dissenting:
At issue in this case is whether the terminal areas controlled by the Port Authority at Kennedy, LaGuardia and Newark airports are traditional public fora, such that the Port Authority’s regulation against leafletting and solicitation must be examined under the strict scrutiny standard articulated in Perry Ed. Ass’n v. Local Educators’ Ass'n, 460 U.S. 37, 103 S.Ct. 948, 74 L.Ed.2d 794 (1983), and its progeny. Although this issue has remained open until today in this Circuit, two of our sister Circuits have recognized that airport terminals are traditional public fora. See Jamison v. City of St. Louis, 828 F.2d 1280, 1282-84 (8th Cir.1987), cert. denied, 485 U.S. 987, 108 S.Ct. 1289, 99 L.Ed.2d 499 (1988); Jews for Jesus, Inc. v. Board of Airport Comm’rs, 785 F.2d 791, 793-95 (9th Cir.1986), aff'd on other grounds, 482 U.S. 569, 107 S.Ct. 2568, 96 L.Ed.2d 500 (1987). In addition, at least three other Circuits have recognized airport terminals as generic “public fora,” without distinguishing whether they are of the traditional or designated category. See United States Southwest Africa/Namibia Trade & Cultural Council v. United States, 708 F.2d 760, 763-68 (D.C.Cir.1983); Fernandes v. Limmer, 663 F.2d 619, 626-27 (5th Cir. Unit A. Dec.1981); Chicago Area Military Project v. City of Chicago, 508 F.2d 921 (7th Cir.), cert. denied, 421 U.S. 992, 95 S.Ct. 1999, 44 L.Ed.2d 483 (1975); see also Gannett Satellite Information Network v. Berger, 716 F.Supp. 140, 149 (D.N.J.1989) (“Newark Airport is a public forum.”), aff'd in part & rev’d in part, 894 F.2d 61 (3d Cir.1990). But see International Caucus of Labor Comms. v. Metropolitan Dade County, 724 F.Supp. 917, 923 (S.D.Fla.1989) (Miami International Airport terminal not a traditional public forum). Moreover, we have previously noted in dicta that Krishna devotees practice san-kirtan in “various public forums — airports, *583bus stations, state fairs,” International Soc’y for Krishna Consciousness v. Barber, 650 F.2d 430, 434 (2d Cir.1981), and have held that the Port Authority’s bus terminal in New York City is a “public forum.” See Wolin v. Port of New York Auth., 392 F.2d 83, 88-91 (2d Cir.), cert. denied, 393 U.S. 940, 89 S.Ct. 290, 21 L.Ed.2d 275 (1968).1
Against this backdrop of authority, the majority acknowledges that before the Supreme Court’s recent plurality opinion in United States v. Kokinda, — U.S. —, 110 S.Ct. 3115, 111 L.Ed.2d 571 (1990), it was prepared to hold that the Port Authority’s airport terminals were traditional public fora. However, the majority believes that Kokinda has “altered” the Supreme Court’s public forum analysis, which it claims now must focus on the distinction between “passageways or other facilities that exist solely to facilitate the public’s carrying on of a particular endeavor and [those] that enable the public to carry out the multitude of purposes persons pursue in their daily life.” Supra. While the latter may be traditional public fora, the majority finds that Kokinda stands for the proposition that the former are categorically not public fora. Because I disagree with this assessment,' and instead believe that Kokinda is neither controlling nor a substantial deviation from the Court’s public forum analysis, I respectfully dissent.
Generally, traditional public fora include locations such as streets and public parks, “clearly held in trust, either by tradition or recent convention, for the use of citizens at large,” where discussions of public questions might ordinarily take place. Members of the City Council v. Taxpayers for Vincent, 466 U.S. 789, 815 n. 32, 104 S.Ct. 2118, 2134 n. 32, 80 L.Ed.2d 772 (1984); see also Perry, 460 U.S. at 45, 103 S.Ct. at 954. But the physical characteristics of a forum do not necessarily determine traditional public forum status. See Kokinda, 110 S.Ct. at 3120 (“The mere physical characteristics of the property cannot dictate forum analysis.”). Even a public street, for example, may not be a traditional public forum along that portion that runs within the confines of a military base. See Greer v. Spock, 424 U.S. 828, 96 S.Ct. 1211, 47 L.Ed.2d 505 (1976) (military reservation, including public street running through it, not traditional public forum). Nor does unlimited public access to a forum necessitate a finding that it is a traditional public forum. See id. at 836, 96 S.Ct. at 1216.
