with whom NEWMAN, Chief Judge, KEARSE, mahoney, McLaughlin, Jacobs, WALKER, LEVAL, CALABRESI, and CABRANES, Circuit Judges, join, concurring in the result:
I concur in the result. I write separately because I believe that Judge Oakes’s opinion is an application of a somewhat broader principle than it acknowledges, namely that the First Amendment does not, in any context, protect coercive or obstructionist conduct that intimidates or physically prevents individuals from going about ordinary affairs.
Where advocacy is involved, the core notion of the First Amendment is the familiar marketplace of ideas. Abrams v. United States, 250 U.S. 616, 630, 40 S.Ct. 17, 22, 63 L.Ed. 1173 (1919) (Holmes, J., dissenting). The goal is to keep channels of communication free of govemmentally imposed impediments so that individuals who want to persuade their fellow citizens to adopt a particular belief or to alter intended conduct will have an opportunity to do so. Kovacs v. Cooper, 336 U.S. 77, 87, 69 S.Ct. 448, 453, 93 L.Ed. 513 (1949) (plurality opinion).
*395The notion of a marketplace of ideas gives primacy to the right to persuade largely for two reasons. First, we justifiably fear that those who govern will be relentlessly tempted to censor their critics. It is difficult to conceive of a working democracy that does not in some way protect those seeking to speak to voters in the hope of unseating those in power. See New York Times Co. v. Sullivan, 376 U.S. 254, 276, 84 S.Ct. 710, 724, 11 L.Ed.2d 686 (1964) (“broad consensus” that restraint upon criticism of government and public officials is “inconsistent with the First Amendment”); Mills v. Alabama, 384 U.S. 214, 218-19, 86 S.Ct. 1434, 1436-37, 16 L.Ed.2d 484 (1966) (“major purpose” of the First Amendment is to protect free discussion of governmental affairs). Second, we want citizens to have exposure to competing ideas in the faith that truth and humane values will have the best chance of prevailing under such a system. See Cohen v. California, 403 U.S. 15, 24, 91 S.Ct. 1780, 1787, 29 L.Ed.2d 284 (1971); Whitney v. California, 274 U.S. 357, 375-76, 47 S.Ct. 641, 648, 71 L.Ed. 1095 (1927) (Brandeis, J., concurring). This is, as stated, a matter of faith and not empirically testable. It may also be easy to quarrel with intellectually, until one is asked to fashion a workable, alternative system for the real world. Where advocacy is concerned, therefore, the First Amendment is concerned with persuasion.
However, the goal of freeing advocacy from governmental control has important, sometimes overriding, implications for audiences as well as speakers. Rowan v. United States Post Office Dep’t, 397 U.S. 728, 736, 90 S.Ct. 1484, 1490, 25 L.Ed.2d 736 (1970) (“[T]he right of every person ‘to be let alone’ must be placed in the scales with the right of others to communicate.”). A legal rule implementing a market-place of ideas protects advocates in the interests of the audience. Pacific Gas & Elec. Co. v. Public Utils. Comm’n, 475 U.S. 1, 8, 106 S.Ct. 903, 907, 89 L.Ed.2d 1 (1986) (“By protecting those who wish to enter the marketplace of ideas ..., the First Amendment protects the public’s interest in receiving information.”) (plurality opinion). It inexorably requires that individuals who are subject to advocacy be left free to make up their own minds and to adhere to their convictions or intended courses of conduct should they remain unpersuaded. Advocates have a right to a full opportunity to appeal to reason, interest, policy, or morals, but they have no right to seek to alter the mind or conduct of the audience by coercion or obstruction. See Rowan, 397 U.S. at 738, 90 S.Ct. at 1491 (“[N]o one has a right to press even ‘good’ ideas on an unwilling recipient.”); Kovacs, 336 U.S. at 86-87, 69 S.Ct. at 453 (A visitor cannot “insert a foot in the door and insist on a hearing,” and a passer-by may be “offered a pamphlet in the street but cannot be made to take it.”); Schneider v. State, 308 U.S. 147,162, 60 S.Ct. 146, 151, 84 L.Ed. 155 (1939) (acknowledging right to distribute literature “to one willing to receive it”); cf. International Soc’y for Krishna Consciousness, Inc. v. Lee, 505 U.S. 672, -, 112 S.Ct. 2701, 2708, 120 L.Ed.2d 541 (1992) (“risk[] of duress” presents an “appropriate target of regulation”). Those who seek to do more than peacefully persuade individuals, therefore, are themselves acting in a way quite inconsistent with a marketplace of ideas.
We thus need not tolerate coercive or obstructionist conduct solely because it serves some passionate ideology or interest. The First Amendment protects peaceful communication, not self-indulgence. Nor need we tolerate such conduct because it makes the advocate feel good. The First Amendment is not a governmental version of the psychiatrist’s couch.
