Last October 26, in a press conference, President Nixon treated us to a little history lesson:

You remember the famous case involving Thomas Jefferson where Chief Justice Marshall, then sitting as a trial judge, subpoenaed a letter which Jefferson had written which Marshall thought or felt was necessary evidence in the trial of Aaron Burr. Jefferson refused to do so, but it did not result in a suit. What happened was, of course, a compromise in which a summary of the contents of the letter which was relevant to the trial was produced by Jefferson and the Chief Justice of the United States, acting in his capacity as Chief Justice, accepted that. That is exactly, of course, what we tried to do in this instant case.

For accuracy, that statement conforms to the Nixon average: the letter mentioned was not written by Jefferson but to him by General James Wilkinson, who tried to implicate Burr in a conspiracy to dismember the union; it was Burr who argued this letter’s relevance; Marshall never sat on the case “in his capacity as Chief Justice”; and no “summary” of a letter was ever delivered. Most of these errors are probably inadvertent, though the words have all the marks of a Nixon lie: when he is trying to put something over on us, he none-too-subtly affects to be repeating what everybody knows (“you remember…famous case…of course…of course“).

But on one point Nixon seemed to have a case. Scholars as respected as Henry Adams and Edward Corwin have maintained that Jefferson disobeyed and defied Justice Marshall’s subpoena. By last October, President Nixon’s lawyers had used the Jefferson precedent in arguing against Judge Sirica’s subpoena of White House tapes before the Court of Appeals, where they convinced two judges (MacKinnon and Wilkey) that their version of the Jefferson incident was the correct one.

Although the other five judges on the panel disagreed with this reading, Nixon was growing fond of the episode he now elevated to the status of a “rule.” At Disney World on November 17, he expatiated:

I could just say in that respect, too, that I referred to what I call the Jefferson rule. It is the rule, I think, that we should generally follow, a President should follow, with the courts when they want information and a President should also follow with committees of Congress when they want information in his personal files. Jefferson, as you know, in that very, very famous case, had correspondence which, it was felt, might bear upon the guilt or innocence of Aaron Burr. Chief Justice Marshall, sitting as a trial judge, said that—held that Jefferson as President had to turn over the correspondence. Jefferson refused. What he did was to turn over a summary of the correspondence—all that he considered was proper to be turned over for the purposes of the trial—and then Marshall sitting as Chief Justice ruled for the President. Now why did Jefferson do that? Jefferson didn’t do that to protect Jefferson. He did it to protect the Presidency. And that’s exactly what I will do in these cases.

Apparently the President had done some homework since his last use of “the rule.” The letter is no longer by Jefferson in this version, and Marshall does not argue its relevance (though he still, for some reason, sits as Chief Justice). The most significant thing, however, is Nixon’s extension of the rule to cover relations with Congress (something never at issue in the Jefferson incident). God knows what that rule would be covering by now if the greatest living authority on Jefferson’s life had not intervened. Dumas Malone, in a letter to The New York Times—it appeared nine days after the Orlando press conference—took the President to task. Jefferson, he explained, “cannot rightly be said to have rejected it [the subpoena].” This letter seems to have had its effect—Mr. Nixon no longer confidently cites what we all know and remember about “the rule.”

The Malone letter had an interest that ran beyond its immediate purpose. The Burr episode fell within the announced scope of the next volume in his monumental life of Jefferson. Malone gave us a preview of the contribution he meant to make to this incident’s historiography:

The fact does not seem to be generally known that two subpoenas duces tecum were issued to Jefferson on Burr’s motion…. The second subpoena called for the delivery of another letter from Wilkinson. This contained certain passages which, in the opinion of the prosecution, should not be made public. After a good deal of argument, Marshall ruled that the prosecution might not do the deleting but implied that the President himself might. Jefferson struck out passages that reflected on other persons and attested their immateriality. The document was accepted without comment. Thus there was an assertion and recognition of a degree of executive privilege….

