Full opinion text
L. HAND, Circuit Judge. The defendants Dennis and others appeal from a judgment of conviction upon an indictment for violation of Section 3 of the “Smith Act,” that is, for “wilfully and knowingly” conspiring to organize the Communist Party of the United States as a group to “teach and advocate the overthrow and destruction” of the government “by force and violence,” and “knowingly and wilfully to advocate and teach the duty and necessity of overthrowing and destroying” the government by “force and violence.” All the defendants were at one time or another officials of the Party during the period laid in the indictment — April 1,1945, to July 20,1948. The case was tried at great length. The defendants challenged the array, and the trial of that issue extended from January 20, 1949, to March 1, 1949; the trial of the issues began the following week and went on continuously until September 23,1949. The jury brought in a verdict against all the defendants on October 14, 1949, and they were sentenced on October 21, 1949. The trial of the challenge to the array took 23 days; the government’s case on the issues took 40 days, and the appellants, 75 days. Logically the first issue, and incidentally the most important, is whether the evidence was sufficient to support the jury’s verdict that the defendants were guilty of the crime charged in the indictment. There was abundant evidence, if believed, to show that they were all engaged in an .extensive concerted action to teach what indeed they do ■ not disavow — the doctrines of Marxism-Leninism. These doctrines were set forth in many pamphlets put in evidence at the trial, the upshot of which is — indeed at; honest jury could scarcely have found otherwise — that capitalism inescapably rests upon, and must perpetuate, the oppression of those who do not own the means of production; that to it in time there must and will succeed a “classless” society, which will finally make unnecessary most of the paraphernalia of government; but that there must be an intermediate and transitional period of the “dictatorship, of the proletariat,” which can be established only by the violent overthrow of any existing government, if that be capitalistic. No entrenched bourgeoisie, having everything to lose and nothing to gain by the abolition of capitalism, by which alone it can continue to enjoy its privileged position, will ever permit itself to be superseded by the means which it may have itself provided for constitutional change: e. g., by the ballot. No matter how solemnly it may profess its readiness to abide the result, and no matter how honestly and literally the accredited processes of amendment may in fact be followed, it is absurd to expect that a bourgeoisie will yield; and indeed to rely upon such a possibility is to range oneself among the enemies of Marxist-Leninist principles. Therefore the transition period involves the use of “force and violence,” temporary it is true, but inescapable; and, although it is impossible to predict when a propitious occasion will arise, one certainly will arise: as, for example, by financial crisis or other internal division. When the time comes the proletariat will find it necessary to establish its “dictatorship” by violence. The defendants protest against this interpretation of their teaching and advocacy. They say that the use of “force and violence” is no part of their program, except as it may become necessary after the proletariat has succeeded in securing power by constitutional processes. Thereafter, being itself the lawful government, it will of course resist any attempt of the ousted bourgeoisie to regain its position; it will meet force with force as all governments may, and must. If the defendants had in fact so confined their teaching and advocacy, the First Amendment would indubitably protect them, for its protects all utterances, individual or concerted, seeking constitutional changes, however revolutionary, by the processes which the Constitution provides. Any amendment to the Constitution passed in conformity with Article V is as valid as though it had been originally incorporated in it; the only exception being that no state shall be denied “its equal Suffrage in the Senate.” It is unnecessary to quote in detail the many passages in the pamphlets and books, published and disseminated by the defendants, which flatly contradict their declarations that they mean to confine the use of “force or violence” to the protection of political power, once lawfully obtained. The prosecution proved this part of its case quite independently of the testimony of its witnesses, though the jury might have relied upon that, had it stood alone. The sufficiency of the evidence therefore comes down to whether it is a crime to form a conspiracy to advocate or teach the duty and necessity of overthrowing the government by violence, and to organize the Communist Party as a group so to teach and to advocate. This being true, three questions arise: (1) whether the Act is constitutional as the judge construed it, (2) whether his construction was right, and (3) whether the evidence was admissible under the indictment. To the last of these we shall devote no time, for it is patent on the merest inspection that the indictment is sufficient; even had it not been, any variances would have been harmless error. Fed. Rules Crim.Proc. rule 52(a), 18 U.S.C.A. Coming then to the first point, although the interest which the Amendment was designed to protect — especially as regards matters political — does not presuppose that utterances, divergent from current official opinion, are more likely to be true than that opinion, it does presuppose that official opinion may be wrong, and that one way — and perhaps the best way— to correct or supplement it, is complete freedom of criticism and protest. This may convince the officials themselves, and in any event it may rouse up a body of contrary opinion to which they will yield, or which will displace them. Thus, the interest rests upon a skepticism as to all political orthodoxy, upon a belief that there are no impregnable political absolutes, and that a flux of tentative doctrines is preferable to any authoritative creed. It rests upon a premise as yet unproved, and perhaps incompatible with men’s impatience of a suspended judgment when the stakes are high. However, it concerns beliefs alone, not actions, except in so far as a change of belief is a condition upon action. Nobody doubts that, when the leader of a mob already ripe for riot gives the word to start, his utterance is not protected by the Amendment. It is not difficult to deal with such situations; doubt arises only when the utterance is at once an effort to 'affect the hearers’ beliefs and a call upon them to act when they have been convinced. As a new question it might have been held that the Amendment did not protect utterances, when they had this double aspect: i. e., when persuasion and instigation were inseparably confused. In that view the Amendment would give protection to all utterances designed to convince, but its protection would be conditional upon their not being part of, or coupled with, provocation to unlawful conduct, whether that was remote or immediate. True, one does not become an accessory to a crime who “counsels, commands, induces * * * its commission.” unless the crime is committed; but he will be guilty of conspiracy by the mere agreement; and it will not protect him that