In determining whether conduct is coercive or obstructionist, some leeway or legal “breathing space” should be allowed to the advocate so that censorship is not disguised as the keeping of order. See Boos v. Barry, 485 U.S. 312, 322, 108 S.Ct. 1157, 1164, 99 L.Ed.2d 333 (1988); Cohen, 403 U.S. at 26, 91 S.Ct. at 1788. For example, shame is a form of persuasion often expressed in epithets indicating immorality or selfishness that may be regarded by a substantial segment of the public as insulting. However, insults that contain neither threats of coercion that intimidate nor fighting words that create the possibility of imminent violence, see Chaplinsky v. New Hampshire, 315 U.S. *396568, 572, 62 S.Ct. 766, 769, 86 L.Ed. 1031 (1942), must be tolerated. Polarizing debates lead to strong language, and an attempt to impose rules of etiquette would inhibit the debate by forcing speakers to weigh every word and empower government with far too much discretion to define permissible and impermissible modes of advocacy.
The leeway afforded advocates is the greatest where the audience is the general public or is a voluntary rather than involuntary participant. See Madsen v. Women’s Health Center, Inc., — U.S. -, -, 114 S.Ct. 2516, 2527, 129 L.Ed.2d 593 (1994); Frisby v. Schultz, 487 U.S. 474, 484-85, 486, 108 S.Ct. 2495, 2502-03, 2503, 101 L.Ed.2d 420 (1988). The general public is a diffuse body that is not easily coerced by words or non-violent symbolic acts. Where more than fleeting exposure to the advocacy is avoidable, moreover, the chance of coercion is low.
However, where specific individuals are targeted at locations difficult or inconvenient for them to avoid, the First Amendment’s tolerance of plausibly coercive or obstructionist protest is least, precisely because the temptation and opportunity for coercion or obstruction are at their greatest. See, e.g., Frisby, 487 U.S. at 486, 108 S.Ct. at 2503 (affording lesser protection for “targeted picketing” that “inherently and offensively intrudes on residential privacy”). The law need not blind itself to the fact that targeted protest is often directed at individuals who have already given considerable thought to the subject at hand, do not agree with the protestor, and are quite unlikely to be moved by peaceful persuasion. In such circumstances, the protestor may become frustrated, refuse to take “no” for an answer, and be tempted to go beyond persuasion in seeking to change the target’s views or conduct.
Generally, targeted persons, whether they be customers of a business, residents of a particular house, workers at a firm, the trustees of a university, military recruiters, delegates at a political convention, or patients and employees of an abortion clinic, are not voluntarily exposed to the protests. Rather, they have been sought out at a particular location that they can avoid only at a cost, see, e.g., Krishna Consciousness, 505 U.S. at -, 112 S.Ct. at 2708 (upholding restrictions on expression at an airport in part because travelers are often weighted down by baggage, in a hurry, and some “cannot easily avoid ... solicitation”), and the opportunity to bring direct coercive or obstructive pressure on the individuals is great, see Madsen, — U.S. at -, 114 S.Ct. at 2528 (“The First Amendment does not demand that patients at a medical facility undertake Herculean efforts to escape the cacophony of political protests.”). Moreover, the power of televised pictures of coercive or obstructionist conduct to chill others is considerable and an independent reason motivating some protestors to go beyond peaceful persuasion. For example, one may be legitimately skeptical about whether residential picketing is designed to persuade anyone by the strength of the argument presented. See, e.g., Frisby, 487 U.S. at 486, 108 S.Ct. at 2503 (“[P]icket-ing ... narrowly directed at [a] household ... do[es] not seek to disseminate a message to the general public, but to intrude upon the targeted resident, and to do so in an especially offensive way.”). Rather, it seeks to compel the homeowner to buy peace by abandoning convictions and to warn those who might share the homeowner’s convictions to alter their views or suffer the same fate.
In my view, therefore, there is no right to invade the personal space of individuals going about lawful business, to dog their footsteps or chase them down a street, to scream or gesticulate in their faces, or to do anything else that cannot be fairly described as an attempt at peaceful persuasion. Moreover, an extensive record of coercion or obstruction is not required to justify an injunction requiring physical separation and an end to such conduct. The power of even isolated threats or obstructions to exert a chilling effect on some members of society need not be ignored. The timid have a right to go about their business, and it is no embarrassment for a federal court to say so.
My emphasis on protecting targets of protest from coercive or obstructionist conduct should not be mistaken for a failure to acknowledge that the state has *397other interests in preventing such behavior. In Madsen, for example, the Court noted the state’s “strong interest in ensuring the public safety and order, in promoting the free flow of traffic on public streets and sidewalks, and in protecting the property rights of all its citizens.” — U.S. at -, 114 S.Ct. at 2526. Society has a strong interest in permitting businesses, universities, and medical clinics to operate, in protecting commerce and the flow of traffic, and in allowing individuals to have privacy in their residences and otherwise to use their property. These are important interests and are frequently invoked to justify restraints on protestors. However, I believe it worth emphasizing that coercion or obstruction does not gain First Amendment protection simply because no one is physically injured, traffic moves, and private property is not invaded. A placid scene that is the result of citizens not going where they wish to be in order to avoid bullying is hardly consistent with a marketplace of ideas.
