute right of access to the UGDP data under its grant regulations.

And third, they claim that the Food and Drug Administration, which is a separate agency within HEW, has taken the UGDP data into account for the purposes of regulatory decision-making.

According to Petitioners, it is the combination of these three factors that somehow turns the UGDP data which unquestionably are owned, they concede, possessed and controlled by a private organization, into HEW agency records, that must be turned over to any person.

Would it make any difference to you if the contract not only gave the Government the right to access for some limited purposes, but also said that if the Government chooses to make public the underlying data?

Kenneth Steven Geller:

Well, it would not make any difference to our main theory, which is that only documents in the possession and control, not documents that they may have a right to go out and get, are within the FOIA.

Warren E. Burger:

I take it you have already conceded that if the Government made copies and took them over to NIH or FDA they would be agency records?

Kenneth Steven Geller:

I think that is correct.

They do not want to concede that if it were to happen in the future they would necessarily be agency records, because there is some question as to whether it would be a Federal record under the NIH record management plan, but I have said, and this is true, that every part of the UGDP data that is ever come into NIAMDD’s or FDA’s possession in the past has been turned over to petitioners, or presumably any other FOIA requester.

William H. Rehnquist:

What if the United States Attorney in Baltimore decides that there has been fraud against the Government committed in connection with some of these grant programs and subpoenas and copies some of this raw data?

Would you concede that those were Federal records?

Kenneth Steven Geller:

No, I would not.

Kenneth Steven Geller:

I would not concede that, Mr. Justice Rehnquist.

I think in that case you have possession on the part of the Federal Government.

That is why we think the control test may be relevant.

It is not relevant in this case because the Government has never had possession or control of any of the documents at issue.

Your proposed test is not possession or control, but possession and control?

Kenneth Steven Geller:

That is correct.

And the control — you know, this case raises a number of interesting questions about what is an agency record, and we presented a definition of —

Well, that is the whole issue in this case.

Kenneth Steven Geller:

Yes, but this is, I think, a very easy case for the Court because no organ of the Government has ever had possession or control.

Now, there may be more difficult questions coming along in which the Government —

Let us explore your test with one another example: Supposing you had what were admittedly agency records on a given date and they were stolen by a foreign agent or a private party.

Would they cease to be agency records?

Kenneth Steven Geller:

If they were not in Government’s possession at the time the FOIA request came in, we would think they would not be agency records, because we think that —

Even though they remained the property of the Government, they are owned by the Government?

Kenneth Steven Geller:

Yes, I do not think that property concepts had much of a role to play —

Why do you say that?

You do mention in your statement of the question presented, You point out the fact that these were owned by a private organization.

Kenneth Steven Geller:

Well, ownership — I think the only bundle of the rights that ownership gives you that is relevant in this case is whether you have an undisputed right to divulge it to somebody else.

Presumably if you own documents —

Well, normally if you talk about a person’s records, you do assume you are talking about things he owns.

I do not know why you do not make the same assumption here?

Kenneth Steven Geller:

Well, we think that the words “make available” and “withhold,” which are in 552(a) –

Oh, yes, that is a different question.

Now, my example, there would presumably be no withholding, if they have been stolen, because you are not withholding something you do not have possession of.

Would it stop being an agency record simply because a spy emptied the files?

This is rather a strange concept.

Kenneth Steven Geller:

It really would not matter in any practical sense, because —

Well, it matters on how we decide the case, if we buy your example, I mean your definition.

Kenneth Steven Geller:

I think all the Court need to say is that if an agency does not have possession at the time of the FOIA request comes in then the document that is being asked for need not be produced, whether you say it is an agency record or not.

And very frankly, under your test, if the records had been stolen by somebody, they would not be within the possession and control of the agency at that time —

Kenneth Steven Geller:

That is correct.

–At the time of the request.

So, your answer to Mr. Justice Stevens’ question is very clear, it seems to me, under your test.

Kenneth Steven Geller:

That is correct.

There would not be agency records for purposes of the FOIA.

For other purposes they, of course, would be, for Records Disposal Act purposes or Federal Records Act.

We are just dealing here, of course, with the FOIA.

But you did say that the documents placed in a private warehouse remain agency records.

Kenneth Steven Geller:


What is the difference between the stolen document and the —

Kenneth Steven Geller:

Well, because the agency warehouse situation is a case where the Government does have possession.

It is a constructive possession.

You can go and get it any time it wants.

There is no one else with a claim to it.