the objective of the conspiracy is lawful, and only the means contemplated are illegal. Had this view of the Amendment been taken, although the utterances of these defendants so far as they attempted to persuade others of the aims of Communism would have been protected, they would have lost that protection, coupled as they were with the advocacy of the unlawful means. And that is probably in fact true of utterances not political or religious; for it is at least doubtful whether other kinds of utterance, however law ful in so far as they were persuasive only, would retain their privilege if coupled with appeals to unlawful means. One can hardly believe that one would be protected in seeking funds for a school,, if he suggested that they should be obtained by fraud. His privilege would be conditional upon separating the means from the end. However, that may be, it is not true of political agitation and the question is ,what limits, if any, the advocacy of illegal means imposes upon the privilege which the aims or purposes of the utterer would otherwise enjoy. The Supreme Court has certainly evinced a tenderness towards political utterances since the first World War. In Schenck v. United States, 249 U.S. 47, 39 S.Ct. 247, 63 L.Ed. 470, the accused had been convicted of distributing a broadside which obstructed the draft, and his conviction was affinned. It is clear that this was upon the theory that his purpose had been to bring about the “substantive evil,” with which it was within the power of Congress to deal. Holmes, J., went on to say, however, by way of limitation, that even in such cases the danger that the “evil” might be realized must be “clear and present”; though he must have supposed that the distribution of the broadside created such .a danger. In Abrams v. United States, 250 U.S. 616, 40 S.Ct. 17, 63 L.Ed. 1173, the majority of the Court on the other hand appear to have found it enough that the purpose of the accused was to bring about the “substantive evil”; at least they did not raise any question as to its gravity or its imminence; but Holmes and Brandéis, JJ., dissented, first, because they did not think that the accused’s purpose was to hinder the prosecution of the war, and second, because the danger was trivial' and remote. Schaefer v. United States, 251 U.S. 466, 40 S.Ct. 259, 64 L.Ed. 360, concerned a situation substantially the same, and adds nothing, so far as we can see. Next came Gitlow v. People of State of New York, 268 U.S. 652, 45 S.Ct. 625, 69 L.Ed. 1138, where a New York statute, under which the accused had been convicted, was so close to parts of the statute at bar that it must have been its model pro tanto. The majority held that it was enough that the pamphlet, broadcast by the accused, “advocates arid urges in fervent language mass action which shall * * * overthrow and destroy organized parliamentary government”; and that it used “the language of direct incitement”. 268 U.S. at page 665, 45 S.Ct. at page 629, 69 L.Ed. 1138. Once more Holmes and Brandéis, JJ., however, dissented because “there was no present danger of an attempt to overthrow the government by force on the part of the admittedly small minority who share the defendant’s views. * * * If the publication * * * had been laid as an attempt to induce an uprising * * * at once and not at some indefinite time in the future, it would have presented a different question,” unless perhaps even then “it. was not futile and too remote from possible consequences”. 268 U.S. at page 673, 45 S. Ct. at page 632, 69 L.Ed. 1138. This case arose under state law, but all the judges agreed that the First Amendment applied to it, though possibly without so strong a grip as though it had been a federal law. It has been often cited in the twenty-five years that have passed, never with disapproval and frequently as authoritative. On the other hand it would be uncandid to say that the dissent did not make the immediacy of the “substantive evil” a condition of the unlawfulness of even direct incitement to rebellion; though it is to be noted that the dissenters also dwelt upon the improbability that the paltry efforts of the accused could effect their purposes within any period which need to be reckoned with. Whitney v. People of State of California, 274 U.S. 357, 47 S.Ct. 641, 71 L.Ed. 1095, which followed shortly thereafter, presented no different question, and the concurrence of Brandéis, J., was explicit' that he would hold a conspiracy protected, though its execution were deferred to the first propitious moment. 274 U.S. at pages 376, 377, 47 S.Ct. 641, 71 L.Ed. 1095. Yet here too the reasoning should be remembered, for it was that delay in execution would give opportunity for the corrective of public discussion. It does not follow that he would have been of the same opinion, if the conspirators had sought to mask their purposes by fair words, as they did in the case at bar. Moreover, it is to be particularly observed that the concurrence said that the Court had not yet fixed the standard by which to determine “when a danger shall be deemed clear; how remote the danger may be and yet be deemed present”. 274 U.S. at.page 374, 47 S.Ct. at page 648, 71 L.Ed. 1095. In Stromberg v. People of State of California, 283 U.S. 359, 51 S.Ct. 532, 75 L.Ed. 1117, 73 A.L.R. 1484, the majority reversed a conviction because it thought the statute too vague to serve as a guide to conduct; but it explicitly recognized “that the State may thus provide for the punishment of those who indulge in utterances which incite to violence and crime and threaten the overthrow of organized government by unlawful means. There is no constitutional immunity for such conduct abhorrent to our institutions. Gitlow v. People of State of New York, supra; Whitney v. People of State of California, supra”, 283 U.S. at pages 368, 369, 51 S.Ct. at page 535, 75 L.Ed. 1117, 73 A.L.R. 1484. In Herndon v. State of Georgia, 295 U.S. 441, 55 S.Ct. 794, 79 L.Ed. 1530, the majority found it possible to avoid the constitutional question, but Brandéis, Stone and Cardozo, JJ., dissented. Cardozo, J., who wrote the dissent, discussed in detail, Gitlow v. People of State of New York, and very carefully avoided any suggestion that it had not been correctly decided. He thought that “the effect of all this” (that is, of the later opinions) “was to leave the question open whether in cases * * * where the unlawful quality of words is to be determined not upon their face but in relation to their consequences, the opinion in Schenck v. United States supplies the operative rule”. 295 U.S. at page 451, 55 S.Ct. at page 798, 79 L.Ed. 1530. We read this as meaning that the rule of “clear and present danger” might be properly limited to situations in which the words were not themselves those of direct instigation. We do not see that De Jonge v. State of Oregon, 299 U.S. 353, 57 S.Ct. 255, 81 L.Ed. 278, adds anything except that here also Gitlow v. People of State of New York and Whitney v. People of State of California were cited with apparent approval. Herndon v. Lowry, 301 U.S. 242, 57 S.Ct. 732, 81 L.Ed. 1006, reversed a conviction because a state statute was too vague, like that before the court in Stromberg v. People of State of California, supra; Gitlow v. People of State of New York being discussed at length without any intimation of dissidence. All the foregoing cases concerned the validity of statutes which had made it unlawful to stir up opposition to the Government or a state in the discharge of some vital function. There followed several which held that an ordinance or statute might not trench upon freedom of speech in order to promote minor public convenience : e.g., preventing the streets from being .littered by broadsides, Schneider v. State, 308 U.S. 147, 60 S.Ct. 146, 84 L.Ed. 155; requiring a license to solicit contributions for societies, Cantwell v. State of Connecticut, 310 U.S. 296. 