Similarly, the purposes for which a forum is designed and to which it is put are not alone determinative of its status. Certainly, the purposes of a forum are highly relevant to the traditional public forum inquiry. In Greer, for example, the exigent national security purposes of a military post weighed heavily in favor of the Supreme Court’s finding that the Fort Dix Military Reservation is not a traditional public forum. See 424 U.S. at 837-38, 96 S.Ct. at 1217-18. However, this observation does not compel what I believe is the majority’s simplistic conclusion that a forum’s purpose can alone dictate its status. Virtually all Government-owned property has an ostensible “purpose” other than the promotion of free speech, assembly, and debate. It can scarcely be argued that the Government builds streets, which are ordinarily traditional public fora, for the primary purpose of fostering First Amendment activities. At least, the Supreme Court does not think it can be. See Schneider v. State, 308 U.S. 147, 160, 60 S.Ct. 146, 150, 84 L.Ed. 155 (1939) (“Municipal authorities, as trustees for the public, have the duty to keep their communities’ streets open and available for movement of people and property, the primary purpose to which the streets are dedicated,.”) (emphasis added); see also Wolin, 392 F.2d at 89 (acknowledging that Port Authority bus terminal is “designed for precisely the purpose of transit,” but nevertheless finding it to be a public forum).
Further undermining the argument that purpose alone is determinative of a forum’s status are the Supreme Court’s numerous decisions holding that some governmental fora that serve no purpose other than communication of ideas are nevertheless not traditional public fora. See Cornelius v. NAACP Legal Defense and Education Fund, 473 U.S. 788, 804, 105 S.Ct. 3439, 3450, 87 L.Ed.2d 567 (1985) (federal charity fundraising drive); Perry, 460 U.S. at 46, 103 S.Ct. at 955 (school district’s internal *584mail system); United States Postal Serv. v. Council of Greenburgh Civic Ass’ns, 453 U.S. 114, 132, 101 S.Ct. 2676, 2686, 69 L.Ed.2d 517 (1981) (residential mailboxes); Lehman v. City of Shaker Heights, 418 U.S. 298, 302-04, 94 S.Ct. 2714, 2716-18, 41 L.Ed.2d 770 (1974) (plurality opinion) (advertising space inside city buses).
Nor do I think that the Supreme Court’s recent decision in United States v. Kokinda, — U.S. —, 110 S.Ct. 3115, 111 L.Ed.2d 571 (1990), announces a new rule that changes a forum’s purpose from being an important factor in public forum analysis to being the sole determinant of a forum’s status. At the start, it must be emphasized that Kokinda is of questionable precedential value, in that only four Justices agreed that the post office sidewalk at issue was not a traditional public forum. The fifth Justice who cast his vote in favor of the plurality’s result agreed that the regulation at issue was a reasonable time, place, and manner restriction, and conceded that “there remains a powerful argument that, because of the wide range of activities that the Government permits to take place on this postal sidewalk, it is more than a nonpublic forum.” See id. 110 S.Ct. at 3125 (Kennedy, J., concurring).
Moreover, even reading Kokinda as controlling, there is no basis to conclude from the plurality’s opinion that a forum’s purpose will, ipso facto, illuminate its status. In concluding that the postal sidewalk was not a traditional public forum, Justice O’Connor assessed not only the postal sidewalk’s purpose, but also its location, see id. 3121, the degree of public access afforded by the sidewalk, see id. at 3120, and whether such sidewalks had “ ‘traditionally served as a place for free public assembly and communication of thoughts by private citizens.’ ” Id. at 3121 (quoting Greer, 424 U.S. at 838, 96 S.Ct. at 1217). In fact, it is the latter inquiry that Justice O’Connor concluded “has illuminated our traditional public forum analysis, and that we apply today.” Id. This pivotal language demonstrates that the majority is simply wrong in concluding that Kokinda stands for the immutable proposition that facilities that exist solely for the “public’s carrying on of a particular endeavor ... are nonpublic fora.”
