concurring in part and dissenting in part:
This case presents the question whether the City of Pasco, Washington (“Pasco”) violated the First Amendment when it invited local artists to display their work in *1084its new city hall and then refused to permit certain pieces of art to be placed or remain on display after the art provoked controversy. I concur in the reversal of the award of summary judgment to Pasco, but dissent from the award of partial summary judgment to the artists based on the majority’s conclusion that Pasco created a designated public forum. I respectfully dissent because factual issues of Pasco’s intent remain, United States Supreme Court precedent dictates a different rule regarding the creation of a designated public forum in light of the summary judgment standard, and the Ninth Circuit should wait for development of a full factual record before creating a new rule. I also dissent because the majority’s ruling is unfair to and unworkable for cities within the Ninth Circuit and likely will discourage cities from experimenting with public art displays, which ultimately will be more harmful for artists than permitting Pasco to have its day in court.
I
In 1994, Pasco remodeled a former school building and converted it into a new city hall. In an effort to decorate the bare walls of the new space, Gary Crutch-field, the City Manager, and his administrative assistant, Kurt Luhrs, decided to institute an experimental program whereby local artists would be invited to display their work in the newly remodeled building. Because Crutchfield did not want city employees to devote time and energy administering this program, Luhrs sought assistance from the Arts Council of Mid-Columbia Region (“the Arts Council”), a private entity that promoted the arts. Luhrs’ primary contact at the Arts Council was its director, Barbara Gurth. Luhrs and Gurth agreed that the Arts Council would locate the art and administer the program in exchange for a $500 quarterly fee.
During initial discussions, Luhrs made clear to Gurth that Pasco was concerned about generating controversy: He instructed her that, in selecting art for the program, work of a “questionable nature” should not be displayed. In a letter to Gurth, he described Pasco’s concerns about this experimental program:
During our conversation I got the impression that your board felt that our approach may not be a commitment to-the long term management of such a project. This is far from the case. Both Gary and I feel that this approach will ensure support when I bring this item to council in a public meeting. Personally, my greatest fear is bringing such a program to council and having various citizens with a conservative “bent” raise issues that have caused trouble for the National Endowment of the Arts, i.e. offensive or politically motivated art. Through our discussions, I feel assured that the Arts Council will not use the City Hall Gallery as a venue for controversy. Nevertheless, without a demonstration, I feel that the unfounded fears of a few citizens would ruin this great opportunity for introducing the arts to our citizens.
Consistent with Pasco’s concerns and shortly after announcing the project, Gurth sent a letter to artists who had expressed interest in the project and warned of the need to avoid controversial subject matter:
Subject matter: Wide open, but with the restraints that would be expected with a public arts project paid for with public money. To offer a quote from a city official’s letter regarding this project “... my greatest fear is ... having various citizens ... raise issues that have caused trouble for the National Endowment for the Arts, i.e., offensive or politically motivated art. Through our discussions, I feel assured that the Arts Council will not use the City Hall Gallery as a venue for controversy.”
Indeed, the Arts Council will not. I have worked for five years to bring the cities on board for the arts, and I will not jeopardize our progress. Additionally, I do not think that regional art in this area presents a problem in this *1085regard, but the Council will reserve the right to reject subject matter that the committee feels may present a problem for a conservative public sector. As an art historian whose major is in post-WW II American Art, this is an enormous compromise for me, but as an arts administrator motivated by the belief that the arts are necessary to the construction of a well balanced society, I realize compromise is necessary. Rejection of a particular piece should not, therefore, be construed by an exhibiting artist as censorship, but as a means of compromise to encourage public partnerships for aits action.
Thus, in announcing the program to the arts community, Gurth suggested that controversy would destroy the project and that artists submitting work could expect to be censored if their work was controversial.