So, even though Malone would not let President Nixon get away with a statement that Jefferson rejected the subpoena, he did side with the third President against John Marshall, treating this incident as an early assertion of “executive privilege.” His account of the second subpoena is fresh, concise, and relevant to several controversies. The only thing wrong with it is that it is wrong. The “second subpoena” referred to was not issued to Jefferson, did not concern “another” (a different) letter, and was not the direct subject of Marshall’s ruling.


How could a man of Malone’s stature be wrong on such a point? Was he saying something other than what a first reading might indicate? The next glimpse we had of his treatment just reinforced the first version. Writing to The New York Review of Books (October 4, 1973) to explain this affair, again on the basis of Malone’s research, Steven H. Hochman, Malone’s assistant, told the same story: “In September the defense again requested papers and moved that a second subpoena duces tecum be sent to Jefferson…. Jefferson again complied with the court’s request, but he did not officially acknowledge the subpoena.”

Now we have Malone’s book itself, which singles out for “detailed consideration” the whole subpoenaing of a President. Two chapters are devoted substantially to this one point, and again we are given the Malone version: “the issuance of another subpoena duces tecum to the President…[who] returned the subpoena” without responding to it. And I repeat: that version is wrong.

The proceedings in court were not as confused or confusing as Malone makes them appear. But we have to remember that there were two 1806 letters from James Wilkinson in dispute (that of October 21 and of November 12), two subpoenas issued in 1807 (on June 13 and September 4), and two relevant opinions delivered (on June 13 and September 4). Since the subpoenas and the opinions were delivered on the same dates, and the first letter was primarily at issue on that first date, while the second was being discussed on the second date, it is easy to be tempted toward a reductive symmetry, to say that the first letter was demanded in the first subpoena, the second in the second.

But it is not as simple as that. Burr argued that the first subpoena covered both letters (R ii 502);1 George Hay, the government prosecutor, never contested this particular point, and Marshall referred to it as established in his second opinion (C iii 35, R ii 534). So a question naturally suggests itself—if the letter at stake had already been subpoenaed, why duplicate that earlier subpoena? Now we see why, for reasons other than symmetry, Malone must save the second letter for a separate subpoena: his two subpoenas, which are directed to a single person, are differentiated only by reason of the documents they demand. He ignores the fact that Marshall’s second ruling made a further subpoena superfluous.

The Chief Justice described the situation the court faced on a Friday, September 4, that was to be a very busy one: “Upon an affidavit made by the accused, a subpoena duces tecum has been awarded to the president of the United States, requiring the production of this [i.e., the November 12] letter” (R ii 534). Marshall is not talking of a new subpoena just issuing, but of the first one, as his next sentence indicates: “In consequence of this process the letter was transmitted to the attorney for the United States, accompanied with a communication from the president, authorizing the attorney to exercise his discretion in the case” (ibid.). Hay admitted to possession of the November 12 letter as a consequence of the first subpoena (which called for all the documents relied on by Jefferson when writing his message to Congress about the Burr conspiracy), and he offered to show the complete letter to Burr’s counsel and the court, but insisted on deletions before it could be shown to Burr or read in public.

At this point Benjamin Botts, on Burr’s team of lawyers, moved that a subpoena be issued to Hay, forcing him to surrender the letter (C iii 23-24, R ii 507). This is the action that is debated in the next few pages, until the subpoena is actually served (C iii 27, R ii 513). Hay responded, on the spot, by turning over a copy of the letter with deletions. Burr called this inadequate compliance (C iii 28, R ii 514). Botts argued that the normal course would be to move for the “attachment” (arrest) of Hay, but that he did not want to take such an extreme course—instead he moved that the “cause stand continued” until the letter could be produced in its entirety (C iii 28, R ii 514-519).


