73 U.S. 499
18 L.Ed. 890
6 Wall. 499
December Term, 1867
ERROR to the Circuit Court for the Southern District of New York.
The Constitution of the United States says, under the legislative head, as follows:
'Every bill which shall have passed the House of Representatives and the Senate shall, before it becomes a law, be presented to the President of the United States; if he approve he shall sign it, but if not, he shall return it, with his objections, to that House in which it shall have originated. . . . If any bill shall not be returned by the President within ten days (Sunday excepted) after it shall have been presented to him, the same shall be a law in like manner as if he had signed it.'
And an act of September 15th, 1789, creating the Department of State, provides that whenever a bill, order, resolution or vote of the Senate and House of Representatives, having been approved and signed by the President of the United States, or not having been returned by him with his objections, shall become a law, or take effect, it shall forthwith thereafter be received by the said Secretary from the President, and he shall carefully preserve the originals, and cause them to the recorded in books provided for that purpose. An act of July 7th, 1838, dispenses with this recording in a book.
With these provisions in force, Congress passed through both houses, in December, 1861, a bill which declared 'that from and after the date of the passage of the act,' the duties on tea should be twenty cents per pound. A previous statute had fixed the duty at fifteen cents. The roll of the engrossed bill was taken to the President, and by him thus signed, no year being indicated:
'APPROVED December 24.
The record kept in the office of the Secretary of State showed, however, that this enrolled statute, with the President's approval on it, was filed in that office December 26th, 1861, and the journal of the House of Representatives, in Congress, showed that a message was received from the President January 6th, 1862, stating that on the 24th day of the preceding month he had approved this bill.
In the volume of the statutes of the United States, published by authority in 1863, the act was presented with an approval thus indicated:
'APPROVED December 24 .'
In this state of things, Gardner, in 1864, entered at the custom-house in New York certain packages of tea, on which the collector of the customs there, assuming that there was a statute laying that duty, required him to pay twenty cents per pound. Gardner declined to pay twenty cents per pound on the ground that there was no statute fixing that duty, but offered to pay fifteen cents, the duty fixed by what he asserted to be the only act in the case. Being compelled to pay the twenty cents, and having paid it under protest, he brought suit in the court below to recovert he excess. The court below gave judgment against him, and on error here the question was, whether the bill fixing the twenty cents had passed, or, in other words, whether it was a law on the 28th April, 1864, when the teas in question were entered.
Mr. George Ticknor Curtis, for the plaintiff in error:
The President's certificate on the roll is a record of the strictest character. It cannot be explained, controlled or aided by any other evidence whatever. This follows from the requirement of the Constitution, that if he approve the bill he shall sign it. That the roll is in all cases the record or 'original' of a statute, is shown by the acts of Congress of September 15th, 1789, and of July 7th, 1838.
Now, whatever methods may be adopted of proving copies of the roll itself, no proof can be admitted to supply what the record does not contain. A law may be construed, judicially, if the means of ascertaining the legislative intention exist in the law itself. But if there is a positive omission in a law of what is essential to its operation, the omission cannot be supplied. A record imports absolute verity. It is of so high a nature, says Lord Coke, that it can be tried only by itself. And that this principle extends by the common law to the records of statutes, is evident from the rule that a statute cannot be proved from a journal of Parliament, but must be proved from the roll, which is the record; and from the further rule, that if it purports to be a general statute, the judges will take notice from the record whether it be a statute or not, and thus the plea of nul tiel record, or denial that there is such a record, cannot be interposed.1 These principles appear to have been adopted into our legislation, which makes the bill signed by the President, and deposited in the Department of State, the 'original,' or record of the statute.
If, then, there be in the Department of State a record which purports to contain a general statute, the judges will take notice of that record, and on it will proceed to determine whether it be a statute or not.