60 S.Ct. 900, 84 L.Ed. 1213, 128 A.L.R. 1352; requiring a union leader to register his name and union affiliation with the Secretary of State, Thomas v. Collins, 323 U.S. 516, 65 S.Ct. 315, 89 L.Ed. 430. The opinions in all these cases did however repeat the rubric of Schenck v. United States, supra, though none of them attempted to define how grave, or how imminent the danger must be, or whether the two factors are mutually interdependent. Moreover, the situation in all was wholly different from that in the preceding decisions. It is one thing to say that the public interest in keeping streets clean, or in keeping a register of union leaders, or in requiring solicitors to take out licenses, will not justify interference with freedom of utterance (and the last decision of the Court is to the contrary, Kovacs v. Cooper, 336 U.S. 77, 69 S.Ct. 448, 93 L.Ed. 513, 10 A.L.R.2d 608); but it is quite another matter to say that an organized effort to inculcate the duty of revolution may not be repressed. It does not seem to us therefore that these decisions help towards a solution here. Three decisions involving punishment for contempt of court stand upon a different footing: Bridges v. State of California, 314 U.S. 252, 62 S.Ct. 190, 86 L.Ed. 192, 159 A.L.R. 1346; Pennekamp v. State of Florida, 328 U.S. 331, 66 S.Ct. 1029, 90 L.Ed. 1295; and Craig v. Harney, 331 U.S. 367, 67 S.Ct. 1249, 91 L.Ed. 1546. In each the question was whether the Fourteenth Amendment, incorporating as it does the First, protected against criminal contempt one who published abuse of a judge while a case was pending before him. These opinions also repeated the rubric, but throw no light upon its meaning, and in any case, there could be no issue as to the imminency of the danger, for, whatever effect the abuse might have upon the judge’s decision, it would be either immediate, or at most delayed no longer than his period of deliberation. Moreover, in Pennekamp v. State of Florida, the majority said that courts must “appraise the comment on a balance between the desirability of free discussion and the necessity for fair adjudication, free from interruption”. 328 U.S. at page 336, 66 S.Ct. at page 1032, 90 L.Ed. 1295. One must “weigh the right of free speech * * * against the danger ,of the coercion and intimidation of courts”, 328 U.S. at page 346, 66 S.Ct. at page 1037, 90 L.Ed. 1295. And again, “What is meant by clear and present danger to a fair administration of justice? No definition .could give an answer”. 328 U.S. at page 348, 66 S.Ct. at page 1038, 90 L.Ed. 1295. Schneiderman v. United States, 320 U.S. 118, 63 S.Ct. 1333, 87 L.Ed. 1796, also stands apart; it held that the prosecution had not adequately proved that Schneider-man, though a Communist, was not “attached to the principles of the Constitution,” when he was naturalized. The majority thought that being a Communist might involve no more than- what the defendants at bar say that it does involve: to foster revolutionary changes, but only by lawful methods. All that can be thought relevant to the case at bar is a passage in the opinion, which may have been meant to imply that only “agitation and exhortation calling for present violent action which creates a clear and present danger of public disorder or other substantive evil”, 320 U. S. at page 157, 63 S.Ct. at page 1352, 87 L.Ed. 1796, will show that one is not attached to the “principles of the Constitution.” Of the eight justices who took part in the decision, three dissented, and of the five who concurred two wrote separate opinions. It is true that both these said that they joined in the opinion in chief; but we should hesitate to say that by this they meant to commit themselves to the proposition that a man may be attached to the principles of a constitution, whose violent overthrow he will continue to advocate and teach, because he knows that the execution of his purpose must be deferred for a time. We should feel bound to await a more definite declaration before accepting a doctrine, which, with deference, seems to us so open to doubt. In Hartzel v. United States, 322 U.S. 680, 64 S.Ct. 1233, 88 L.Ed. 1534, a prosecution under the Espionage Act, the question was whether the accused had “wilfully” tried to cause insubordination, and the majority held that “wilfully” confined the forbidden utterance to what was specifically intended to bring about the evil, and then only in case there was a “clear and present danger” of its success. 322 U.S. at pages 686, 687, 64 S.Ct. at page 1236, 88 L.Ed. 1534. Four justices thought that no specific intent had been proved and did not find it necessary to pass upon whether there was the prescribed danger; four thought the specific intent proved and said nothing as to the proximity of the danger; Roberts, J., merely said that the evidence was insufficient. Chaplinsky v. State of New Hampshire, 315 U.S. 568, 62 S.Ct. 766, 86 L.Ed. 1031, and Terminiello v. Chicago, 337 U.S. 1, 69 S.Ct. 894, 895, 93 L.Ed. 1131, concerned statutes or ordinances which made unlawful speeches that were likely to create disorders or other “breaches of the peace.” Concededly the accused had not intended the “substantive evil” to occur, though he was probably indifferent about it, and may have thought disorder likely to result. In the first case the conviction was affirmed; in the second it was. reversed. Whatever danger there was, was sure to be immediate, and the cases are not helpful here. Nor is the law as to enjoining peaceful picketing altogether plain. In Thornhill v. State of Alabama, 310 U.S. 88, 66 S.Ct. 736, 84 L.Ed. 1093, the Court declared unconstitutional a state statute which forbad such picketing in an ordinary labor dispute. It held that picketing, when unaccompanied by threats or violence, was no more than an appeal to. others to side with the union. However, in Giboney v. Empire Storage & Ice Co., 336 U.S. 490, 69 S.Ct. 684, 93 L.Ed. 834, it upheld a state statute which authorized courts to enjoin such picketing in what seems to us to have amounted to a secondary boycott. An ice peddlers’ union wished to compel some unwilling peddlers to join, and in order to do so picketed an ice plant which refused to stop supplying ice to the recalcitrants. That conduct the state statute forbad, and the Court held that it was constitutional, as part of the remedy, to enjoin picketing, although that consisted only in trying to persuade others to join in the unlawful effort. The “evil” — injury to the defendant’s business — was “present” enough, and apparently it was important enough to justify even a prevention of the utterance in advance. In International Brotherhood of Teamsters, etc., Union Local 309, v. Hanke, 339 U.S. 470, 70 S.Ct. 773, a partnership of four was conducting a business which included the sale of secondhand motor cars. They had no employees, but their predecessor had conducted a union shop, and the senior partner had joined the union when he bought the business. Later the partners resigned from the