dissenting:
I share the majority’s reluctance to affirm this conviction on the basis of the views expressed by Judge Nebeker. See majority opinion at 1246 n. 1. I had *1250not thought that anything in Terry v. Ohio, 392 U.S. 1, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968), changed the long-standing rule that evidence resulting from a search incident to an unlawful arrest is inadmissible. Nor' had I doubted that when a policeman on the beat announces, “You are under arrest,” he means what he says, and his words are sufficient to effect an arrest.1 It follows that, if the arrest is invalid, the incident search is invalid as well, even if it is confined to the sort of superficial frisk which would have been permissible following a Terry stop. See Clarke v. United States, D.C.C.A., 256 A.2d 782, 786 n. 7 (1969). Any less stringent rule would mean that whenever the police had the reasonable suspicion necessary for a Terry stop, they could instead arrest the suspect and subject him to all the indignities which an arrest implies without losing the benefit of the evidence seized so long as that evidence might have been uncovered by a superficial frisk. As this court said in United States v. Cunningham, 138 U.S.App.D.C. 29, 30, 424 F.2d 942, 943, cert. denied, 399 U.S. 914, 90 S.Ct. 2218, 26 L.Ed.2d 572 (1970): “A lack of probable cause cannot be made up in hindsight by a hypothetical variation in the basis on which a search was conducted.” Cf. United States v. Morris, 142 U.S.App.D.C. 196, 197, 440 F.2d 224, 225 (1970).
I do not agree, however, that this conviction may be rehabilitated on the post hoc theory that probable cause existed for a disorderly conduct arrest.2 Nor do I approve the majority’s cavalier treatment of appellant’s argument that the words he spoke were not in that context so offensive as to justify an arrest under the principles explained in Gooding v. Wilson, 405 U.S. 518, 92 S.Ct. 1103, 31 L.Ed.2d 408 (1972), Cohen v. California, 403 U.S. 15, 91 S.Ct. 1780, 29 L.Ed.2d 284 (1971), and Williams v. District of Columbia, 136 U.S.App.D.C. 56, 419 F.2d 638 (1969) (en banc.) Although the majority concedes that “conventions and fashions in language change,” it nonetheless proceeds to hold that appellant’s words still retain “shock quality.” In so doing, my brothers resolve a question of fact about which this court has no expertise and on which no record was made below.
In my view, it is not the function of judges to decide on the basis of their own sensibilities which words retain “shock quality” and which “are today’s common idiom.” No case that I know of authorizes use of the judicial process to figuratively wash out the mouths of criminal defendants who use language which some judge, on his own, considers “dirty” or “offensive.”3 Writing only a few months ago in Huffman v. United States, 152 U.S.App.D.C. 238, 248, 470' F.2d 386, 396 (decided October 7, 1971), Judge Leventhal warned: “Judges concerned with the many elements comprised in our free, democratic society must take care lest they decide these cases on the basis simply of their indignation and disgust * * Because I for one take that warning seriously, I think this case should be remanded for a factual determination as to whether the arresting officer had probable cause to believe that *1251appellant’s words were so offensive where spoken as to threaten a breach of the peace.
After Williams v. District of Columbia, supra, it should be clear that a conviction under 22 D.C.Code § 1107 (1967) is permissible only in very special circumstances. It is not enough that “obscene” or “profane” language be spoken in a public place. In addition “the language [must] be spoken in circumstances which threaten a breach of the peace. And for these purposes a breach of the peace is threatened either because the language creates a substantial risk of provoking violence, or because it is, under ‘contemporary community standards,’ so grossly offensive to members of the public who actually overhear it as to amount to a nuisance.” 136 U.S.App.D.C. at 64, 419 F.2d at 646. (Footnotes omitted.) Any broader reading of the statute would interfere with the “unwritten amenities [which] have been in part responsible for giving our people the feeling of independence and self-confidence, the feeling of creativity” and which “have encouraged lives of high spirits rather than hushed, suffocating silence.” Papachristou v. City of Jacksonville, 405 U.S. 156, 164, 92 S.Ct. 839, 844, 31 L.Ed.2d 110 (1972).