There is, however, ambiguous and contradictory evidence about who would take responsibility for screening submitted artwork. Gurth’s announcement letter suggested Arts Council responsibility (“the Council will reserve the right to reject subject matter that the committee feels may present a problem for a conservative public sector”). City Manager Crutchfield also understood that the Arts Council would screen works for suitability, testifying that the “Aits Council would do everything.” Similarly, Crutchfield’s assistant Luhrs testified that Pasco had no expertise in art and that the Arts Council should ensure that artists presenting political or offensive art would not be invited to participate. On the other hand, after the onset of this litigation, Gurth testified that she believed Pasco — not the Arts Council— would be responsible for screening. She said that the Arts Council viewed itself as a “middleman” and although it planned to convene a screening committee if the program became permanent, it did not intend to do so, and did not do so, in the initial stage. More specifically, she testified:
Mr. Luhrs did express a concern on the telephone that work of a questionable nature not be submitted and I assured him that I didn’t think that was a problem. This was not a progressive area, where the visual arts are concerned, and it’s made up mainly of landscape and flower painters, and there had never been a question about work being questionable in its content. So I assured him I didn’t think it would be a problem, but that they [Pasco] would have to make that judgment.
On cross-examination, however, Gurth conceded that she may have agreed with Luhrs that “[w]e would both probably have a hand” in content control.
The course of events of the short-lived experiment is not entirely clear. The exhibits were scheduled to run three months at a time. The first exhibit ran from August through October 1995 and the second ran from November 1995 through January 1996. There is no evidence that either Pasco or the Arts Council screened the artwork before its display in these two exhibits. Although some individuals criticized a few pieces, there is no evidence that Pasco or the Arts Council prevented the display of any work or that they removed any work from display in response to public complaints.
The art displayed during these two exhibits generated very little controversy. Some works depicted nudity, but Gurth testified these were “abstracted” and nothing in the record indicates that they caused a genuine controversy. The record indicates, however, some controversy surrounding a sculpture known as “The Starving Man.” Gurth testified, “from the very first exhibit, some people wanted one of the works removed. It was a clothed figure .... ■ The figure was [a] life-size ceramic figure, and he called it ‘The Starving Man.’ ” When asked about this criticism, Gurth testified: “It didn’t send a good message, a starving man.” But. she also testified that there was never any serious discussion about removing it. Apart from “The Starving Man,” the first two exhibits were uneventful.
*1086Appellants Hopper and Rupp were invited to participate in the third exhibit, scheduled to run from February through April 1996. Luhrs testified that before the third exhibit, Gurth called to tell him she had not reviewed Hopper’s work because she was out of town and that Luhrs might want to “check it out.” This call prompted Luhrs to review Hopper’s pieces when they arrived, although he had not reviewed any previous submissions. Gurth denied this telephone conversation took place. In any event, when Hopper submitted prints showing a naked couple in varied scenes, Crutchfield and Luhrs refused to hang the prints.
Rupp submitted a sculpture titled “To the Democrats, Republicans, and Biparti-sans,” which depicted a woman’s bare buttocks “mooning” the viewer. Although Crutchfield and Luhrs initially permitted Rupp’s piece to be displayed, complaints ultimately prompted Crutchfield to direct its removal.
Hopper and Rupp sued both Pasco and the Arts Council asserting, inter alia, a cause of action against Pasco under 42 U.S.C. § 1988 for the alleged violation of the First Amendment. Hopper and Rupp sought declaratory and injunctive relief, damages, and attorney’s fees. On cross-motions for summary judgment, the district court ruled for Pasco and against Hopper and Rupp. This appeal followed.
II
In Perry Education Assn. v. Perry Local Educators’ Assn., 460 U.S. 37, 103 S.Ct. 948, 74 L.Ed.2d 794 (1983), the Supreme Court adopted a “forum analysis” framework for determining when restrictions of speech on government property are permissible. Within this framework, the Court recognized three types of forums: traditional, designated, and nonpublic. Id. at 45 — 46,103 S.Ct. 948.
There is no question that Pasco’s city hall is not a traditional public forum. Instead, as the majority correctly recognizes, resolution of the dispute in this case turns largely on whether Pasco converted its city hall from a non-public forum, where it would have relatively broad power to regulate expression, into a designated public forum, where its constitutional ability to regulate speech would be strictly limited. More specifically, we are not dealing here with an attempt by Pasco to censor or preclude the showing of artwork by Hopper, Rupp, or any other artist in traditional public forums. Nor are we concerned with an attempt generally to preclude the showing of “controversial” artwork at all or selected locations within Pasco city limits. Instead, this case is about Pasco’s right to regulate expression in a space that normally is not a public forum and that can become a public forum only through Pas-co’s intentional action creating one. “[T]he First Amendment does not guarantee access to property simply because it is owned or controlled by the government.” United States Postal Service v. Council of Greenburgh Civic Ass’ns., 453 U.S. 114, 129, 101 S.Ct. 2676, 69 L.Ed.2d 517 (1981).