union because they were unwilling to abide by the union rules as to the hours of business. Although there was no statute which made the union’s conduct unlawful, a majority of the Court held that the question was of balancing conflicting interests, and that the state was free to decide which interest should prevail, whether by its legislature or by its courts. The last decision — in some ways the most important of all — is American Communications Association, C. I. O. v. Douds, 339 U.S. 382, 70 S.Ct. 674, in which a majority of the Court held that § 9(h) of the Labor-Management Act was constitutional in requiring the officers of a union to take an oath that they were not Communists, as a condition to according the union the benefits of the Act. All the opinions seem to us to have accepted it as a condition upon any limitation upon freedom of utterance that there must be some “clear and present danger” that the utterance will succeed in creating a “substantive evil” within the control of Congress. The opinion in chief, however, cautioned that this was not to be applied “as a mechanical test * * * without regard to the context of its application.” “It is the considerations that gave birth to the phrase * * * not the phrase itself, that are vital”. 339 U.S. at page 394, 70 S.Ct. at page 681. Again, “even harmful conduct cannot justify restrictions upon speech unless substantial interests of society are at stake. But * * * it was never the intention of this Court to lay down an absolute test measured in terms of danger to the Nation”. 339 U.S. at page 397, 70 S.Ct. at page 683. “When particular conduct is regulated in the interest of public order, and the regulation results in an indirect, conditional, partial abridgment of speech, the duty of the courts is to determine which of these two conflicting interests demands the greater protection under the particular circumstances presented”. 339 U.S. at page 399, 70 S.Ct. at page 684. “We must therefore, undertake the ‘delicate and difficult task * * * to weigh the circumstances and to appraise the substantiality of the reasons advanced in support of the regulation of the free enjoyment of the rights.’ Schneider v. State, 1939, 308 U.S. 147, 161, 60 S.Ct. 146, 84 L.Ed. 155”. 339 U.S. at page 400, 70 S.Ct. at page 684. The significance of the decision for the purposes of the case at bar does not, however, lie in the language used, so much as in the decision itself; not indeed because the conflict of interest which the Court there resolved, was the same as that before us, but because the test applied was the same as in the case of a criminal prosecution for utterances threatening the stability of the government: i.e., “clear and present danger”; and because this was used to weigh the danger to commerce considering its gravity and proximity, against the repression of political activity involved — indeed a minority included even repression of political belief. The danger in that case included “political” strikes, and the danger to commerce from such strikes is closer than the danger to the existence of the government is to the teachings of the defendants; but the second danger is vastly graver if it be realized. We do not pretend that the decision is authoritative here; indeed the Court carefully declared that it was limited to the situation then at bar. j What we do say is that no longer can there be any doubt, if indeed there was before, that the phrase, “clear and present danger,” is not a slogan or a shibboleth to be applied as though it carried its own meaning; but that it involves in every case a comparison between interests which are to be appraised qualitatively. From this wearisome analysis of the decisions of the Supreme Court it has appeared, as we indicated at the outset, that to deprive an utterance of the protection of the Amendment it is not always enough that the purpose of the utterer may include stirring up his hearers to illegal conduct— at least, when the utterance is political. The same utterance may be unprotected, if it be a bare appeal to action, which the Amendment will cover, if it be accompanied by, or incorporated into, utterances addressed to the understanding and seeking to persuade. The phrase, “clear and present danger,” has come to be used as a shorthand statement of those among such mixed or compounded utterances which the Amendment does not protect. Yet it is not a vade mecum; indeed, from its very words it could not be. It is a way to describe a penumbra of occasions, even the outskirts of which are indefinable, but within which, as is so often the case, the courts must their way as they can. In each case they must ask whether the gravity of the “evil,” disc.ounted, by its improbability, justifies such invasion of free speech as is necessary to avoid the danger. We have ^"purposely substituted “improbability” for “remoteness,” because that must be the right interpretation. Given the same probability, it would be wholly irrational to condone future evils which .we should prevent if they were immediate; that could be reconciled only by an indifference to those who come after us. It is only because a substantial intervening period between the utterance and its realization may check its effect and change its importance, that its immediacy is important; and that, as we have said, was the rationale of the concurrence in Whitney v. People of State of California, supra. We can never forecast with certainty; all prophecy is a guess, but the reliability of a guess decreases with the length of the future which it seeks to penetrate. In application of such a standard courts may strike a wrong balance; they may tolerate “incitements” which they should forbid; they may repress utterances they should allow; but that is a responsibility that they cannot avoid. Abdication is as much a failure of duty, as indifference is a failure to protect primal rights. \ In the case at bar the defence seems to us to kick the beam. One may reasonably think it wiser in the long run to let an unhappy, bitter outcast vent his venom before any crowds he can muster and in any terms that he wishes, be they as ferocious as he will; one may trust that his patent impotence will be a foil to anything he may propose. Indeed, it is a measure of the confidence of a society in its own stability that it suffers such fustian to go unchecked. Here we are faced with something very different. The American Communist Party, of which the defendants are the controlling spirits, is a highly articulated, well contrived, far spread organization, numbering thousands of adherents, rigidly and ruthlessly disciplined, many of whom are infused with a passionate Utopian faith that is to redeem mankind. It has its Founder, its apostles, its sacred texts — perhaps even its martyrs. It seeks converts far and wide by an extensive system of schooling, demanding of all an inflexible doctrinal orthodoxy. The violent capture of all existing governments is one article of the creed of that faith, which abjures the possibility of success by lawful means. That article, which is a common-place among initiates, is a part of the homiletics for novitiates, although, so far as conveniently it can be, it is covered by an innocent terminology, designed to prevent its disclosure. Our democracy, like any other, must meet that faith and that creed on the merits, or it will perish; and we must not flinch at the challenge. Nevertheless, we may insist that the rules of the