I do not understand the majority to argue that Von Sleichter’s arrest was permissible under the first branch of the Williams test — that is, because probable cause existed to believe that Von Sleichter’s words created a risk of imminent violence. Nor would such an argument be tenable. Von Sleichter directed his epithet at the arresting officer, and it is clear that the risk of provoking a policeman to violence cannot justify an arrest since the police, as representatives of the law, are held to a higher standard than ordinary citizens. See Landry v. Daley, N.D.Ill., 288 F. Supp. 183,187, 192-193 (1968). Cf. Williams v. District of Columbia, supra, 136 U.S.App.D.C. at 64 n. 23, 419 F.2d at 646 n. 23. True, there is some evidence in the record that some other passers-by heard Von Sleichter’s epithet, although even this fact is not unambiguously established.4 But there is no evidence at all that these bystanders were about to resort to violence in order to retaliate for a verbal assault not even directed at them. Hence the situation is essentially the same as that described in Cohen v. California, supra:
“ * * * While the four-letter word * * * is not uncommonly employed in a personally provocative fashion, in this instance it was clearly not ‘directed to the person of the hearer.’ * * Nor do we have here an instance of the exercise of the State’s police pow- ' er to prevent a speaker from intentionally provoking a given group to hostile reaction. * * * There is * * no showing that anyone * * * was in fact violently aroused or that appellant intended such a result.”
403 U.S. 15 at 20, 91 S.Ct. 1780 at 1786. If appellant’s arrest is to be justified at all, then, it must be supported on the basis of the second branch of the Williams test — that is, because his words are, “under ‘contemporary community standards,’ so grossly offensive to members of the public who actually overhear [them] as to amount to a nuisance.” Williams v. District of Columbia, supra, 136 U.S.App.D.C. at 64, 419 F.2d at 646. It seems to me, however, that grave constitutional questions can be raised as to the continuing validity of the second branch of Williams after the Supreme Court’s decisions in Cohen and Gooding. In Cohen the Supreme Court held that “the mere presumed presence of unwit*1252ting listeners or viewers does not serve automatically to justify curtailing all speech capable of giving offense.” 403 U.S. at 21, 91 S.Ct. at 1786. True, Cohen involved an expletive which was written rather than shouted and which was, in some sense, political. But the same cannot be said for the speech in Gooding, which was shouted at a policeman in a manner strikingly similar to that employed in this case. See 405 U.S. at 519-520 n. 1, 92 S.Ct. at 1103. The Supreme Court nonetheless found the statute under which the defendant was prosecuted facially invalid because its scope was not limited to “fighting words” likely to provoke immediate violence. Cf. Chaplinsky v. New Hampshire, 315 U.S. 568, 62 S.Ct. 766, 86 L.Ed. 1031 (1942). In so doing the Court explicitly rejected the state’s argument that it had an interest in preventing speech which was merely “offensive” even if it was unlikely to lead to immediate violence. A statute which “makes it a ‘breach of peace’ merely to speak words offensive to some who hear them * * * sweeps too broadly.” 405 U.S. at 527, 92 S.Ct. at 1108. I fail to understand how this holding can be reconciled with the second branch of Williams, which makes punishable words “offensive to members of the public who actually overhear [them].”
Moreover, even if the second branch of the Williams test is still good law, there remains an insurmountable barrier to its use here. Williams requires a measurement of the language used against “contemporary community standards.” But in order to perform such a measurement, the content of those standards must first be determined. Unfortunately, the record is altogether barren on this subject. There is no indication whatever that anyone who “actually overhear [d]" Von Sleichter’s verbal assault was thereby offended, or that his words in any way departed from the standards of the community in which the incident occurred. The majority seeks to fill the lacuna by taking judicial notice of the fact that Von Sleichter’s words “have at least some minimal shock quality.” Majority opinion at 1249. With all respect, it seems to me this formulation represents a considerable watering down of the Williams test. It is one thing to find that this expletive has “at least some minimal shock quality,” and quite another to find that it is, “under ‘contemporary community standards,’ so grossly offensive * * as to amount to a nuisance.”