The key to whether the government has created a designated public forum is governmental intent. Cornelius v. NAACP Legal Defense and Educational Fund, Inc., 473 U.S. 788, 802, 105 S.Ct. 3439, 87 L.Ed.2d 567 (1985). The Supreme Court has identified three primary factors to consider in determining whether the government intended to create a designated public forum: (1) what the government said; (2) what the government did; and (3) the compatibility of the space with the expression.
The government does not create a public forum by inaction or by permitting limited discourse, but only by intentionally opening a nontraditional forum for public discourse. Accordingly, the Court has looked to the policy and practice of the government to ascertain whether it intended to designate a place not traditionally open to assembly and debate as a public forum. The Court has also examined the nature of the property and its compatibility with expressive activity to discern the government’s intent.
*1087Id. at 802, 105 S.Ct. 3489 (citing Perry, 460 U.S. at 47, 103 S.Ct. 948) (internal citations omitted). We must consider these factors to determine whether they “indicate an intent to designate a public forum dedicated to expressive activities.” DiLoreto v. Downey Unified Sch. Dist. Bd. of Educ., 196 F.3d 958, 965 (9th Cir.1999). This is precisely where, in my view, the majority goes astray as it does not consider each of the Cornelius factors in light of the correct summary judgment standard.
In ruling on a motion for summary judgment, “[cjredibility determinations, the weighing of the evidence, and the drawing of legitimate inferences from the facts are jury functions, not those of a judge.... The evidence of the non-movant is to be believed, and all justifiable inferences are to be drawn in his [or her] favor.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). Summary judgment is proper only if there is “no genuine issue as to any material fact.” Id. at 258, 106 S.Ct. 2505. A “genuine issue” exists, precluding summary judgment, as long as “the evidence is such that a reasonable jury could return a verdict for the nonmoving party.” Id. at 248, 106 S.Ct. 2505. “[A]ll that is required is that sufficient evidence supporting the claimed factual dispute be shown to require a jury or judge to resolve the parties’ differing versions of the truth at trial.” Id. at 249, 106 S.Ct. 2505 (quoting First Nat’l Bank of Arizona v. Cities Serv. Co., 391 U.S. 253, 288-89, 88 S.Ct. 1575, 20 L.Ed.2d 569 (1968)).
In light of these cautionary and salutary principles, summary judgment for either party is inappropriate because the three Cornelius factors support reasonable inferences in favor of either party on the issue of Pasco’s intent to create a designated public forum.
Ill
If the evidence is viewed in the light most favorable to Hopper and Rupp, the non-moving parties with respect to Pasco’s motion for summary judgment, and if all permissible inferences are drawn in their favor, there is no question the district court erred in granting Pasco summary judgment based on the finding that the city hall was a non-public forum. To this extent, I agree with the majority’s analysis.
Pasco’s initial communications with the Arts Council indicated a concern to avoid controversial subject matter. Viewing the evidence in the light most favorable to Hopper and Rupp, however, Pasco’s conduct does not reflect its expressed intent. Before the incidents with Hopper and Rupp, neither Pasco nor the Art’s Council pre-screened the artwork. Evidently all submitted art was displayed. Although some art generated a small degree of controversy and was arguably political (e.g., the “Starving Man”), Pasco made no effort to remove any art. The “City Hall Art Gallery” was located in the lobby of city hall, which a reasonable juror might perceive to be a natural place for the display of all types of art if all inferences are given the artists. Moreover, there is no evidence that the displays, even if controversial, would have directly affected the running of the city government. Cf. Cornelius, 473 U.S. at 792, 105 S.Ct. 3439 (before regulations “the increasing number of entities seeking access to federal buildings and the multiplicity of appeals disrupted the work environment and confused employees who were unfamiliar with the groups seeking contributions”).