The date of the executive approval of a bill is an essential part of this record, and it is, under our Constitution, and in the modern English practice, necessarily the date of the passage of the law. In England, prior to 1792-3, all acts took date from the first day of the session. Great mischiefs followed from the enforcement of this rule. They culminated in Latless v. Holmes,2 where an act by its terms was to take effect from and after its passage. It was held that the time of its actual passage could not be shown; and that there could be no relief against its retrospective operation, great as the hardship manifestly was. This led to the act, 33 Geo. III, chap. 13, which directed that the day, month and year of the royal assent be indorsed on the roll, and that such indorsement be taken as a part of the act, and as the date of its commencement when no other commencement is therein provided.
As our Constitution does not permit any bill to become a law before it has been presented to the President, and requires him, if he approves it, to sign it, those who fixed our first precedents, which have never been departed from, established by them the rule that the President must record the date of his approval, and this must be the date of the passage of the law. Matthews v. Zane3 fully recognizes this rule.
Now, a law which is to operate from the date of its passage, and which has yet no date, can have no operation, especially if its provisions would supersede some former law. The former law remains in force,4 and in order to make a date, the year, as well as the day of the month, must appear. How is it to be known, judicially, in what year occurred that 24th day of December on which the President signed this roll?
Parol evidence is out of the question. That would break in upon the rule that a record must prove itself, and would oblige the citizen, when seeking the date of the President's approval, to inquire whether there are living witnesses who can prove that date.
Other records, such as the journals of the two Houses, would require an inference to be drawn, that is to say, they supply defects in one record by arguments from another, and so break down the rule that a record must prove itself.
A resort to the calendar would afford no aid here, because the 24th day of December occurs in every year, and consequently the calendar of any year since the passage of the former law, which made the duties fifteen cents per pound, would determine nothing as to the year in which the President signed a law making the duties twenty cents.
Our conclusion is,
1. That the President alone can make the record which is to show the date of his approval.
2. That if the President's record is defective in respect to the year when it was made, no resort can be had to extrinsic evidence to supply that defect.
Mr. Ashton, contra:
1. The Constitution says that if the president approves, he shall SIGN. It requires nothing but his signature. The word 'approved' is surplusage. And for the same reason, and to the same extent, in a legal sense, is the date of approval. In a practical sense it may be, and is important. The best evidence of that time is, of course, the contemporaneous memorandum of the signer himself: and usage has accepted that memorandum, not as a record, perhaps, but as the best proof that the nature of the case admits. But what is the rule when the usual and conclusive evidence of the time of signing is absent? That the bill has been approved is certain. If so, it has become a law. Shall this be made null—declared not to have been at all—the approval, the signing, the going into offect, contrary to the truth of the case, because we cannot admit unquestionable evidence of the day when the final act was done? The question seems to carry its own answer.
2. But conceding that the time of the approval can only be proved by the record, what is the 'record?' It consists of the recorded proceedings connected with, and leading to, and following after, the law—the journals of Congress, the records and files of the office of the Secretary of State,—and all these may be resorted to for the purpose of determining the time when the approval of the President occurred.
In the Matter of Welman,5 the court say:
'It may be necessary and admissible in some instances, particularly when an act becomes a law by not being either signed or returned with objections, to carry back the inquiry to the legislative journals. But it would be unsafe, as it would be unfit, to allow the commencement of a public law, whenever the question may arise, to depend on the uncertainty of parol proof, or upon anything extrinsic to the law and the authenticated recorded proceedings in passing it.'
So cases settle that the court may inspect those journals to correct clerical mistakes, or carelessness.6 Even to correct an erroneous entry of the date of approval;7 or to ascertain whether an act was passed by ayes and nays.8 The court may inquire whether an act, coming within the twothird clause of the Constitution, have passed by the requisite number of votes.9 Parol evidence will be received, too, to show that an actual signing of the bill, as approved, was done by mistake.10
In Pennsylvania, the constitution does not require a bill to be signed by the speakers of the two houses of the legislature; and it was there held, that the signatures, though proper, were not essential to the validity of the law. The fact of its passage and approval was held to be provable by the certificate of the Secretary of State that the bill was duly enrolled in his office.11
Mr. Justice MILLER delivered the opinion of the court.