When the correct test is applied, I think one can seriously question on a factual level whether these words still fit within it. Thus, although upper New England is hardly known for its lax community standards, Chief Judge Aldrich, writing for the First Circuit Court of Appeals about the same language used in this case, has stated, “[W]e cannot think that it is unknown to many students in the last year of high school,” and noted that the words are used “by young radicals and protesters from coast to coast.” Judge Aldrich went on to observe, “If * * * students must be protected from such exposure, we would fear for their future.” Keefe v. Geanakos, 1 Cir., 418 F.2d 359, 361 (1969).
But it is not my intention to engage in arid debate about whether certain language is or is not consonant with modern community standards since there is, I think, a more important point to be made. In my view, this court commits a serious impropriety when it draws a conclusion about community standards without the benefit of any evidence in the record and without permitting argument on the point before the trial judge.5
I have never been one who believed that judges should check in their common *1253sense when they assume the bench or that studied myopia is a good substitute for judgment and wisdom. But neither do I think the case for judicial activism need rest on the obviously false premise that judges are omnipotent and omniscient. I share Professor Jaffe’s view that “constitutional courts of this country are the acknowledged architects and guarantors of the integrity of the legal system.” L. Jaffe, Judicial Control of Administrative Action 589 (1965). But that is not to say that judges can be everywhere and see everything. In fact, most judges by necessity lead somewhat insulated lives, and this very detachment which gives them the perspective to decide broad issues of policy makes them poor judges of community standards on the street.
Moreover, even if one assumes — erron-. eously, in my view — that judges are qualified to determine community standards in general, it still does not follow that they have the specialized knowledge which Williams requires. Under Williams, Von Sleichter can be punished only if his language was “so grossly offensive to members of the public who actually overhear it as to amount to a nuisance.” 136 U.S.App.D.C. at 64, 419 F.2d at 646. (Emphasis added.) Hence the relevant community under the Williams test is not the nation as a whole or even the city of Washington. It is, rather, the immediate neighborhood in which the words were spoken. Surely we cannot assume that judges can speak authoritatively as to the community standards of a portion of the city about which they may know nothing and which they may (never have even entered. When faced with the impossible task of articulating such standards on their own, it is almost inevitable that judges will slip over into the sort of individualistic, subjective judgments which the Williams test was designed to avoid.
It is fears such as these which have led some of our most respected judges to eschew the task of giving content on their own to community standards — a task which the majority casually assumes today. Thus Judge Thornberry, writing for a panel of the Fifth Circuit, recently held that judges “cannot take judicial notice, without even a scintilla of evidence, of what constitutes the community standard of decency at this or any other time. If such a standard exists at all, we would expect that it would be in a constant evolutionary and even revolutionary flux, the fact of which militates against our exercising uninformed judgment at any particular point in time. At best it would be a matter of pure chance as to whether we as a Court, or as individuals left to our own devices and without the aid of evidence, could determine the correct standard.” United States v. Groner, 5 Cir., 475 F.2d 550, 557 (decided January 11, 1972). Similarly, in a scholarly and careful opinion by Mr. Justice Tobriner, the California Supreme Court has held that an obscenity conviction cannot stand unless expert evidence is introduced at trial as to contemporary community standards.
“Relying principally on the well established doctrine that jurors should not be endowed with the prerogative of imposing their own personal standards as to the test of criminality of conduct, we hold that expert testimony should be introduced to establish community standards. We cannot assume that jurors in themselves necessarily express or reflect community standards; we must achieve so far as possible the application of an objective, rather than a subjective, determination of community standards. An even-handed application of the criminal law, even with evidentiary guidance * *, is sufficiently difficult in an area so confusing and intricate as obscenity. To sanction convictions without expert evidence of community standards encourages the jury to condemn as obscene such conduct or material as is personally distasteful or offensive to the particular juror. * * * ”
In re Giannini, 69 Cal.2d 563, 72 Cal. Rptr. 655, 663, 446 P.2d 535, 543 (1968). (Footnote omitted.) And of course I have already quoted Judge Leventhal’s *1254eloquent warning against individualistic, judicially imposed community standards. See text at pages 1250-1251 supra. The majority does not begin to explain why this court’s past refusal to “rest on our own untutored view” of the character of the activity involved, Huffman v. United States, supra, 152 U.S.App.D.C. at 254, 470 F.2d at 402, should not extend to this case. See also United States v. Klaw, 2 Cir., 350 F.2d 155, 170 (1965); Comment, Expert Testimony in Obscenity Cases, 18 Hastings L.J. 161 (1966).