Viewing this evidence in the light most favorable to Hopper and Rupp, a reasonable juror could conclude that even if Pas-co initially intended to permit display only of selected, non-controversial art, it abandoned that intent. Similarly, in this light, the nature of the space and Pasco’s conduct could support the conclusion that Pas-co intended to open its city hall to the unrestricted display of art.
The district court assumed the role of trier of fact when it concluded that Pasco’s stated policy trumped any reasonable inferences that might be drawn from Pasco’s actual practice or from the nature of the space. This was error and, as the majori*1088ty correctly determines, summary judgment for Pasco must be reversed. See Liberty Lobby, 477 U.S. at 250-51, 106 S.Ct. 2505.
IV
I respectfully disagree, however, with the majority’s further holding that Hopper and Rupp are entitled to partial summary judgment. We must view the evidence in the light most favorable to Pasco when considering Hopper and Rupp’s cross-motion for summary judgment. Viewing the evidence in this light, a reasonable juror could find that Pasco did not intend to open its city hall as a public forum.
It is undisputed that Pasco’s stated intent at the outset of the experimental project was not to open its city hall to unrestricted or indiscriminate artistic activity. Viewed in the light most favorable to Pasco, this undisputed evidence and reasonable inferences therefrom alone are sufficient to preclude a partial summary judgment for Hopper and Rupp on the public forum and free speech issues. To grant summary judgment to Hopper and Rupp in light of this evidence of Pasco’s intent, we must either: (1) weigh this important evidence and determine it counts for nothing; (2) conclude that Pasco did not mean what it said; or (3) conclude that Pasco meant what it said but later changed its mind. However, we may not weigh and interpret evidence because this is a function solely reserved for the trier of fact.
Reasonable inferences that a trier of fact may draw from the nature of the space also preclude granting partial summary judgment to Hopper and Rupp. Both Pasco and the Arts Council expressed concern that continued funding might be jeopardized by a display of art that generated controversy. Luhrs’ letter to the Arts Council stated: “my greatest fear is bringing such a program to council and having various citizens with a conservative ‘bent’ raise issues that have caused trouble for the National Endowment of the Arts, i.e. offensive or politically motivated art.” In response, Gurth’s announcement letter noted: “I have worked for five years to bring the cities on board for the arts, and I will not jeopardize our progress.” Viewing these statements in the light most favorable to Pasco, a trier of fact could reasonably conclude that the space was not, in fact, compatible with an open public forum.
Similarly, Pasco’s conduct was not unambiguous and does not necessarily support the inference that Pasco intended to create a designated public forum. Even assuming that Pasco did not perform any screening until it encountered Hopper’s and Rupp’s pieces, this does not compel the conclusion that Pasco intended to create a designated public forum. A reasonable juror, viewing the evidence in the light most favorable to Pasco, could conclude that any lack of screening was the result not of a conscious plan, but of inaction caused by inadvertence or a breakdown in communications. While Pasco thought the Arts Council was screening the artwork to avoid controversial material, the Arts Council believed it was a middleman subject to Pasco’s final determination to exclude objectionable work. As Cornelius makes clear, “[t]he government does not create a public forum by inaction or by permitting limited discourse, but only by intentionally opening a nontraditional forum for public discourse.” 473 U.S. at 802,105 S.Ct. 3439.
Similarly, a reasonable juror could conclude that the brief experimental period before Pasco excluded Hopper’s and Rupp’s works is too short, and the record too sparse, to show any unambiguous intent to open the forum. Viewing the evidence in the light most favorable to Pasco, there was no substantial controversy before the controversy regarding Hopper’s and Rupp’s pieces. As noted above, the earlier abstracted nudes generated virtually no comment. And although the “Starving Man” piece generated some dispute, Gurth testified:
Question: Did you have a discussion about removing it, a serious discussion?
Answer: No, it didn’t escalate to that point. I simply alerted the artist that it *1089may happen, and we were prepared to take it out if Pasco insisted. And so the artist said that was fíne by him. He had no qualms with it, it would only enhance his reputation.
Viewing this evidence in the light most favorable to Pasco, a reasonable juror could infer not that Pasco failed consistently to enforce a policy of excluding controversial art, but that before encountering Hopper’s and Rupp’s pieces, Pasco never faced occasion or need to enforce the policy. The majority’s conclusion that Hopper and Rupp are entitled to summary judgment depends on its weighing of these conflicting inferences — a function properly reserved for the trier of fact.