In short, I believe that Kokinda reaffirms the basic proposition that traditional public forum status does not turn on any single factor or characteristic. Rather, a more complex balancing determination is necessary, in which it must be determined whether “the character of the place, the pattern of usual activity, the nature of its essential purpose and the population who take advantage of the general invitation extended make it an appropriate place for communication of views on issues of political and social significance.” Wolin v. Port of New York Auth., 392 F.2d 83, 89 (2d Cir.), cert. denied, 393 U.S. 940, 89 S.Ct. 290, 21 L.Ed.2d 275 (1968); see also Fernandes, 663 F.2d at 626 (adopting Wolin’s factors for determining public forum status). In other words, we balance the governmental purposes of the forum on the one hand against the tradition of public access to that forum, and the interests of those who wish to use the forum for another purpose, 'on the other hand. See Cornelius, 473 U.S. at 800, 105 S.Ct. at 3448.2
Employing this balancing test, I would agree with the district court that the airport terminals at issue are traditional public fora. For the reasons set forth below, I see no meaningful distinction, in terms of the balance between the public’s tradition and interest in access and the governmental interest in controlling its property, between airport terminals and ordinary public streets or sidewalks. Both airport terminals as well as public streets and sidewalks *585are generally accessible to the public, available and appropriate at all times for discussion of public questions. See Jamison, 828 F.2d at 1283 (“[A]n airport terminal is much like a busy city street. Both are lined by shops, restaurants, newsstands, and other businesses, with travelers or other members of the general public coming and going as they please.”); United States Southwest Africa/Namibia, 708 F.2d at 764 (“[I]t seems clear that the public places in these airports are far more akin to such public forums as streets and common areas than they are to such nonforums as prisons, buses, and military bases.”); Fer-nandes, 663 F.2d at 627 (noting “lack of restrictions on entry by the general public, and the commercial, street-like character of the terminal concourses”). In fact, it is conceded that the terminal areas at Kennedy, LaGuardia and Newark contain restaurants, post offices, banks, drug stores, and many other commercial establishments that one would normally expect to encounter on the main thoroughfare of any town in America. The Government’s interest in controlling movement and traffic in both fora, while substantial, is no greater in the context of an airport terminal complex than in the context of the public roadway system. Accordingly, I see no greater incompatibility between governmental interests and free-speech interests on an ordinary public street than in the general circulation areas of New York’s major airport terminals.
Nonetheless, the Port Authority contends that airport terminals differ from traditional public fora such as streets, sidewalks, or parks in five different ways: their unique design and usage, their congestion problems, their security problems, their specialized user-fee financing, and the presence of captive audiences. I consider each of these in turn and find none of them individually or collectively persuasive.
Airport terminals and streets or sidewalks alike are designed and used for the efficient movement of travelers. I see no distinction between the amount of planning going into the design of airport passenger terminals compared to our modern urban roadway systems. Both are used for the same purposes — transportation—and I do not see how we can attach any constitutional significance to the fact that one involves travel by airplane and the other by automobile.
The Port Authority puts particular weight on the fact that airport terminals are designed to be geographically isolated from other parts of the community.3 Indeed, the Kokinda plurality emphasized the separation of a post office sidewalk by a parking lot from other municipal sidewalks as an important factor in determining that the post office sidewalk at issue was not a traditional public forum. See Kokinda, 110 S.Ct. at 3121 (“[T]he location and purpose of a publicly-owned sidewalk is critical to determining whether such a sidewalk constitutes a public forum.”) (emphasis added). However, as stated above, Kokinda is of questionable precedential value. Moreover, I am considerably less than certain that if the post office building in Kokinda had also included the vast array of shops, restaurants, and service centers characteristic of Kennedy, LaGuardia, and Newark Airports that even four Justices would have found that the sidewalk outside was not a traditional public forum. “The Postal Service [had] not ... open[ed] postal sidewalks to any First Amendment activity.” Id. at 3121. The Port Authority, by contrast, has invited and welcomed extensive expressive activity within its airport terminals.