This motion led to considerable argument (cf. Wickham for continuance, R ii 525-528, Wirt against continuance, R 528-530)—and it was on this subject that Marshall delivered his opinion, as the opening words of his ruling indicate: “It is not without regret that I find myself constrained to deliver an opinion on the present application. To overrule the motion may, at least, have the appearance of imposing a hardship on the prisoner, and to grant it may occasion delay in a case which all must desire to terminate (R iii 538). The “present application” is for “continuance,” for suspension of the trial until the next court term. At C iii 35 Marshall says, “It is necessary, before the Court could order a continuance of the case, that it should be well satisfied of the materiality of the evidence asked for and refused….”

Malone seems to be confused by Marshall’s ruling because he reads it as an opinion on the issuing of another subpoena, instead of on a request for continuance. But there was no task a subpoena could perform here—the one to Hay had been delivered and responded to, and the defense had dropped the further pursuit of Hay for contempt; and the letter was already covered, so far as the President was concerned, by the first subpoena. What Marshall decided was that there was not sufficient evidence of materiality to grant the motion of continuance, but that the defense was blocked from proving materiality by the prosecution’s refusal to show the letter to the defendant. “It is a very serious thing to proceed to trial [the result of denying continuance] under such circumstances” (R ii 536). “The Court would be very glad to hear farther testimony on the subject, in order to determine properly upon the materiality of the paper called; but that is not offered” (C iii 37). After discussing the various justifying grounds that the President might have for withholding immaterial sections of the letter, he said that these must be presented by the President himself:

It does not even appear to the court that the President does object to the production of any part of this letter. The objection, and the reasons in support of the objection, proceed from the attorney himself, and are not understood to emanate from the president. He submits it to the discretion of the attorney. Of course, it is to be understood that he has no objections to the production of the whole if the attorney has not. Had the president, when he transmitted it, subjected it to certain restrictions, and stated that in his judgment the public interest required certain parts of it to be kept secret, and had accordingly made a reservation of them, all proper respect would have been paid to it; but he has made no such reservation (R ii 536-537, cf. C iii 37).

After this indication that he would listen to direct representations of the President, he made his ruling: “The only ground laid for the court to act upon, is the affidavit of the accused; and from that the court is induced to order that the paper be produced, or the cause be continued” (R ii 537, emphasis added; cf. C iii 37). Then Marshall added a suggested procedure for dealing with the letter on its production.

In short, Marshall granted the defense request unless the prosecution produce the letter. There was no direct demand for evidence from the prosecution (as in a subpoena). If Hay was willing to let the case be dropped, no further claims were made on him or Jefferson. That is the way Hay understood the ruling in his first response to it: “Mr. Hay stated that he would consult general Wilkinson and if he consented, he would produce the letter under the restrictions ordered by the court; preferring that to a continuance of the cause” (R ii 537, cf. C iii 38). The ball was clearly in the prosecution’s court now. They had to volunteer something to keep the case alive. The defense, after getting this ruling in its favor, had no reason to take another (repetitive) step of subpoenaing the letter. It just had to wait for the President not to deliver it, and the misdemeanor trial would for all purposes be dropped. This was probably the real aim of this defense maneuver, occurring as it did just before the impanelment of another jury. The occurrence of all these things on the same day—the demand for the letter, refusal, the issuing of the subpoena to Hay and his prepared response to it, the debate on continuance followed by Marshall’s ruling—suggests that the defense moves were expected.

Only on this reading of things does the prosecution’s action on the morning of Saturday, September 5, the day after Marshall’s ruling, make any sense. Alexander MacRae, a prosecution lawyer, offered a suggestion: “As the President is now at Monticello, is it not practicable to send an express to him, with a subpoena duces tecum?” (C iii 38). Is this, then, a third subpoena, which would be the second one issued to Jefferson? If so, it would not be, as Malone put it, “issued to Jefferson on Burr’s motion,” but on the part of a desperate prosecution trying to keep its case alive against an order for continuance. As MacRae said: “It is advantageous on every consideration to have the trial completed at this term. Almost all the witnesses are present, and perhaps it would be impossible to collect them here again. An express might, perhaps, return in four days from Monticello; and is it not far better to postpone the trial for that period, than to the next term of the court?”