game be observed, and the rules confine the conflict to weapons drawn from the universe of discourse. The advocacy of violence may, or.,may not, fail; hut in neither case can there be any “right” to use it. Revolutions are often “right,” but a “right of revolution” is a contradiction in terms, for a society which acknowledged it, could not stop at tolerating conspiracies to overthrow it, but must include their execution. The question before us, and the only one, is how long a government, having discovered such a conspiracy, must wait. When does the conspiracy become a “present danger”? The jury has found that the conspirators will strike as soon as success seems possible, and obviously, no one in his senses would strike sooner. Meanwhile they claim the constitutional privilege of going on indoctrinating their pupils, preparing increasing numbers to pledge themselves to the crusade, and awaiting the moment when we may be so far extended by foreign engagements, so far divided in counsel, or so far in industrial or financial straits, that the chance seems worth trying. That position presupposes that the Amendment assures them freedom for all preparatory steps and in the end the choice, of initiative, dependent upon that moment when they believe us, who must await the blow, to be worst prepared to receive it. We need not say that even so thoroughly planned and so extensive a confederation would be a “present danger” at all times and in all circumstances;' the question is how imminent: that is, how probable of execution — it was in the summer of 1948, when the indictment was found. We must not close our eyes to our position in the world at that time. By far the most powerful of all the European nations had been a convert to Communism for over' thirty years; its leaders were the most devoted and potent proponents of the faith; no such movement in Europe of East to West had arisen since Islam. Moreover in most of West Europe there were important political Communist factions, always agitating to increase their power; and the defendants were acting in close concert with the movement. The status quo, hastily contrived in 1945, was showing strains and tensions, not originally expected. Save for the unexpected success of the airlift, Britain, France and ourselves would have been forced out of Berlin, contrary to our understanding of the convention by which we were there. We had become the object of invective upon invective; we were continuously -charged with aggressive designs, against other nations; our efforts to reestablish their economic stability were repeatedly set down as a scheme to enslave them; we had been singled out as the chief enemy of the faith; we were the eventually doomed, but the still formidable, protagonist of that decadent system which it was to supplant. Any border fray, any diplomatic incident, any difference in construction of the modus vivendi-^such as the Berlin blockade we have just mentioned — might prove a spark in the tinder-box, and lead to war. We do not understand how one could ask for a more probable danger, unless we must wait till the actual eve of hostilities. The only justification which can be suggested is that in spite of their efforts to mask their purposes, so far as they can do so consistently with the spread of the gospel, discussion and publicity may so weaken their power that it will have ceased to be dangerous when the moment may come. That may be a proper enough antidote in ordinary times and for less redoubtable combinations; but certainly it does not apply to this one. Corruptio optimi pessima. True, we must not forget our own faith; we must be sensitive to the dangers that lurk in any choice; but choose we must, and we shall be silly dupes if we forget that again and again in the past thirty years, just such preparations in other countries have aided to supplant existing governments, when the time was ripe. Nothing short of a revived doctrine of laissez faire, which would have amazed even the Manchester School at its apogee, can fail to realize that such a conspiracy creates a danger of the utmost gravity and of enough probability to justify its suppression. We hold that it is a danger “clear and present.” However, although for the foregoing reasons we agree with the Eighth Circuit' that the Smith Act is constitutional, so limited, it does not follow that as matter of interpretation it may be so limited, or that the judge was justified in charging the jury that they need only find the defendants meant to use violence “as speedily as circumstances would permit it” (the overthrow or destruction of the government) “to be achieved.” The words of the Act are unconditional and forbid advocacy or teaching of such a violent overthrow at any time and by anyone, weak or strong;' literally, they make criminal the fulminations of a half crazy zealot on a soap box, calling for an immediate march upon Washington. Therefore, the defendants argue that, if the Act is to be limited as it must be to be valid, Congress must limit it; and further, that, if it were so limited, it would become too vague to serve as a. definition of crime. We think not. Section 40 of the original act provided by a not uncommon clause, not only that “if any provision” were “held invalid, the remainder of the Act,” should not be affected; but also that, if “the application thereof to any person or circumstance is held invalid * * * the application * * * to other persons or circumstances” should not be affected. Even when there is no “separability” clause of any kind, the doctrine of United States v. Reese, 92 U.S. 214, 23 L.Ed. 563, does not always apply; the Supreme Court has often limited general words in a statute so as to make it constitutional, although in such cases a court must hazard the inference that Congress would have enacted the statute in the limited form,. if it had known that in its broad scope it would be unconstitutional. We have no such problem here, because there can be no doubt as to the intent; Congress has explicitly declared that it wished the words to govern all cases which they constitutionally could. Nor do we think that, so limited, the Act becomes too vague to stand up, which is the only challenge on the score of vagueness that deserves discussion. In the first place, it is to be observéd that it would have been impracticable to provide against the evil and yet to define the forbidden conduct more definitely. True, one might have added in § 2(a) (1) [1948 Revised Criminal Code 18 U.S.C.A., § 2385] after the word “teach” the clause: “when that constitutes a clear and present danger”; and in § 2(a) (3) the same clause after the words “help to organize.” But that would not have helped to define the forbidden conduct; for, not only are those words imprecise in themselves, and were never intended to be otherwise; but, as we have seen, they presuppose balancing the repression necessary to avoid the evil, against the evil itself, discounted by the improbability of its occurrence. That is a test in whose application the utmost differences of opinion have constantly arisen, even in the Supreme Court. Obviously it would be impossible to draft a statute which should attempt to prescribe a rule for each occasion; and it follows, as we have said, either that the Act is definite enough as it stands, or that it is practically impossible to deal with such conduct in general terms. Such a consideration is relevant in judging the ■constitutionality of