The date of the President's approval of the bill is undoubtedly the date at which it became a law, if it ever did. In the volume of the statutes now before us, published in 1863, the approval is dated December 24th , but the figures 1861 are in brackets, by which it is understood that no such figures are found in the original enrolled act on file in the Department of State. And it is conceded that on inspection, the roll shows on the face of the bill no other date for the approval of the President than the day of the month already stated.
It is not denied that the President's signature to the bill is genuine, and that he did approve it. The volume of the United States Statutes at Large, which contains this act, was published by authority the year before the entry was made of his tea by the plaintiff. The record kept in the office of the Secretary of State shows that this enrolled statute, with the President's approval on it, was filed in that office, December 26th, 1861. The journal of the House of Representatives in Congress shows that a message was received from the President, January 6th, 1862, stating that on the 24th day of the preceding month he had approved this bill. So that, if we can look to any of these sources of information, the court can have no doubt that the bill was in force as a statute at the time the duties on plaintiff's tea became chargeable.
The whole of the very able and ingenious argument of counsel for plaintiff rests on these two propositions, as stated in his own language: 'That the President alone can make the record which is to show the date of his approval; and that if the President's record is defective in respect to the year when it was made, no resort can be had to extrinsic evidence to supply that defect.'
The first of these propositions assumes that no act of Congress can become a valid statute, unless some official written statement is found in it of the precise date when the President approved it, and that it is a part of the duty of the President to make this statement; a duty so important that unless made by him, and by no one else, all the previous proceedings of the two Houses of Congress, and the approval of the President, and his signature attesting that approval, are all vain and nugatory.
We should reasonably expect to find a duty so very important as this, the neglect of which is followed by such serious consequences, prescribed by some positive and express provision of the Constitution, or, at least, by some act of Congress.
The only duty required of the President by the Constitution in regard to a bill which he approves is, that he shall sign it. Nothing more. The simple signing his name at the appropriate place is the one act which the Constitution requires of him as the evidence of his approval, and upon his performance of this act the bill becomes a law.
'Every bill which shall have passed the House of Representatives and the Senate shall, before it becomes a law, be presented to the President of the United States; if he approve, he shall sign it, but if not, he shall return it, with his objections, to that House in which it shall have originated.' 'If any bill shall not be returned by the President within ten days (Sunday excepted) after it shall have been presented to him, the same shall be a law in like manner as if he had signed it' Here are two courses of action by the President in reference to a bill presented to him, each of which results in the bill becoming a law. One of them is by signing the bill within ten days, and the other is by keeping it ten days, and refusing to sign it. Even in the event of his approving the bill, it is not required that he shall write on the bill the word approved, nor that he shall date it.
If a date by the President is essential to the validity of the statute, it must be as essential when he retains the bill and fails to sign it as when he signs it. It is his action in retaining the bill for ten days which makes it a law as much as it is in signing it. Yet, in the latter case, no evidence is required of the President, either by the Constitution or in actual practice, to show that he had ever received or considered the bill.
It is not possible, therefore, to hold that the Constitution, either expressly or by just implication, imposes upon the President the duty of affixing a date to his signature to a bill.
Nor does any act of Congress require him to do this. The statutes of September 15th, 1789, and of July 7th, 1838, so far from requiring the President to affix a date to his act of signing bills, provide another means of ascertaining when a statute takes effect, namely, by finding it on file in the office of the Secretary of State; for by this statute all such bills, orders, resolutions or votes of Congress as shall become laws, or shall take effect, are to be received from the President and filed in that office. The duty, then, of making such memorandums as shall show when they were received by this Department, in which the rolls are to remain permanently, and where alone they can be inspected, is much clearer than any such duty on the part of the President. As the only valuable purpose of having a date is to determine when the statute takes effect, it is reasonable that this should be made by the officer who receives it from the President forthwith, and who is to be the future custodian of the statute—who alone can give certified copies of it, and from whose office the legally authorized publisher receives the copy from which it is printed.