V
I respectfully do not agree with the majority’s reliance on Christ’s Bride, Planned Parenthood, Grace Bible, and United Food to grant summary judgment to Hopper and Rupp. When fully examined, these cases highlight the ambiguity of the evidence in this case as well as the need for a full trial to develop an adequate record.
In Christ’s Bride Ministries, Inc. v. SEPTA, 148 F.3d 242, 244 (3rd Cir.1998), cert. denied, 525 U.S. 1068, 119 S.Ct. 797, 142 L.Ed.2d 659 (1999), the Third Circuit reviewed the decision of the Regional Transit Authority (“RTA”) to remove a poster from a bus submitted by an antiabortion group, which stated “Women Who Choose Abortion Suffer More & Deadlier Breast Cancer.” There, following a bench trial the district court found that the Transit Authority had not created a public forum and accordingly could permissibly remove the poster. Id. The Third Circuit, having the benefit of a full record before it, reversed. The court noted that the RTA had a long practice of accepting numerous ads on a wide range of topics, including religious and political messages, explicit ads regarding safe sex, abstinence and AIDS, and ads in favor of abortion rights. Of the many such controversial advertisements submitted, the RTA rejected or requested modification of only three — none of which related to abortion. Id. at 251-52. The Third Circuit thus concluded that the RTA’s practice of permitting numerous controversial advertisements and several abortion-related advertisements indicated an intent to create a public forum for the display of material relating to abortion. Id. at 252. Moreover, the court found that the RTA’s stated policy also supported the inference that it intended to create a public forum because the policy did not provide for the type of limitations on speech reflected by the removal of the anti-abortion poster. Id.
Here, in stark contrast, there is no record of a long practice allowing controversial artwork. During the experimental period, there was only a scintilla of evidence suggesting some inchoate controversy surrounding one piece1 and no evidence of any serious controversy. We cannot correctly conclude as a matter of law that Pasco had a consistent practice of permitting controversial artwork. In the short time the project proceeded, Pasco was faced with few instances of even arguably controversial art. And unlike Christ’s Bride, Pasco’s stated policy is not consistent with an intent to create a designated public forum. Christ’s Bride does not support the majority’s conclusion that Hopper and Rupp are entitled to partial summary judgment rather than a trial.
Similarly, neither Planned Parenthood Ass’n v. Chicago Transit Auth., 767 F.2d 1225 (7th Cir.1985), nor Grace Bible Fellowship, Inc. v. Maine Sch. Admin. Dist. #5, 941 F.2d 45 (1st Cir.1991), supports the proposition that summary judgment is appropriate where intent to create a designated public forum is ambiguous, as it is here. In Planned Parenthood, following a bench trial the district court found that the Chicago Transit Authority (“CTA”) created a limited public forum on its buses *1090and thus improperly excluded the plaintiffs abortion-related advertising. 767 F.2d at 1228. The district court found that the CTA’s alleged policy of excluding certain material (such as the advertisement at issue) was contrived for the lawsuit and that over a ten-year period the CTA permitted a wide range of controversial advertising, including advertisements relating to abortion. Id. The only issue on appeal was whether these factual findings were clearly erroneous. Id. at 1228-29. Similarly, in Grace Bible, following a bench trial the district court found that the defendant school district created a designated public forum. 941 F.2d at 46^47. The First Circuit affirmed on the ground that the school district’s written policies indicated an intent to open its facilities indiscriminately to the community, rather than to restrict access to school-related uses. Id. at 47-48. Further, the district’s practice of allowing many organizations to use its facilities reflected this intent. Id.
Both Planned Parenthood and Grace Bible involved an extensive record of a past practice permitting the type of speech the government later sought to ban, which contradicted the government’s claim that it limited access to its property. Both involved either an express policy demonstrating an intent to open the space or a post hoc policy contrived to conceal censorship. And in both cases a trial was conducted, the record fully developed, and factual determinations made at trial were controlling. Here, in contrast, the record of Pasco’s past practice is barely existent and Pasco’s express policy genuinely does not show an intent to create a designated public forum. And perhaps most importantly, neither party has had the opportunity to develop a full record at trial and key facts cannot properly be determined on this appeal.