any statute. United States v. Petrillo, 332 U.S. 1, 7, 67 S.Ct. 1538, 91 L.Ed. 1877. There is ah added reason leading to the same result. Both § 2(a) (1) and § 2(a) (3) make advocacy and teaching of revolution criminal only when it is with a specific intent — the intent to do so by “force or violence.” Obviously one cannot teach or advocate,the use of violence without specifically intending to bring about its use; a'fortiori must that be true, if one organizes a group so to teach. All discussion as to the use of the word “wilfully” in the alternative in § 2(a) (1) is. therefore irrelevant; the sections carry their own specific intent, and that intent violates the accepted mores of our society, which discountenance resort to violence as a means of political change. When a statute is directed against conduct which offends accepted moral standards, and particularly when the moral offense is heinous, ambiguities do not count against its validity as much as they do when the proscribed conduct has no ethical significance, as, for instance, when it is only an economic regulation, United States v. L. Cohen Grocery Co., 255 U.S. 81, 41 S.Ct. 298, 65 L.Ed. 516, 14 A.L.R. 1045. Thus, in Gorin v. United States, 312 U.S. 19, 61 S.Ct. 429, 85 L.Ed. 488, a conviction was affirmed under a statute which punished making copies of “anything connected with the national defense,” or “relating to” it, because the statute was limited to cases where the forbidden act was done “for the purpose of obtaining information * * * with intent or reason to believe that the information to be obtained is to be used to the injury of the United States.” Those words, the court said, required “those prosecuted to have acted in bad faith,” which was “sufficiently definite to apprise the public of prohibited activities” 312 U.S. at page 28, 61 S.Ct. at page 434, 85 L.Ed. 488. Again, in United States v. Ragen, 314 U.S. 513, 62 S.Ct. 374, 86 L.Ed. 383, although the crime depended upon whether the accused paid more than “reasonable” commissions, the statute was held valid because it was limited to deductions, made “willfully.” “That such acts of bad faith are not beyond the ready comprehension either of persons affected by the act or of juries called upon to determine violations need not be elaborated.” “A mind intent upon willful evasion is inconsistent with surprised innocence”. 314 U.S. at page 524, 62 S.Ct. at page 378, 86 L.Ed. 383. This is a rational exception to the doctrine that the words of a statute must not be too vague to guide those who wish to conform to it; it means that, though the forbidden conduct be itself defined in ambiguous terms, the ambiguity will be cleared up, if the statute adds as a condition that the conduct is criminal only in case the accused knows that what he intends is wrong. To apply the doctrine to such persons would only serve to relieve them of the risk of incorrectly guessing the right meaning of the words, when engaged in an undertaking that they know to be wrong. In the case at bar the doctrine would mean that those, who set in train of execution an elaborate scheme of violent revolution, will escape because the literal meaning of the words covers others, too insignificant to be noticed. The defendants have no standing as vicarious champions of such offenders. We hold the Smith Act to be constitutional, as the judge construed it; and that evidence supported the verdict of the jury.? It will be most convenient now to consider also the defendants’ exception to that part of the judge’s charge which took from the jury all questions regarding the constitutionality of the Act, and only left to them whether the defendants’ intent was to overthrow the government “as speedily as circumstances would permit it to be achieved.” As we have said, “clear and present danger” depends upon whether the mischief of the repression is greater than the gravity of the evil, discounted by its improbability; and it is of course true that the degree of probability that the utterance will bring about the evil is a question of fact. On the other hand, to compare the repression with the evil, when discounted, is not a question of fact at all; for it depends upon a choice between conflicting interests. Ordinarily such choices are for a legislature, whose chief function it is indeed to make them, since a legislature is best qualified to represent the divergent interests of society. However, as we have just said, it is at times impracticable to make such choices in general propositions, because the occasions which will arise within the ambit of the general purpose, are multiform. Sometimes, these choices are delegated initially to administrative tribunals, subject to a court review; sometimes — notably in the case of the Anti-Trust Acts— they are left to the courts, first and last. It is true that similar choices — choices of the proper standard to be applied ad hoc— are sometimes also left to juries: negligence and reasonable notice are instances. But these usually concern only individuals involved in a private suit, not the conflict of momentous public interests; and it would be improper to take them as exemplifying a universal principle. When the dispute does involve interests' of high moment, and Congress, thinking it impracticable to deal with them specifically, makes the courts its surrogate, the choices so delegated must be treated as questions of law. Were it not so, there would be no chance for review, for the verdict would be final; moreover, different juries might give different verdicts, and any approach to uniformity, short as that can be in any event in this field, would be impossible. We do not forget that the majority of the Court in Pierce v. United States, 252 U.S. 239, 250, 40 S.Ct. 205, 209, 64 L.Ed. 542, said that “the probable effect of distributing it” (a challenged pamphlet) “in the mode adopted” was among those issues which should be left to a jury. The question to which that remark was addressed, was however only whether the evidence supported the verdict; and, so far as appears, the accused had- not asked that the court should pass upon the pamphlet. For these reasons the judge appears to us to have been right, when in the case at bar he took upon himself the duty of declaring that the defendants were guilty, if the jury found that they organized and supported the Party ‘for the purpose, among others, of spreading the doctrine of violent revolution, that purpose to be realized as soon as it was feasible. II. The Challenge to the Array The next question concerns the challenge to the “array” which we mentioned at the outset, and upon which the defendants conccdedly had the burden of proof. Their, position is — to quote from their brief — “that none of the jury lists, panels and arrays * * * is, or for a long period of time has been, representative of a cross-section of the community, but has been representative only of the upper economic and social groups.” They do not assert that jury lists must “give representation, proportional or otherwise, to every economic” group, but they do assert “that there was, on the one hand, systematic and gross discrimination against and limitation of the excluded groups * * • * and on the other hand, systematic and gross favoritism towards and intentional inclusion of the rich, the propertied and the well-to-do strata of the community.” We assume for argument that if the jury list — that is, the list . from which separate panels are drawn — was unlawful, any separate panel based upon that list was unlawful, though taken by itself it was unexceptionable. That is to say, we will assume that any party to a suit, civil or criminal, is entitled to have the particular panel which tries his case, drawn at random from a list which is not unlawfully weighted, and that he may complain even though he has not shown that the imbalance ' has prejudiced him. The question therefore becomes whether the grand panel and the petit panel which respectively found the indictment and tried the case, were drawn from an unlawful list. The list contained those names which had been carried over from the past to which were added in 1947 and 1948 hew names, culled out of some 25,000 notices sent out during those years. The practice of the clerk’s office was for a deputy clerk, to whom the duty was assigned, either himself, or through assistants who-acted under his direction, to prepare the list, confined substantially to the Boroughs of Manhattan and the Bronx with a sprinkling from Westchester and the other northern counties. The challenge is, not only to the sources from which he sent out no-, tices to prospective jurors, but also that he unlawfully selected those whom he preferred from among those who appeared in answer to the notices and were in fact qualified. There is not enough evidence of this last challenge to demand discussion, and we shall confine ourselves to the first. It does not appear what proportion of those notified appeared, but it must have been small, for, after those who did appear, but failed to qualify were excluded, there remained only fifteen to twenty per cent— roughly one-sixth. In order to “qualify” a, juror, the deputy required him to swear to a “questionnaire” stating his name, address, place and date of birth, education,, employment and employer and other details. When “qualified” the name was put on the list, and a card was made out, and added to those from which the clerk and the commissioner drew panels as they were needed. The Federal Grand Jury Association, as its name implies, was a voluntary association of present or former grand jurors in the Southern District. Beginning with the year 1938 and until some time in 1941 it prepared lists of prospective jurors, compiled by its officers from the following sources: Who’s Who in New York, Who’s Who in Engineering, The Social Register, volunteers recommended by the Association, the alumni directories of Princeton, Columbia, Harvard, Yale and Dartmouth, Poor’s Register of Executives, and the Directory of Directors. The lists so prepared it .submitted to the clerk and deputy clerk, who until May, 1942, used them as the principal source of persons to be notified. The total number of names so submitted by ■the Association through May 1942, was 16,277. The record contains no figures as to what was the condition of the list in 1938, except the testimony of the Chief Judge of the District that it contained an inordinate number of people on relief. In January, 1942, the Supreme Court decided the case of Glasser v. United States, 315 U.S. 60, 62 S.Ct. 457, 86 L.Ed. 680, in which Murphy, J., said: “deliberate selection of jurors from the membership of particular private organizations definitely does not conform” to the requirements of “a cross-section of the community”. 315 U.S. at page 86, 62 S.Ct. at page 472, 86 L.Ed. 680. Whether or not because of this, after May, 1942, the Grand Jury Association prepared and sent no lists except 2,014 up to April, 1943, of which 514 were those of women volunteers, Negroes and other undistributed persons. After April, 1943, it sent only 121 names through the next five years. Therefore, even though we count all these as improperly used, the jurors who qualified out of them were presumably not more than 350, too little to deserve notice. A new deputy clerk was appointed in 1940 and from him we know that in July, 1941, the list contained about TO,-000 names, of whom about 8,000 were petit jurors, and 2,000 were grand jurors. It was about the same size in 1942; but between 1940 and July, 1942, the Association had sent about 12,500 names to the clerk, out of which there were presumably entered upon the list say 2,000 new names, though the judge found that 1,900 names in all had been added between July, 1941, and July, 1942. From July, 1942, to April, 1943, the clerk notified nearly 12,000 persons taken from the voting lists, and 5,000 taken from the Association and volunteers. From the voting lists presumably he secured about 2,000 jurors, and from the Association 800 or more. Again in 1945 and 1946 he added about 2,500 more; so that at the end of that year, except for wastage, the list should have contained between sixteen and seventeen thousand names. In fact the number was about 13,000, and at the end of that year the clerk conducted a sort of purge, called an “inventory,” which was designed to strike out those who had ceased to be qualified for one reason or another. In 1947 and 1948, as we have said, he notified 25,000 new persons — 22,000 from the voting lists — out of which again we are to assume that there resulted a little over 4,000 new qualified jurors. The list in 1948 was about 13,000 (apparently the additions once more only made up for the losses caused by the “inventory”). It is of course impossible to say how many of the names originally sent to the clerk by the Jury Association remained upon that list. As we have seen, even though we do not allow for wastage they were presumably no more than a sixth of 18,000, or about 3,000 out of 13,000. The wastage over eight years must have been substantial, although we can do no more than guess what it was. Since the defendants have the burden, we shall do no injustice if we assume that no more than 3,000 of the 13,000, as the list stood in 1948, had been suggested by the Association. The clerk was free to call all these persons if he wished; he had no reason to reject them; at most his only mistake was that he allowed others to suggest them to him; and, as we shall show, that was not unlawful, for he was free to go to any source for persons to call. The only question is whether they so weighted the list as a whole as to make it an improper source for the panels drawn in the case at bar. Glasser v. United States, supra, did not condemn the practice of accepting suggestions from groups such as the Association ; what it did condemn was calling from any-one group — women in that case — members of a single association only. We do not forget that of the grand jurors in the 70 panels examined by the defendants, 38 per cent of all who served, had been suggested by the Association. The most that could be inferred is that there may 'have been a disposition to favor Association members as grand jurors while they continued upon the list. In conclusion we hold that, even though we were to concede that a list would have been invalid which was wholly made up by persons recommended by the Association, it was not proved that enough of those so recommended.remained upon the list in 1948 to invalidate it. as a whole. The defendants next challenge the list because of the use made by the deputy clerks of the voting lists after the middle of 1942. With great detail and labor they prepared a series of charts designed to prove that the clerks must have deliberately selected from the voting lists those who came from the wealthier districts of the City in preference to those who came from