Warren E. Burger:

Suppose it were stolen, in a hypothetical, by a foreign agent and taken off to a foreign country, impossibility of performance would be a defense to any demand, would it not?

Kenneth Steven Geller:

I think that is correct.

You could not say the Government is withholding something if it cannot turn it over if it wanted to.

That is why I am puzzled that you use the definition, because you would win that lawsuit because there is no withholding, and I am just not quite clear on why you are so firm on the notion they will stop being agency records.

Kenneth Steven Geller:

I think it is important, though, to realize we are talking about agency records for FOIA purposes, and the definition exists in the same sentence with the words “withhold” and “make available.”

We think they give meaning to what Congress meant by the words “agency records.”

We believe that the Court should reject the essentially standardless approach to the definition of “agency records” that petitioners offer.

As I have already mentioned, there is nothing in the Act itself or in its legislative history to suggest that Congress meant it to reach papers or other materials in private hands.

Perhaps more importantly, petitioner’s approach would lead to more problems than it could possibly solve, because unless the exception carved in the FOIA at petitioner’s behest were to be limited to the UGDP documents alone or to other private documents that were produced under the circumstances virtually identical to the UGDP data, in other words all the four factors they rely on existed.

It would be difficult, if not impossible, to identify limiting criteria in order to prevent literally billions of other private documents from falling within the disclosure requirements of the Act.

Now, petitioners, as I have noted, have stressed three, or now they say four particular factors that they believe make the private UGDP records in this case sufficiently Federal to require their release under the FOIA, but some of these factors would apply equally to almost every meaningful document in the hands of a private group that receives Federal funding.

And I might add at this point that there were over 16,000 Government research grants in existence in 1978 and that all Federal grants of one sort or another accounted for nearly one-fifth of the Federal Government’s budget outlays that year.

In other words, by making the producibility of a document in private hands turn on a balancing of an unknown number of variables, including such subjective considerations as the substantiality of the Federal agency’s funding or monitoring of the private project that produced the records, or the scope of the granting agency’s access rights to the records, or the degree of the agency’s reliance on the final report prepared by the grantee through the use of the records –

Mr. Geller, excuse me for persisting in this one thought, but supposing you had an agency proceeding, like a lawsuit, where a long record and transcript and briefs were compiled and that whole record was stolen by lawyers, would you say that would cease to be an agency record?

Kenneth Steven Geller:

Presumably the agency has no other copies?

That is right, they stole the original.

Kenneth Steven Geller:

Yes, I would still maintain that for FOIA purposes, it is not an agency record.

But as I say, the same result would, you would reach the same result you could not —

I understand that, but I —

Kenneth Steven Geller:

I think that is the way we would think Congress meant by using the words “withhold” and “make available” agency records.

Warren E. Burger:

You have to stand on possession and control?

Why is it not the answer to that that you cannot give someone something that you do not have?

Kenneth Steven Geller:

Well, the Act requires Federal agencies to make available and not to withhold a category of documents called agency records, and I think that is right, Mr. Chief Justice.

If there is something you cannot turn over, then —

Warren E. Burger:

Why have you reached on agency?

The question of the definition of an agency record, if the answer to the request can be, “We are very sorry, but we have not got what you are asking for.”

Kenneth Steven Geller:

Well, I am certain that that is what would happen in practice, but I was attempting to address Mr. Justice Stevens.

John Paul Stevens:

You would win the lawsuit I would propound because you say, “We are not withholding it.

They stole it.”

Kenneth Steven Geller:


Well, that in fact — that is —

John Paul Stevens:

But the thing that puzzles me as why you take this what seems to me rather strange position that that stops being an agency record.

Kenneth Steven Geller:

If that situation would ever in practice to arise, the Chief Justice is right, we would send a letter saying —

John Paul Stevens:

Maybe somebody could donate the documents to the Library of Congress, too.

Kenneth Steven Geller:

That might arise, also.

Thank you.

But it might remand an agency record if so determined under the Federal Records Act, might it not?

Kenneth Steven Geller:

Oh, absolutely.

So, it is important to bear in mind which statute you are talking about?

Kenneth Steven Geller:

That is the point I was attempting to make, only we are not talking about agency records for FOIA purposes, for turning it over to the public at large, not for other purposes.

Thank you.

Warren E. Burger:

Mr. Plank.

Thomas E. Plank:

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

There are three reasons why we believe that Congress never intended the Federal Freedom of Information Act to apply to the records of private citizens and private and state institutions that receive Federal grants, and also why the Act should not be extended to such records.