I thus think the physical isolation of a forum is relevant only in light of the reasons for its isolation. Only when the Government isolates a particular forum for the purpose of regulating public access does location emerge as a relevant factor in favor of denying public forum status. So, for example, the isolation or separation of a military base from the surrounding-community in order to discourage prying eyes would render location highly relevant to the public forum inquiry. Where physical isolation of a forum is the product of happenstance or for functional purposes unrelated to regulation of public access, location becomes far less relevant. It is obvious why airport terminals are isolated and separated from the buildings and skyscrapers of downtown Manhattan and to a certain extent from quiet residential areas. *586The isolation — if it can be called that — of airports from other parts of the community is in deference to the physics and audio-phonics of aviation and not to any notion that the movement of people and ideas should be restricted. This physical separation does not imply that the airport is any less socially integrated into the everyday flow of people and ideas around the New York metropolitan region. To the contrary, few are the places where one can meet so many persons from diverse parts of the New York metropolitan area as inside its airport terminals.
As for problems of congestion, public streets and sidewalks no less than airport terminals are subject to tremendous traffic problems all over the New York metropolitan area. New Yorkers and others who have had the misfortune of trying to fly out of one of the New York airports during rush hour have discovered the hard way that the greatest risk of missing their flight might well come from the delay of traffic on public streets leading to the airport rather than from the crowds of people circulating within the airport terminals. Moreover, congestion might well weigh in favor of finding the airport to be a traditional public forum. The airport terminals are “an appropriate place for expressing one’s view precisely because the primary activity for which [they are] designed is attended with noisy crowds ... some unrest and less than perfect order.” Wolin, 392 F.2d at 90 (footnote omitted). I am confident that if the Port Authority can withstand the inefficiencies of movement occasioned by travelers’ stopping to visit the numerous vendors and boutiques lining the terminal’s corridors (and from which the Port Authority, of course, derives revenue), it can also tolerate at least some inconveniences attendant upon the exercise of plaintiffs’ fundamental First Amendment freedoms.
Similarly, security problems are of equal concern on public streets as in airport terminals, even if security problems on the streets are more innocuously labeled as “crime” rather than “terrorism.” No one would seriously contend that streets located in dangerous city neighborhoods are not traditional public fora by virtue of the Government’s greater need to promote public safety in these areas. The Port Authority no doubt has a powerful interest in maintaining security at its airports. But these concerns are redressable through reasonable time, place, and manner restrictions regulating access to secured areas.
The specialized financing of airports through user-fees, rather than general tax revenue, has no relevance to whether the terminals are traditional public fora. Many streets are also financed by user-fees, e.g., by tolls or by subdivision exactions. By contrast, the top secret areas of military bases, which no one would argue are traditional public fora, are financed by general tax revenues.
Finally, the Port Authority’s fear that travelers are a “captive audience” at the mercy of the Krishnas is not well-founded. The Krishnas seek access to persons only in general circulation areas and not check-in areas or baggage claim areas. On these facts, the Constitution requires those who are offended by the Krishnas to “avoid further bombardment of their sensibilities ... by averting their eyes.” Cohen v. California, 403 U.S. 15, 21, 91 S.Ct. 1780, 1786, 29 L.Ed.2d 284 (1971).
In light of my conclusion that the airport terminals are traditional public fora, it follows that the Port Authority may not impose a complete prohibition on First Amendment activities absent a compelling governmental interest. I would find no such interest here. Accordingly, the Port Authority’s virtual blanket prohibition on First Amendment activities inside the terminal areas it controls cannot be upheld. This much even the majority concedes insofar as leafletting is concerned.