To be sure, there are some jurisdictions where juries, at least in some cases, are allowed to evaluate putatively obscene material without expert guidance and measure it against their own notions of what the community standard is. See, e. g., United States v. Wild, 2 Cir., 422 F.2d 34 (1969), cert. denied, 402 U.S. 986, 91 S.Ct. 1644, 29 L.Ed.2d 152 (1971); Kahm v. United States, 5 Cir., 300 F.2d 78, cert. denied, 369 U.S. 859, 82 S.Ct. 949, 8 L.Ed.2d 18 (1962). But these cases are premised on the notion that juries themselves function as the embodiment of the community and therefore require no experts to tell them what the community thinks. Appellate judges are not, of course, the embodiment of the community. Indeed, their decisions derive legitimacy from the fact that the judiciary is insulated from transient community passions and prejudices. “Since an appellate court certainly does not in any sense compose a cross-section of the community, it cannot effectively carry out [its review] function in the absence of evidence in the record directed toward proof of the community standard.” In re Giannini, supra, 72 Cal.Rptr. at 664, 446 P.2d at 544. (Footnote omitted.)
Moreover, even in those jurisdictions permitting convictions without expert evidence as to community standards, the defendant is at least allowed to introduce such evidence as part of his rebuttal case. As Mr. Justice Frankfurter made clear years ago:
“ * * * Since the law through its functionaries is ‘applying contemporary community standards’ in determining what constitutes obscenity, * * * it surely must be deemed rational, and therefore relevant to the issue of obscenity, to allow light to be shed on what those ‘contemporary community standards’ are. Their interpretation ought not to depend solely on the necessarily limited, hit-or-miss, subjective view of what they are believed to be by the individual juror or judge. * * * ”
Smith v. California, 361 U.S. 147, 165, 80 S.Ct. 215, 225, 4 L.Ed.2d 205 (1959) (concurring opinion).
Yet in this case Von Sleichter was effectively deprived of the opportunity to prove that his conduct was in conformance with community standards — a deprivation which, according to Mr. Justice Frankfurter, “goes to the very essence of the defense and therefore to the constitutional safeguards of due process.” Id. at 165, 80 S.Ct. at 225. True, Von Sleichter never made an actual offer of proof as to community standards before the trial judge, and none of his evidence was excluded. But a close reading of the hearing record makes plain .that the prosecutor did not argue that there was probable cause to arrest Von Sleichter for disorderly conduct. Rather, his argument — and presumably the decision of the trial judge as well — was premised on the existence of probable cause to make a narcotics arrest.6 Surely we expect too much if we require the defendant to anticipate and rebut an argument which the Government failed to make either at *1255the hearing or in its appellate brief and which made its first, unheralded appearance in an opinion by one of the reviewing judges of the District of Columbia Court of Appeals. Placing this burden on the defendant would require him to introduce evidence of a lack of probable cause for arrest for every crime listed in the statute books. Thus, even if I agreed with the majority that judges are qualified to discern the community standard on their own, I would still hold that this case should be remanded to allow appellant to introduce evidence on this vital question.