the poorer. The clerks themselves categorically denied this, and the judge believed them and made a number of findings that they had not had any such purpose; but the defendants answer that the distribution which actually resulted so contradicts the testimony that the findings are “clearly erroneous.” This argument is drawn not only from the 70 grand jury panels, which we have just mentioned, but from 28 petit jury panels, taken more or less at random. The defendants say that the large disproportion of jurors on these panels coming from the wealthier districts of the City, presupposes a similar proportion among those called, and that this could not have been the result of accident, and proves that the clerks preterred the wealthier districts because they wished to pad the list with wealthier jurors. We may say at the outset that it is to the last degree improbable that the clerks themselves should have had any preference for wealthy jurors; the only rational assumption is that the judges had directed them so to weight the list, which the Chief Judge of the district,, who had charge of the matter, flatly denied. The defendants make an additional argument based upon classification of the jurors-on the same panels by their occupations ^ but all such classifications are inevitably arbitrary and these incidentally do not conform to the Census. This kind of reasoning the Supreme Court refused to accept in Fay v. People of State of New York, 332 U.S. 261, 67 S.Ct. 1613, 91 L.Ed. 2043; and in any event, whatever can be inferred-from it would add nothing to the argument drawn from the distribution by districts. We shall therefore confine our consideration to the territorial charts. These were themselves shown to be far from immune from attack in their origin and in detail. They were prepared by two witnesses, Wilkerson and Rodman, who supervised the work of a number of assistants, none of whom were called as witnesses; and the charts were therefore not competent as original documents. However, since they were used only as argumentative compilations from the panels which showed the jurors’ addresses and had been put in evidence, it was proper to use them. Concededly all contained errors, and the judge discredited the accuracy, if not the good faith, of both Wilkerson and Rodman. Although, as we have said, the 28 panels of petit jurors were chosen at random, the sample territorial charts were drawn from only thirteen of .these. The first six were indeed taken alternately; but the record does not show how the other seven were chosen; for all that appears they may have been deliberately selected because of their especial fitness to prove the defendants’ contention. Moreover, quite aside from these infirmities, there are good reasons for be- . lieving that the panels were not a trustworthy basis for learning the composition of the lists of persons to whom notices were sent. A large, though uncertain, percentage of those called never appear at all; as we have said already, only fifteen to twenty per cent of those called qualify, and the record does not show how large a part of this deficit is made up of those who do not appear. It may be a very large part, and it is so costly and troublesome to proceed against the defaulters, that the clerk merely calls others in their place. It would be an unproved assumption that the proportion of defaulters is the same among the poorer groups as among the wealthier; and indeed there is perhaps some reason for thinking that it may be considerably greater; not indeed because the wealthier groups have a higher sense of civic duty, but because they may think themselves more subject to pursuit. Be that as it may, the Census shows that less than one-half the urban population lives in the same quarters for seven years, and again it was not proved that this mobility was not greater among the poor than among the wealthier. Besides, a well-to-do person is more likely to leave a forwarding address than is a poor one, and so to receive a notice directed to his past residence. Again, it was not proved that the clerk did not excuse a larger percentage of the poorer persons notified who did- appear. Jury duty is a burden to all who serve except to the relatively few who are not gainfully employed and have leisure — such as retired business men, housewives whose children have grown up, and those who welcome any break in the monotony of their lives. But the hardship which it imposes upon a manual worker, especially now that the cost of living has so much increased, is much greater than that upon “executives” for example, who, ordinarily without serious disarrangement, can be made to serve at specified months. But all times are equally inconvenient to one who must work by the day, week or month to support a family, or even himself alone. No one who has not been charged with the duty of excusing such persons knows how strong such an appeal can be. This factor alone may have accounted for a wide difference between the proportion of those notified and those who “qualified.” That such excuses are illegal Thiel v. Southern Pacific Company, 328 U.S. 217, 66 S.Ct. 984, 90 L.Ed. 1181, 166 A.L.R. 1412, did not decide. The jury clerk had there systematically refused to call any who worked for a daily wage. True, his excuse was that he had found that the judges uniformly excused all such when they were drawn for a panel; but what vitiated the list was the completeness of the exclusion, and the foolishness of the line between daily, weekly or monthly workers. The majority were careful so to limit the decision; saying 328 U.S. on page 224, 66 S.Ct. 984, 987, 90 L.Ed. 1181, “It is clear that a federal judge would be justified in excusing a daily wage earner for whom jury service would entail an undue financial hardship,” and quoting with approval the remarks of Judge Knox before the House Committee of the Judiciary that “to require them to do so” (to serve) “is nothing less than the imposition upon them of extreme hardship”. 328 U.S. at page 224, 66 S.Ct. at page 987, 90 L.Ed. 1181. “Only when the financial embarrassment is such as to impose a real burden and hardship does a valid excuse of this nature appear. Thus a blanket exclusion of all daily wage earners, however well-intentioned and however justified by prior actions of trial judges, must be counted among those tendencies which undermine and weaken the institution of jury trial”. 328 U.S. at page 224, 66 S.Ct. at page 987, 90 L.Ed. 1181. Finally 328 U.S. at page 223, 66 S.Ct. at page 987, 90 L.Ed. 1181: “the pay period of a particular individual is completely irrelevant to his eligibility and capacity to serve as a juror.” Whether any such power was properly lodged in the deputy clerks is another matter; even though we are to concede that it was not, it would not prove that they had deliberately chosen wealthier jurors and discarded poorer ones. Therefore this too might have accounted for a large part of the disparity between the panels who served and a roster of those called, if we had one. However, after making allowance both for the discovered errors in the charts, and for the unanswered possible disparities we have just discussed, it appears to us that it would not be candid to say that these would account for the territorial distribution of the panels analyzed in the charts. If notices were sent out upon the basis of the voting lists alone without any principle of selection, they would not, we think, have res