I would also note that my views are not an application of doctrines concerning public or non-public fora, see Krishna Consciousness, 505 U.S. at -, 112 S.Ct. at 2705; Hague v. CIO, 307 U.S. 496, 515-16, 59 S.Ct. 954, 963-64, 83 L.Ed. 1423 (1939) (Roberts, J.), or captive audiences, see Frisby, 487 U.S. at 484-85, 108 S.Ct. at 2502-03; Public Utils. Comm’n v. Pollak, 343 U.S. 451, 468, 72 S.Ct. 813, 823, 96 L.Ed. 1068 (1952) (Douglas, J., dissenting), although those doctrines inform some of the distinctions discussed above. Those doctrines define circumstances in which government may restrict speech that is otherwise protected. My point is that coercive or obstructionist conduct is not protected by the First Amendment in any forum and regardless of the nature of the audience. As noted, the legal breathing space afforded the protestor in defining coercion or obstruction may vary according to the forum and ability of the audience to avoid the protest at little cost. However, actual coercion or obstruction is not protected speech at any time or in any place.
Turning to the present ease, affirmance of the district court is dictated generally by this principle. The protests here were not directed at the general public. The targets were particular individuals who either worked at, or were patients of, the abortion clinic. The location was selected precisely because the targets had to choose either to pass by the protestors or to quit work or forgo services at the clinic. There was, moreover, under either the view of the panel majority or panel dissent, a considerable amount of obstruction and bullying by both the protestors and the self-described counselors. It appears undisputed that a physical blockade of the clinic was planned and was forestalled only by the issuance of a temporary restraining order. Pro-Choice Network v. Project Rescue, 799 F.Supp. at 1424. Moreover, numerous coercive and obstructionist acts occurred before the order issued. The protestors “crowd[ed] around people trying to enter the facilities in an intimidating and obstructing manner, and gráb[bed], push[ed] and shov[ed] the patients, patient escorts and staff ...” Id. The self-described counselors would become “angry and frustrated” when persons showed indifference to their proffers and would “then turn to harassing, badgering, intimidating and yelling at the patients and patient escorts -” Id. at 1425. The counselors would “often crowd around patients, invade their personal space and raise their voices to a loud and disturbing level.” Id. After issuance of the restraining order, at least two further acts of obstruction occurred.
In Madsen, the Supreme Court approved an injunction barring anti-abortion protestors from entering a 36-foot zone around the entrances to an abortion clinic. — U.S. at -, 114 S.Ct. at 2527. The majority of the panel that originally decided this appeal believed that the record in Madsen revealed a considerably greater history of disruption than does the record in the instant matter. Pro-Choice Network v. Schenck, 67 F.3d 359, 371 (2d Cir.1994). I agree. However, the 15-foot buffer zone is not a significant infringement upon speech and does not require an extensive record of bullying to be justified. A nose-to-nose confrontation is hardly essential to the conveying of the protestors’ views and will, given their history of intimidation, be reasonably perceived by those seeking to enter the clinic as intimidating. *398I also see no constitutional violation in the cease and desist order directed at the counselors. They distinguish their role from that of protestors by claiming that they seek only to provide patients with information relating to the availability of alternative services, etc. However, some of the counselors in the instant matter have in the past resorted to bullying as soon as their proffer of advice was rebuffed, and I see no First Amendment barrier to a requirement that, once the would-be counselors have had an opportunity to offer advice and it is declined, they must respect the buffer area applicable to protestors.
Moreover, the court is unanimous in the view that some form of injunction was justified, and Judge Arcara’s order was an entirely reasonable response to the events that occurred. Indeed, an order that commands a specific physical separation but does not prevent communication in an ordinary tone of voice gives certainty to the advocate as to what is lawful and thus avoids the risk of a contempt adjudication for violating a vague order not to interfere with entry or exit. One irony of the appellants’ position in this court is that they attack the specific portions of the order as unconstitutionally rigid and the more general portions of the order as impermissibly vague.1
I thus concur in the result.
. I do not agree with the panel’s view that, because the protestors generally (but not entirely) complied with the restraining order, a more specific buffer zone and cease and desist provision could not be imposed by the preliminary injunction. The bullying and planned blockade did not disappear from the record because most of the protestors were on good behavior while the preliminary injunction was under consideration. In my view, the district court had ample discretion to impose specific restraints in the preliminary injunction to insure that such conduct was not renewed.