First, such extension of the Federal Freedom of Information Act would be contrary to the very basic premise of the Freedom of Information Act, which is the right of the people to know what their government is doing.

The second reason is that such an extension of the Freedom of Information Act to the records of grant recipients would significantly alter the basic relationship between the Federal Government and grant recipients.

Thomas E. Plank:

Third, the application of the act to the records of such grant recipients, such as the university records of the University Group Diabetes Program, would have a staggering adverse impact upon grant recipients whether they be researchers or whether they be state and local governments.

Before I discuss these points further, I would like to respond to some of the points that were made earlier.

I would just like to point out to the Court that the record is very clear in this case that not only did only the UGDP have possession of the records, but they had control of the records.

The District Court so found, and Dr. Whedon’s affidavit at page 147 of the Appendix clearly sets forth that the day-to-day management of the University Group was in the hands of the University Group and not the National Institutes of Health or any other agency of the Federal Government.

It is also helpful to remember what are not issues in this case.

The validity of the findings of the University Group study is not an issue in this case.

The propriety of the proposed changes in the labeling of oral hypoglycemic drugs by the Food and Drug Administration is not an issue in this case, and neither is the propriety of the reliance by the Food and Drug Administration upon the findings of the University Group study.

Similarly, the issue of whether the University Group records ought to be released to the public at some time in the future is not an issue in this case.

The issue here is whether individuals and private and state institutions that receive federal grants but otherwise retain their independence and autonomy, or in the case of a state or local government, their sovereignty, the right of those individuals to determine when, to what extent, and under what circumstances their records will be revealed to the public.

Pardon me asking, I suppose the contract may determine it?

Thomas E. Plank:

Yes, Your Honor.

Let us say if the Federal Government in some way interferes with the independence of the that may be relevant.

One of the important principles that is recognized in our society is the right of individuals and private and state entities to pursue their own goals with a minimum of interference from the Federal Government.

William H. Rehnquist:

Can someone who accepts a large grant from the Federal Government complain about interference with its privacy or its goals?

Thomas E. Plank:

Well, I think the point, Mr. Justice Rehnquist, is, there is of course a certain amount of interference entailed in a federal grant, but the purpose of the federal grant is not to interfere with or to direct the subject matter of the research.

William H. Rehnquist:

Well, the purpose of the federal grant is what Congress determines it to be, is it not?

Thomas E. Plank:

That is correct, Your Honor, but in this case and in the cases of most federal grants, the purpose is to support and to assist these institutions and entities as they pursue their own goals.

Warren E. Burger:

Could Congress make all of these records public records by definition?

Thomas E. Plank:

I believe the Congress perhaps could, if it wanted to define agency records to include the records of grantees.

I think they could make that definition.

In this case they have not.

Warren E. Burger:

The agency could make the grant on virtually any terms they wanted, could they not?

Thomas E. Plank:

That is also correct.

Yes, Your Honor.

But I think the point is here that any time the Congress has circumscribed the freedom of a grant recipient to take action.

Grant recipients may not discriminate on the basis of race, may not discriminate against the handicapped.

Congress has specifically so provided, either in the enabling legislation for the grant or in another Act.

In this case, Congress has not stated that the records of grant recipients are federal agency records, and I think that that fact in itself is conclusive.

In addition, the fact of the staggering impact that the extension of the Act would have upon conducting research and the federal program of supporting research suggests very strongly that Congress would not and did not intend the Act to apply to these records simply because they did not address these policy issues.

These policy issues are substantial, as the Court of Appeals pointed out, and Congress at the very least, if they had wanted to do this and had intended to do this, would have discussed those policy questions.

Thomas E. Plank:

Thank you, Your Honor.

Warren E. Burger:

You have just one minute left, Mr. Sonnenreich.

Michael R. Sonnenreich:


I simply want to point out few things to the Court, and that is with respect to the Right of Access.

The Right of Access is clear in the contract made between the Government and the UGDP, and the regulations of HEW, Section 74.23 clearly allow the right of access to copy the records, take the records, analyze the records, and make an audit, so that the contract between the parties is very clearly allowing that.

In addition, I point out to the Court that the FDA which stepped in to do the audit has its own regulation, 21 CFR 20.105(d), which equally makes it, all the raw data, available to any research that had been conducted by the FDA.

The only question I posit for the Court is the question that what we are involved with here is two agencies, one of which would have allowed under its own regulation all the raw data given, as opposed to another agency that does not, and the question is, we are in a question of agency shopping.

Thank you.

Warren E. Burger:

Thank you, gentlemen.

The case is submitted.