If neither the Constitution nor the statutes impose this duty upon the President, we are equally unable to find anything in the practice of the English Parliament to sustain this view. The custom there anciently was for the enrolled bill, on receiving the assent of the King, generally given by commission in Parliament, to be delivered, with the statement of this fact indorsed on it, to the clerk of Parliament. From thence transcripts were sent to the sheriffs of the counties, who were ordered to proclaim them in their county courts, where the transcripts were filed for reference. Since the art of printing, this latter custom has been abandoned. But an act of 33 George III, chap. 15, requires the clerk of Parliament to indorse the date of the King's approval upon the roll of each statute, which is to be the date from which it shall take effect.12 The enactment of such a statute shows that no rule had previously existed, that the date was affixed by the King or by the commissioners who, in his name, gave his assent to the bill.
The second proposition, that 'if the President's record is defective in respect to the year when it was made, no resort can be had to extrinsic evidence to supply that defect,' is still more at variance with both principle and authority than the one we have just considered.
The statute under consideration is a public statute, as distinguished from a private statute. It is one of which the courts take judicial notice, without proof, and, therefore, the use of the words 'extrinsic evidence' are inappropriate. Such statutes are not proved as issues of fact as private statutes are. But if we suppose the phrase to have been used to express the sources of information to which the court may resort, the proposition is still inadmissible.
In point of morl force in producing conviction in the mind that a bill was signed on a given day, there may be often found stronger evidence than the date accompanying the signature. It is general experience that mistakes are often made in such dates. So well is this understood that the general rule of law that parol evidence cannot be received to contradict a written contract, does not apply to the date, which, though forming a part of the written instrument, may be contradicted whenever it is material to the issue to do so. So also written contracts, or other instruments having no date on their face, may have the time of their execution proved by parol or other competent testimony. It is believed that this principle would be applicable to any instrument in writing offered to a jury on an issue of fact even if it were a private statute, always requiring, however, the best evidence of the date that exists. But the argument we are considering imposes upon the judges who are to take judicial notice of a statute, a more limited range of search for information than that which is open to a jury, when the rule of judicial notice, as we shall show hereafter, was adopted for the purpose of enlarging it.
The record of the Secretary of State of the time of filing such a paper, the journals of the two Houses of Congress, the message of the President, and other circumstantial facts, may produce stronger conviction of the day and of the year in which the bill was signed, than the date affixed by the President. There is no reason, then, on sound principle, why the court should confine itself to the date made by the President, or, if he has made none, should reject all other sources of knowledge. The judicial notice of the court must extend, not only to the existence of the statute, but to the time at which it takes effect, and to its true construction.
This view of the subject is well supported by authority.
In the learned work of Mr. Dwarris on Statutes13 we are told that the principal reason of the rule that the courts should take judicial notice of public statutes, and should not permit them to be put in issue as private statutes are, was that many ancient statutes were no longer to be found, which yet were within the time of legal memory, and could not, therefore, be treated as common law. In order to prevent their existence being brought to the test of proof by record, the principle was adopted that the court should take notice of them; and that the judges are to inform themselves in the best way they can.