But what would the subpoena MacRae suggested call for? The November 12 letter? That was already in Richmond, and had been subpoenaed last June. What was needed now, and had been suggested by Marshall, was an argument by the President himself for making specific deletions; and a subpoena is not the proper document for eliciting an argument. Hay seems to have realized this, and proposed something more in line with Marshall’s suggested compromise: “Mr. Hay then observed, that he did not conceive himself at liberty to put this letter into the hands of the defendant; but that he would immediately send an express to Monticello for instructions; and that the return would probably be made by Tuesday evening”(ibid).

Does this passage describe the issuing of a subpoena? Hay said that he was sending, not the court; for instructions, not a response. This Saturday session seemed to end in a general willingness to work out some arrangement. “Mr. Burr observed, that some agreement might be made between his counsel and the prosecution, respecting the letter, and that they might go to trial on Monday. The Chief Justice then observed that the Court would meet on Monday; as some arrangement might be, in the mean time, made, respecting the letter.” Hay said he must wait for word from Jefferson, and Luther Martin, the bellicose defense lawyer, said, “Gentlemen need not be so scrupulous, Sir, upon this subject; for we can compel the appearance of the President before this court, with that letter.” We can—not we have taken the step, or will.

The next exchange is the final one recorded on that day. “Mr. Hay. Shall I furnish such parts, in the mean time, as I am disposed to surrender? Mr. Burr. Yes; under the reservation, that such a step does not impair my right to demand the remainder”(ibid). The parties are to meet again Monday and discuss the matter, though no response could be made to a subpoena by then. It is inconceivable that this amicable exchange (amicable except for Martin’s threat, which itself speaks of a weapon held in readiness, not one already used) could be taking place under the affront of a second subpoena freshly issued to the President.

Yet this is the very day Hay is supposed to have sent off that court document with his express messenger, a slave named Bob. His covering letter begins: “Inclosed you will receive a subpoena duces tecum and the letter of the 12th November which it requires. A subpoena was served on me yesterday, upon which I made the return stated in the 3d col. of the 3d page of the inclosed gazette.” He explains that, though the newspaper account shows that he offered the letter on Friday if General Wilkinson did not object, he brought up further reservations when he returned to court on Saturday (the day on which he is writing). He encloses the Wilkinson letter to show, by underlining he had made on it, which parts he had deleted in his response to the subpoena. “I send Bob up expressly for the purpose of procuring your return to the [?indecipherable?] with such parts of the letter as you may think proper to communicate….” Mr. Steven Hochman kindly read me Professor Malone’s transcription of this letter (which is not included in the book) and said that he would conjecture “subpoena” belongs where the decipherment is impossible.

Jefferson responded September 7: “Dear Sir,—I have received, late last night, your favor of the day before, and now re-enclose you the subpoena. As I do not believe that the district courts have a power of commanding the executive government to abandon superior duties & attend on them, at whatever distance, I am unwilling, by any notice of the subpoena, to set a precedent which might sanction a proceeding so preposterous.”

The references to a subpoena in these two letters are the only pieces of evidence that a second subpoena was ever issued to Jefferson. Malone never mentions the subpoena to Hay, which caused a good deal of debate in the transcriptions; yet he does not find it odd that there is not a single trace in those records, or anywhere else, of the issuance of this third subpoena. When Hay took the public part of Jefferson’s September 7 response into court—a certificate vouching for the immateriality of all deleted parts in the letter—he did not mention that a subpoena was at stake, and neither did anyone else, not even Luther Martin. And Justice Marshall did not sound particularly anxious to hear what the President had to say: “The Chief Justice did not know whether there was any necessity for it, if there was any difficulty of the bar. Mr. Hay. None at all, I assure you.” One of Burr’s lawyers volunteered that he would like to hear the certificate. Hay read it, and we hear no more about it after that. This seems a rather snoozy enforcement of a subpoena that Marshall is supposed to have been intent upon issuing only four days earlier.