United Food & Commercial Workers Union, Local 1099 v. Southwest Ohio Reg’l Transit Auth., 163 F.3d 341 (6th Cir.1998), similarly does not support granting summary judgment for Hopper and Rupp. There, the Sixth Circuit relied on Cornelius and reasoned that “[w]e ... must closely examine whether in practice [the transit authority] has consistently enforced its written policy in order to satisfy ourselves that [its] stated policy represents its actual policy.” Id. at 363. Because the union did not identify any advertisements accepted by the transit authority that violated the written policy, the court noted “we have no reason based on the record at this time to believe [the transit authority] applies its written policy on an ad hoc basis.” Id. However, the court observed “[s]hould [the union] introduce evidence at trial demonstrating that [the transit authority] has not consistently followed its written policy, but instead has maintained an ad hoc.policy where the acceptability of an advertisement depends on the whim of the decision-maker, this would strongly suggest that [the transit authority] has created a public forum.” Id. at n. 6. As previously discussed, here there is no evidence demonstrating Pasco’s inconsistent enforcement of its written policy because the experimental exhibitions generated little to no controversy with which to test its stated policy. United Food does not support the majority’s use of a scant record to grant partial summary judgment to Hopper and Rupp on the public forum issue; instead, United Food supports remanding this issue for trial.
Certainly inconsistent enforcement of a policy may be evidence of governmental intent to create a designated public forum. And a jury instruction on the standards for determining intent might properly encourage consideration of this factor. If enforcement history is considered at a trial, and a trier of fact finds the history supports a determination of intent to create a designated public forum, this evidence may be properly credited by an appellate court. Notwithstanding, these principles do not justify the majority’s approach of making an ambiguous enforcement practice — not considered by a trier of fact — the sole determining factor.
VI
The Supreme Court instructs that governmental intent is the key to forum anal*1091ysis because the government does not convert its property into a public forum absent an intent to do so. Cornelius, 473 U.S. at 802, 105 S.Ct. 3439; see also General Media Communications, Inc. v. Cohen, 131 F.3d 273, 279 (2d Cir.1997) (intent is “touchstone” of forum analysis); Sentinel Communications Co. v. Watts, 936 F.2d 1189, 1202 (11th Cir.1991) (same); Gregoire v. Centennial Sch. Dist., 907 F.2d 1366, 1386 (3d Cir.1990) (same); Stewart v. District of Columbia Armory Bd., 863 F.2d 1013, 1016 (D.C.Cir.1988) (same). It necessarily follows that whether a government entity intended to open a forum is an “inherently factual inquiry that should not be resolved without due attention to an underlying record.” Air Line Pilots Ass’n, Int’l v. Dept. of Aviation of the City of Chicago, 45 F.3d 1144, 1152 (7th Cir.1995); Stewart, 863 F.2d at 1018. Where important constitutional issues turn on the outcome of the intent inquiry, summary judgment is likely to be inappropriate. Searcey v. Crim, 815 F.2d 1389, 1392-93 (11th Cir.1987) (summary judgment on issue of government intent to create limited public forum inappropriate); May v. Evansville-Vanderburgh Sch. Corp., 787 F.2d 1105, 1115 (7th Cir.1986) (same); see also, e.g., Perry v. Sindermann, 408 U.S. 593, 598, 92 S.Ct. 2694, 33 L.Ed.2d 570 (1972) (where government’s motive was element of § 1983 claim, summary judgment was improper).
Summary judgment almost certainly is inappropriate in a case such as this where intent is at issue. “In many constitutional and civil rights cases, a necessary element of the claim for relief presents an inquiry into the state of mind of one or more of the parties.... [Cjlaims requiring a determination regarding intentions or motives are particularly unsuitable for summary adjudication.” 10B Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and Procedure, Civil 2d § 2732.2 (1998). In fact, in a related context involving a plaintiffs claim that he had been fired for exercising his First Amendment rights, this court noted that “a fair resolution” of the issues of the employer’s motive and intent — essential elements of the claim — “requires a full trial on the merits,” and that those questions are “plainly ... reserved to the trier of fact.” Peacock v. Duval, 694 F.2d 644, 646 (9th Cir.1982) (internal citations and quotations omitted).