This is confirmed by Sir Matthew Hale in his History of the Common Law.14 Alluding to these statutes, of which there are many that are no longer to be found among the rolls, he says: 'An act of Parliament, made within the time of memory, loses not its being so, because not extant of record, especially if it be a general act of Parliament. For of general acts of Parliament the courts of common law are to take notice without pleading them. And such acts shall never be put to be tried by the record upon an issue of nul tiel record, but it shall be tried by the court, who, if there be any difficulty or uncertainty touching it, or the right of pleading it, are to use for their information ancient copies, transcripts, books, pleadings, and memorials, to inform themselves, but not to admit the same to be put in issue by a plea of nul tiel record. For, as shall be shown hereafter, there are many old statutes which are admitted, and obtain as such, though there be no record at this day extant thereof; nor yet any other written evidence of the same but which is in a manner only traditional, as namely, ancient and modern books of pleadings, and the common received opinion and reputation and approbation of the judges learned in the law.'15
Lord Coke,16 giving an account of the manner in which the statutes were formerly published in the county courts, in regard to h ich he had made diligent search, observes that 'although proclamation be not made in the county, every one is bound to take notice of that which is done in Parliament, for as soon as Parliament hath concluded anything, the law intends that every person hath notice thereof, for the Parliament represents the body of the whole realm, and therefore it is not requisite that any proclamation be made, seeing the statute took effect before.' If this proposition be sound, of which there seems to be no reason to doubt, how can it be held that the judges, upon whom is imposed the burden of deciding what the legislative body has done, when it is in dispute, are debarred from resorting to the written record which that body makes of its proceedings in regard to any particular statute?
The courts of last resort in several of the States have expressly decided that this may be done.17
In the Prince's case,18 the rule on this subject is laid down by the court in the following language: 'As to the fourth point it was resolved, that against a general act of Parliament, or such whereof the judges ex afficio ought to take notice, the other party cannot plead nul tiel record, for of such acts the judges ought to take notice. But if it be misrecited the party ought to demur in law upon it. And in that case the law is grounded upon great reason, for God forbid, if the record of such acts should be lost, or consumed by fire or other means, that it should be to the general prejudice of the commonwealth, but rather, although it be lost or consumed, the judges either by the printed copy, or by the record in which it was pleaded, or by other means, may inform themselves of it.'
In this case the Lord Chancellor was assisted by a judge from each of the common law courts, of whom Coke was one, and the decision as reported by him, and the reason on which it was founded, are entitled to the highest consideration.
We are of opinion, therefore, on principle as well as authority, that whenever a question arises in a court of law of the existence of a statute, or of the time when a statute took effect, or of the precise terms of a statute, the judges who are called upon to decide it, have a right to resort to any source of information which in its nature is capable of conveying to the judicial mind a clear and satisfactory answer to such question; always seeking first for that which in its nature is most appropriate, unless the positive law has enacted a different rule.
Coke, Littleton, 117b, 98b; Comyn's Digest, tit. Parliament R. 3, 4; Pacific Railroad v. The Governor, 23 Missouri, 353; People v. Devlin, 33 New York, 269.
4 Term, 660.
7 Wheaton, 211.
Opinion of the judges of Massachusetts, 3 Gray, 606, 607; Rex v. Biers, 1 Adolphus & Ellis, 327; Langley v. Haynes, Moore, 302; Gibbs v. Pike, 8 Meeson & Welsby, 223.
20 Vermont, 656.
Supervisors v. Heenan, 2 Minnesota, 330; Turley v. County of Logan, 17 Illinois, 151.
Fowler v. Peirce, 2 California, 165.
Spangler . Jacoby, 14 Illinois, 297.
People v. Purdy, 2 Hill, 31; Hunt v. Van Alstyne, 25 Wendell, 605.
People v. Hatch, 19 Illinois, 283.
Speer v. Plank Road Co., 10 Harris, 378.
Bacon's Abridgment Statutes, letter C.
Pages 14, 16.
See 1 Kent's Commentaries, 460; Sedgwick on Statutes and Constitutional Law, 34.
4 Institutes, 26.
Purdy v. The People, 4 Hill, 384; De Bow v. The People, 1 Denio, 9; Spangle v. Jacob, 19 Illinois, 283; Young v. Thomson, 14 Id. 297; Speer v. Plank Road, 22 Penna. State, 376; Matter of Welman, 20 Vermont, 656; Supervisors v. Heenan, 2 Minnesota, 330; Fowler v. Pierce, 2 California, 151.
8 Reports, 28.