Still, there the subpoena is, in the first sentence of both letters. Where did it come from? I think the answer is in Jefferson’s statement that he is “re-enclosing” the subpoena. Why? What was Hay supposed to do with it? Give it back to the court? But he did not do that. And such an act would seem a needless affront to the court, one that Marshall would not take napping. Was Hay supposed to keep it for himself? What for—as a souvenir? Jefferson did not formally respond to the first subpoena, either—but he did what we expect in a man so careful of his records; he put it among his papers, where it is still to be found. The second one he sent back, for no apparent reason, not even explaining the mystery to Hay.

The only logical assumption is that he did not have to give a reason for returning it to Hay because it belonged to Hay. It was the subpoena that had been served on Hay in the courtroom Friday. Hay, rushing to get his message off Saturday after the session was over, included his copy of the subpoena, just as he did the newspaper account of Marshall’s decision and the Wilkinson letter with its deletions marked—so that Jefferson would have a file of everything that was relevant to the instructions he would be sending back. (Jefferson does not return “the gazette,” since that is not Hay’s property in any important sense.)

Jefferson, returning the subpoena without any comment on the use to be made of it (one just naturally returns things to their owner), said he would not dignify the thing with any formal notice, and then took notice of it in blistery terms for Hay’s ears only—he was still testy about the subpoena served on him in Washington. He repeats some of his earlier arguments on the use of subpoena power against “the executive.” True, the subpoena to Hay did not touch Jefferson directly—it was Hay, after all, who was threatened with imprisonment for defiance. But it aimed at countermanding his orders to the district attorney, and it was natural for Jefferson to resent another subpoena meant to take documents out of his custody, however remote.

If we adopt this interpretation, then the major part of Hay’s letter makes a good deal better sense. Though the newspaper account of Marshall’s ruling was not strictly accurate, he told Jefferson, and the ruling itself was “far from being explicit,” Hay concentrates on the point that only Jefferson and not his delegate could single out the deletions, and gives it as his own opinion that Marshall “seemed to think that any return from the Ps. [President] himself would be sufficient.” Is that any way to describe a Chief Justice in so little conciliatory a mood that he has just issued a new subpoena demanding a specific document?

This shows that the whole issue remained, in Hay’s mind, further instructions on deletion in response to Marshall’s ruling on continuance. Hay is assuming that this is all that will be required, and does not even discuss the possibility of Jefferson’s turning over the whole letter in response to a subpoena, which would be in the front of his attention if he had just served up such a demand. (Malone, working from his prior belief in a second subpoena, conjectures that Hay wrote about “your return to the [subpoena] with such parts of the letter as you may think proper to communicate”—making Hay presume ahead of time that Jefferson will defy the subpoena. Surely it is more likely that he spoke of Marshall’s ruling—under one verbal form or another—in the indecipherible spot.)

More than that: after indicating what he thinks will be sufficient to prevent continuance, Hay goes on to suggest that it might be just as well to let the ruling go into effect, since “these prosecutions will terminate in nothing.” He gives a gloomy prognosis of the misdemeanor action just beginning (the treason trial was already concluded by this time) and suggests “it would be well to dismiss these indictments.” Again, the assumption is that Jefferson is at liberty to let things lapse; whereas, if he were under subpoena, he would have to respond.

The whole letter is obviously concerned with response to the continuance ruling, which called for argument of immateriality in the deletions coming straight from the President; it does not directly address the challenge of any fresh subpoena to Jefferson. And the President’s answer perfectly fits the same context. He does three things: sends the proper certificate addressing the court on the subject of the deletions, refuses to notice the returned subpoena, and goes on to urge that the case be prosecuted further, adding Madison’s insistence to his own lest Hay give in to the maneuvers for “continuing.”