Such is the case here. At best the record is incomplete and supports contrary inferences regarding Pasco’s intent, thus precluding summary judgment for either party. A grant of partial summary judgment to Hopper and Rupp is inconsistent with the Supreme Court’s teaching in Liberty Lobby, which bars a court from making factual determinations when the record supports conflicting inferences. “[A]t the summary judgment stage the judge’s function is not himself [or herself] to weigh the evidence and determine the truth of the matter but to determine whether there is a genuine issue for trial.” Liberty Lobby, 477 U.S. at 242-43, 106 S.Ct. 2505.
A failure correctly to apply the Cornelius factors and the proper summary judgment standard to the key question of Pasco’s intent is error.2 The majority *1092in effect creates a new per se rule that the government’s failure systematically to enforce an exclusionary policy — without regard to whether there was even an appropriate occasion to enforce such policy — conclusively establishes that it has created a designated public forum. This new rule conflicts with the Supreme Court’s admonishment in Perry and Cornelius that intent is the key to the public forum inquiry. It also contradicts the Supreme Court’s instruction that a court must look to three factors to determine intent: the government’s policy, the government’s conduct, and the nature of the space. If reasonable inferences from these three factors support a jury finding that, the government did not intend to open up a space for unrestricted expression, partial summary judgment for Hopper and Rupp is inappropriate. See Liberty Lobby, 477 U.S. at 248, 106 S.Ct. 2505. As demonstrated above, each of these factors support such a finding here. Pasco’s intent is an open question and the issue should be submitted to a jury after a trial that fully develops the facts.
Moreover, the majority’s new rule undermines important First Amendment interests. As the Supreme Court recently explained:
The Cornelius distinction between general and selective access furthers First Amendment interests. By recognizing the distinction, we encourage the government to open its property to some expressive activity in cases where, if faced with an all-or-nothing choice, it might not open the property at all. That this distinction turns on governmental intent does not render it unpro-tective of speech. Rather, it reflects the reality that, with the exception of traditional public fora, the government retains the choice of whether to designate its property as a forum for specified classes of speakers.
Arkansas Educ. Television Comm’n v. Forbes, 523 U.S. 666, 680, 118 S.Ct. 1633, 140 L.Ed.2d 875 (1998). Faced with a rule that a government’s ambiguous conduct alone can support a § 1983 action for damages for First Amendment violations, governments may see their choice as being the “all-or-nothing” choice described in Forbes and likely will then refuse to open the property at all. This result harms artists, governments, and the public alike.
It is not necessary to create a new rule that conflicts with Supreme Court precedent when we can.instead remand for a trial and abide that precedent. For the foregoing reasons, I respectfully dissent.
. The sparse record does not contain any photograph, drawing or replica of this piece. The fact that the record is not sufficiently developed to determine the nature of any pri- or "controversy” further supports a remand for a trial on the merits.
. The majority incorrectly assumes that a forum was created, and then argues that there is no genuine issue of material fact concerning Pasco's allegedly inconsistent screening policy. Majority Opinion at 1079 n. 10. With respect, the majority misses the point. There is a genuine issue of material fact on Pasco’s intent. This requires trial to determine if a limited public forum was ever created. As United States Supreme Court and other authority make clear, intent, and not enforcement policy, is the touchstone of forum analysis. The majority’s analysis infers that the “structure of Pasco's arts program suggests an intent to permit unrestricted expression.” Majority Opinion at 1079. It is a stretch for the majority to conclude that the structure of the program proves that Pasco’s intent is unmistakably different from what Pasco stated in its letters. This requires the majority to draw several inferences adverse to Pasco. Perhaps this can be argued, but when all inferences properly are given Pasco on summary judgment, there can be no doubt that Pasco's letters and statements show that there is a genuine issue of material fact regarding its intent. It is puzzling that the *1092majority mistakenly assumes the role of fact finder on the issue of intent, which courts almost universally view as a factual issue. The majority’s reliance on snatches of language from cases that went to trial hardly answers this objection. And under the Supreme Court's controlling precedents, a lack of consistent enforcement policy, properly viewed, is onfy one factor that the trier of fact may consider in assessing intent.