It is of course true that the Port Authority may promulgate reasonable content-neutral time, place, and manner restrictions narrowly tailored to serve a significant governmental interest and permitting alternative channels for communication. See Perry, 460 U.S. at 45, 103 S.Ct. at 955. I have no doubt that the Port Authority’s interest in expediting the movement and safety of travelers is significant. See Heffron v. International Soc’y for Krishna Consciousness, Inc., 452 U.S. 640, 654, 101 S.Ct. 2559, 2567, 69 L.Ed.2d 298 (1981); Fernandes, 663 F.2d at 626; Wolin, 392 F.2d at 93. Therefore, I may well be persuaded that reasonable time, place, and manner regulations that, for example, require permits or badges for solicitors, or that limit the number of solicitors permitted, or that restrict the locations within the terminal buildings where solicitors may conduct their activities would pass constitutional scrutiny. See, e.g., Heffron (upholding limitation on Krishna solicitation at state fair to stationary booths); International Soc’y for Krishna Consciousness v. Rochford, 585 F.2d 263, 268-69 (7th Cir.*5871978) (upholding restrictions on locations of First Amendment activities within airport); Wolin, 392 F.2d at 94 (“[T]he Port Authority may set approximate and reasonable limitations on the number of persons who may engage in such activities at any specific time, the duration of the activity and the specific places within the building where the rights of expression may be exercised.”). However, a complete prohibition of First Amendment activity — such as the Port Authority’s regulation — is, in my mind, not a narrowly drawn regulation that allows for alternative channels of communication, and therefore not constitutionally permissible.
Finally, the distinction between leaflet-ting and fund solicitation that the majority draws I find to be without a difference. Both involve the same congestion, the same irritation, if you will. And may the content of the leaflets permitted be proscribed? If not, will solicitation of funds by leaflet be permitted? If not, what may the leaflets say?
I would affirm the judgment of the district court.
On Petition for Rehearing and Suggestion for Rehearing In Banc
April 25, 1991
A petition for rehearing containing a suggestion that the action be reheard in banc having been filed herein by counsel for the plaintiffs-appellees International Society for Krishna Consciousness, Inc., et al.,
Upon consideration by the panel that heard the appeal, it is
Ordered that said petition for rehearing is DENIED.
It is further noted that the suggestion for rehearing in banc having been transmitted to the judges of the Court in regular active service and to any other judge that heard the appeal and a poll of said judges having been taken, a majority of the Court has voted not to reconsider the decision in banc. Chief Judge Oakes dissents from the denial of the rehearing in banc in a separate opinion, in which Judges Newman and Cardamone concur.
. In Wolin, we did not specify whether the bus terminal was a traditional public forum or a designated public forum, as public forum analysis at that point in time had not yet given rise to the now-familiar tripartite distinction between traditional public fora, designated public fora, and non-public fora. Nevertheless, Wolin's analogy of the bus terminal to a public street, both with large crowds and stores and concessionaires, supports the proposition that under modern public forum analysis, the Port Authority bus terminal is a traditional public forum. See id. at 89.
. The public forum analytic framework has come under attack from many commentators, and even from four Supreme Court Justices, in part because it lends itself to mechanistic labeling of fora as public or non-public on the basis of their physical characteristics alone. See Kokinda, 110 S.Ct. at 3127 & n. 1 (Brennan, J., dissenting). Even if I agreed with the critics and wished to disregard public forum analysis, what the Supreme Court has adopted is not in my power to cast aside. Nevertheless, I see no necessary inconsistency between public forum analysis and a particularized balancing of First Amendment interests against governmental interests. See Greer v. Spock, 424 U.S. 828, 843, 96 S.Ct. 1211, 1220, 47 L.Ed.2d 505 (1976) (Powell, J., concurring). To put it differently, I view public forum analysis, properly applied, not as a device for inflexible categorization of governmental fora as public or non-public per se. Rather, public forum analysis is a useful shorthand method of evaluating the competing interests of individual speakers in exploiting important channels of communication, and of the Government in regulating the use of its own property. See L. Tribe, American Constitutional Law § 12-24, at 987 (2d ed. 1988); Farber & Nowak, The Misleading Nature of Public Forum Analysis: Content and Context in First Amendment Adjudication, 70 Va.L.Rev. 1219, 1234-35 (1984).
. Public highways, the Port Authority points out, isolate the airports from the surrounding communities. These highways, however, are most likely public fora. See Frisby v. Schultz, 487 U.S. 474, 481, 108 S.Ct. 2495, 2500, 101 L.Ed.2d 420 (1988) ("[A]1I public streets are held in the public trust and are properly considered traditional public fora.”). It would be anomalous to conclude that the airport's separation from the rest of the community by highways, which are themselves concededly public fora, could weigh in favor of finding the airport terminals to be non-public fora.