Thus the references to “a subpoena” in these two letters fit much better with the subpoena that we know about, whose existence cannot be doubted, and which is amply attested in the record—the one issued to Hay—and not to a phantom subpoena issued we know not where or when, disappearing without a trace, and casually ignored in every other spot except these letters, and even in the bulk of these letters themselves. Yet, as I said, these two references are the only indication that a second subpoena was ever served on Jefferson. Malone, in a footnote, admits that he cannot find the second subpoena (he has seen the original of the first one and the copy served on Jefferson), but he says that a copy of the court order calling for this second subpoena to Jefferson is contained in the Richmond court records.

Mr. Hochman, once more, generously supplied me with the material Malone is referring to, and it is clear that the document in question is not an independent witness to a second Jefferson subpoena, but a paper Malone misinterpreted because of his prior assumption that a second subpoena must have existed. The text in question is a rough draft on a loose slip of paper, and it is undated. It is a court order, “on the motion of A.B.” (Aaron Burr), “that a subpoena duces tecum be awarded to summon the President of the United States or such of the secretaries of the departments as may have the papers herein mentioned, or any of them.” But only one paper is then mentioned: “A letter from General James Wilkinson to the President of the United States.” The first subpoena called for a number of papers, and has a form different from this court order, so Malone assumed that this order was aimed at the single letter that was still a matter for dispute in September.

But if that were the case, why would Marshall be subpoenaing the President “or secretaries of departments” for a letter that he knew was in Hay’s possession? Besides, the argument over various Wilkinson letters had reached a point in September where it was clear that the November 12 one was the center of attention.

This undated document is clearly the court order for the first subpoena. One document is mentioned in it, exempli gratia, but Burr would be allowed to spell out his demands at leisure (he actually wrote a personal condition on the subpoena after it was drawn up). In the same way, what is generally indicated in the court order (“or such of the secretaries of the departments”) is made precise in the subpoena as it was then drawn up: it was addressed “to Thomas Jefferson, to Robert Smith, to Henry Dearborn, or either of them who may have the papers herein…” (Smith and Dearborn were the Secretaries of Navy and War, the precise men generally referred to in the court order). The court order, though it lists only one letter, envisages a further listing, along with the possibility of collecting papers from three different sourcesnone of which, it would have been known in September, actually possessed the November 12 letter. The court order looks to the circumstances of the first subpoena issued to Jefferson, because that first subpoena was the only one ever issued to him. The order differs from the subpoena in its form because it is not itself a subpoena, but only an instruction that one should be awarded.

It may seem niggling to sort out these details at such length, but the consequences are important. Malone, correcting Nixon, says Jefferson did not exactly defy the subpoena when he delivered the letter with its deletions, since he had complied with the demand in great part. To which it is fair to answer that even partial non-compliance with a subpoena, if deliberate, is a defiance. But now it seems clear that Jefferson was not defying a fresh subpoena by his delivery of the November 12 letter2 in deleted form—on the contrary, he was making exactly the response that Marshall’s ruling had recommended.

Yet more than that is at issue. Malone argues that, by refusing to turn over the whole letter, Jefferson made “an assertion…of a degree of executive privilege,” and that Marshall by his tacit acceptance of this assertion had given a de facto “recognition of a degree of executive privilege.” If so, this would be a striking early example of that slippery doctrine. But ’tain’t so.

In any event, the use of this incident in courts, in Justice Department memoranda (dating all the way back to the famous one issued by William Rogers in 1958), in debates about the limits of the executive and the judiciary, in current journalism about the responsiveness of Richard Nixon to the courts, and in many areas outside the quiet world of historical scholars, makes it a matter of economy, not wasteful profusion, to get the facts sorted out clearly. We could have hoped, up to the appearance of this volume, that Malone would settle the matter with his customary accuracy; but he was misled, as others have been, by those two references to a subpoena in the Hay-Jefferson correspondence.

With another criticism of this emphasis on the Jefferson subpoena I would heartily concur—that the judgment of a major work by a great scholar like Malone should not be concentrated on a single error, no matter how important that error might be to current disputes. That is why I reserve a consideration of the book as a whole for this journal’s next issue.

(This is the second of three articles on recent Jefferson studies.)

This Issue

May 2, 1974