One final point needs to be made. The majority suggests that the principles governing this case are somehow different because the question before us is not whether Von Sleichter actually committed the crime of disorderly conduct, but rather whether the police officer had probable cause to believe that the crime had been committed. But while it is true that the Government’s burden of proof is substantially lower in a probable cause hearing than it would be in a criminal prosecution, see, e. g., Brinegar v. United States, 338 U.S. 160, 175, 69 S.Ct. 1302, 93 L.Ed. 1879 (1949), this difference has never been thought sufficient to transmogrify a question of fact about which evidence is required into a question of law which the judge can decide without evidence. We have held that “[pjrobable cause is a plastic concept whose existence depends on the facts and circumstances of the particular case,” and that “[m]uch less evidence than is required to establish guilt is necessary.” Bailey v. United States, 128 U.S.App.D.C. 354, 357-358, 389 F.2d 305, 308-309 (1967). But we have also made clear that the burden is on the Government to establish probable cause and that it cannot meet this burden by putting on testimony as to the bare facts surrounding the arrest and depending upon the trial judge to fill in the gaps. I agree with the majority that “[w]e must give some credit to the policeman for ability to assess the ordinary impact of the sights and sounds he hears.” But there is no indication in this record that the arresting officer made such an assessment. The officer did not testify that he was aware of community standards in the area, that in his judgment appellant had violated those standards, or that he thought a breach of the peace was imminent. Rather, the officer’s testimony consisted entirely of a skeletal account of Von Sleichter’s conduct prior to the arrest. Our comments in Wrightson v. United States, 95 U.S.App.D.C. 390, 392, 222 F.2d 556, 558 (1955), where the Government testimony was similarly elliptical, seem particularly apt:
“The point here is that at the trial, when the search and the arrest were under attack as illegal, the officer and the prosecutor chose not to reveal what cause there was for the arrest and thus not to support its legality. There is law which governs arrest, that law is binding upon police officers, and persons arrested have a right to invoke it.”
This court has recently gone to great lengths to show that arrests are not somehow insulated from the constitutional protections which surround the rest of the criminal process. See Hall v. United States, 148 U.S.App.D.C. 831, 459 F.2d 831 (en banc). I am surprised and disappointed to see this salutary principle apparently abandoned so soon after its exposition.
By citing Hall, however, I do not mean to suggest that the law upon which I rely is in any way new. In fact, it has been clear for at least 50 years that, when the Fourth Amendment is raised to challenge the admissibility of evidence, the trial judge must hold a hearing to evaluate the competing factual assertions of the parties with respect to how the evidence was obtained. See Gouled v. United States, 255 U.S. 298, 41 S.Ct. 261, 65 L.Ed. 647 (1921); Nardone v. United States, 308 U.S. 338, 60 S.Ct. 266, 84 L.Ed. 307 (1939). Of course, such a hearing would be a meaningless formality if the trial judge were permitted to assume, without the benefit of evidence, that probable cause existed for a chai-*1256lenged search or arrest. See, e. g., United States v. Jeffers, 342 U.S. 48, 51, 72 S.Ct. 93, 96 L.Ed. 59 (1951); United States v. Johnson, 9 Cir., 425 F.2d 630, 632 (1970), cert. dismissed, 404 U.S. 802, 92 S.Ct. 21, 30 L.Ed.2d 35 (1971); Williams v. United States, 5 Cir., 382 F.2d 48, 50 (1967). Hence it is hardly surprising that the Supreme Court has stood ready to reverse a conviction when insufficient evidence appeared on the record to show that the search or seizure comported with the commands of the Fourth Amendment. See, e. g., Recznik v. City of Lorain, 393 U.S. 166, 168, 89 S.Ct. 342, 21 L.Ed.2d 317 (1968). Moreover, these procedural requirements are particularly vital where, as here, First Amendment rights are at stake — -rights which can be stifled as effectively by a wrongful arrest or seizure as by a wrongful conviction. See Monaghan, First Amendment “Due Process”, 83 Harv.L.Rev. 518, 538 (1970). Cf. Marcus v. Search Warrants, 367 U.S. 717, 81 S.Ct. 1708, 6 L.Ed.2d 1127 (1961); A Quantity of Copies of Books v. Kansas, 378 U.S. 205, 84 S.Ct. 1723, 12 L.Ed. 2d 809 (1964).
I am at a loss to understand why these long-standing principles should be ignored in this case. To be sure, usually the Fourth Amendment problem arises in the context of a challenge to the identity of the defendant, whereas here the issue is whether a crime was committed at all. But surely this is not a difference in principle. Probable cause requires not only a reasonable determination by the arresting officer that the suspect is probably guilty, but also a reasonable determination by the officer that criminal conduct has occurred. In this case, criminal conduct can only have occurred if community standards were violated. And we cannot know whether these standards have been violated until we have some evidence as to what the standards are.
It is possible, I suppose, to imagine some hypothetical case where a fact is not sufficiently obvious for judicial notice in a criminal trial but can be noticed under the lower standard of proof applicable in a probable cause hearing. But the theoretical existence of such a possibility should not trouble us in this case. Where, as here, there is no evidence of imminent violence, no evidence that anyone was offended by the defendant’s words, and only inconclusive evidence as to whether the words were overheard at all, I would require at least some testimony as to community standards before sanctioning a disorderly conduct arrest. As Mr. Justice Harlan has reminded us in an only slightly different context, “Any broader view * * * would effectively empower a majority to silence dissidents simply as a matter of personal predilections.” Cohen v. California, supra, 403 U.S. 15 at 21, 91 S.Ct. 1780 at 1786. And lest we underestimate the stakes in criminal prosecutions such as this one, Mr. Justice Harlan offers another admonition:
“To many, the immediate consequence of [First Amendment] freedom may often appear to be only verbal tumult, discord, and even offensive utterance. These are, however, within established limits, in truth necessary side effects of the broader enduring values which the process of open debate permits us to achieve. That the air may at times seem filled with verbal cacophony is, in this sense[,] not a sign of weakness, but of strength. We cannot lose sight of the fact that, in what otherwise might seem a trifling and annoying instance of individual distasteful abuse of a privilege, these fundamental societal values are truly implicated. * * * ”
Id. at 24-25, 91 S.Ct. at 1788.
In my view, the majority has failed to take sufficient cognizance of the “fundamental societal values” implicated in this case. I must, therefore, respectfully dissent.
Supplemental Opinion
. I am thus in full agreement with Judge Gallagher’s observations: “A person can hardly be more under arrest than when an officer plainly tells him he is. In criminal law these are critical words and not mere conversation.” Von Sleichter v. United States, D.C.C.A., 267 A.2d 336, 342 (1970) (dissenting statement).
. It is worth noting that the Government did not even advance this argument before the trial judge or the District of Columbia Court of Appeals. The argument makes its first appearance in Judge Kern’s opinion, in which neither Judge Gallagher nor Judge Nebeker concurred.
. Nor do I know of any case which authorizes a judge to assess the offensive potential of words spoken on the street on the basis of what he would do “if they were shouted in a courtroom, or * * * court corridor.” Majority op. at 1249. On the contrary, I had thought that the Constitution permitted sanctions on speech in and around courts which would not be permissible if applied to the streets. See Cohen v. California, 403 U.S. 15, 19, 91 S.Ct. 1780, 29 L.Ed.2d 284 (1971) ; Cox v. Louisiana, 379 U.S. 559, 562-564, 85 S.Ct. 476, 13 L.Ed.2d 487 (1965).
. The arresting officer testified as follows during the probable cause hearing:
“A. * * * X asked if I could speak with him and he shouted the profanities and began to run.
“Q. You didn’t ask any of these people if they heard this; did you?
“A. No, I didn’t.
“Q. You don’t know if they did or not; is that correct?
“A. That’s correct.”
Transcript at 13.
. The majority contends that appellant implicitly concedes that he violated community standards by arguing that cursing is a useful safety valve. Majority op. at 1249. No doubt this “concession” will come as a considerable surprise to appellant. Since judicial notice seems to be in vogue, I think we can notice the fact that people do all sorts of things to relieve pent up hostility which by no stretch of the imagination violate community standards.
. Thus the prosecutor made the following statement in summation:
“We don’t rely solely on the flight. We rely on three factors — the area; the transaction, the seemingly [sic] transaction the officer saw; and, the flight. Which, in the Government’s position, brings the case over the ledge to attain probable cause. It may be a fine question. That’s the Government’s position. We’ll leave it to the court to decide.”
Tr. at 19.