McCraw v. City of Oklahoma City

FILED
United States Court of Appeals
PUBLISH Tenth Circuit

UNITED STATES COURT OF APPEALS August 31, 2020

Christopher M. Wolpert
FOR THE TENTH CIRCUIT Clerk of Court
_________________________________

CALVIN McCRAW; G. WAYNE
MARSHALL; MARK FAULK; TRISTA
WILSON; NEAL SCHINDLER;
OKLAHOMA LIBERTARIAN PARTY;
RED DIRT REPORT,

Plaintiffs - Appellants,
No. 19-6008
v. (D.C. No. 5:16-CV-00352-HE)
(W.D. Okla.)
CITY OF OKLAHOMA CITY, an
Oklahoma municipal corporation;
WILLIAM CITTY, in his official capacity
as Chief of the Oklahoma City Police
Department,

Defendants - Appellees.
_________________________________

Appeal from the United States District Court
for the Western District of Oklahoma
(D.C. No. 5:16-CV-00352-HE)
_________________________________

Joseph Thai, Oklahoma City, Oklahoma (Erwin Chemerinsky, Berkeley, California;
Ryan Kiesel, Brady Henderson, Megan Lambert, ACLU of Oklahoma Foundation,
Oklahoma City, Oklahoma; Greg Beben, Legal Aid Services of Oklahoma, Inc.,
Oklahoma City, Oklahoma; and Micheal Salem, Salem Law Offices, Norman, Oklahoma,
with him on the briefs)

Amanda Carpenter, Oklahoma City, Oklahoma (Kenneth Jordan, Municipal Counselor;
Catherine Campbell, Phillips Murrah P.C., Oklahoma City, Oklahoma, with her on the
briefs)
_________________________________

Before LUCERO, EBEL, and HARTZ, Circuit Judges.
 _________________________________

LUCERO, Circuit Judge.
_________________________________

This case concerns the First Amendment rights of citizens in the public square—

specifically on medians in public roads. Oklahoma City Ordinance 25,777 prohibits

standing, sitting, or remaining for most purposes on certain medians. Okla. City, Okla.,

Code ch. 32, art. XIII, § 32-458. Plaintiffs are Oklahoma City residents, a minority

political party in Oklahoma, and an independent news organization. They use medians to

panhandle, engage in protests or other expressive activity, mount political campaigns,

cover the news, or have personal conversations. After they were no longer able to engage

in such activity due to the ordinance, plaintiffs sued Oklahoma City and its chief of

police, William Citty, (together, “the City”) alleging violations of their First and

Fourteenth Amendment rights. The district court dismissed plaintiff Trista Wilson’s First

Amendment claim; granted summary judgment favoring the City on plaintiffs’ due

process vagueness claims; and, following a bench trial, entered judgment against

plaintiffs on all other claims. Exercising jurisdiction under 28 U.S.C. § 1291, we reverse

the court’s entry of judgment favoring the City on plaintiffs’ First Amendment claims;

we reverse the dismissal of Wilson’s First Amendment claim; and we affirm on all other

claims.

2
 I

A

As in many other cities, the medians in Oklahoma City are varied and diverse.

They range in length and width: some span an entire city block, others stretch down

several car lengths at intersections. Many contain trails, sidewalks, benches, art, large

signs, landscaping, or wide-open spaces. One even contains an operating fire station.

In 2015, before the enactment of the ordinance at issue in this case, Oklahoma

City’s municipal code prohibited pedestrians from soliciting in roadways without a

permit. Pedestrians could apply for a permit to walk from a median or sidewalk into the

road to solicit, so long as they did not impede traffic and remained in the road only when

cars were stopped at traffic lights. Under this system, political campaigns, panhandlers,

and community fundraisers—including firefighters engaged in their annual Fill the Boot

campaign for the Muscular Dystrophy Association—engaged in various activities on

medians.

In December 2015, the Oklahoma City Council further restricted pedestrian

activity on medians. Ordinance 25,283 (“Original Ordinance”) prohibited standing,

sitting, or staying on any portion of a median either less than thirty feet wide or located

less than two hundred feet from an intersection. Okla. City, Okla., Ordinance 25,283

(Dec. 9, 2015). The ordinance eliminated the prior permit exception for soliciting in

roadways, but it allowed access to medians for certain specified purposes, including

access by public employees and for emergency uses.

Before its passage, city officials and others pointed to panhandlers as the impetus

3
for the Original Ordinance. The ordinance’s author cited complaints she had received

from citizens and businesses regarding panhandling and repeatedly described the Original

Ordinance as addressing panhandling. She also stated that her goal was “to help try to

find a way to redirect the dollars that are going out windows” back to agencies that

provide food and shelter. Before the Original Ordinance was introduced, the City’s

municipal counselor informed City officials that the author was working on an ordinance

to ban panhandling and soliciting on medians; he recognized the potential

unconstitutionality of such a law. Later, at a public hearing regarding the ordinance two

weeks after its introduction, the municipal counselor’s office contended it should be

viewed as addressing public safety and was “not necessarily about panhandlers.” An

assistant city attorney explained that people on medians, regardless of their activity, were

in danger and that panhandling would still be permitted on sidewalks and on the side of

the road.

At the third and final council meeting regarding the ordinance, Chief Citty gave a

presentation. The presentation was originally titled “Panhandler Presentation,” but by the

time Chief Citty gave it, its name had been changed to “Median Safety Presentation.” It

demonstrated that between January 10, 2010, and September 29, 2015, there were 39,833

collisions citywide. This included 16,358 accidents resulting in injuries or fatalities, of

which 76% occurred near intersections. However, it showed no pedestrian-related

accidents on medians.

Chief Citty showed slides and photographs of damaged medians and accidents in

which vehicles entered or crossed onto the median, but he offered no specific evidence of

4
accidents involving pedestrians on medians. He stated that some of the accidents

involved pedestrians but that he did not know the precise number, adding that the number

would not be “very high.” He also stated that much of the damage to medians is caused

by unreported accidents.

According to Chief Citty, it had been the police department’s position for several

years that pedestrian activity on medians was dangerous because of pedestrians’ exposure

to traffic moving in different directions. The City Council disagreed about whether these

safety concerns justified the Original Ordinance, but it passed by a seven-to-two vote.

Plaintiffs sued, claiming that the Original Ordinance violated their First and

Fourteenth Amendment rights. The same month, the City Council amended the city’s

Aggressive Panhandling Ordinance, Okla. City, Okla., Code ch. 30, art. XV, div. 2, § 30-

428 et seq., to expand existing panhandling-free zones and to create new ones. The

ordinance included a ban on panhandling within fifty feet of any mass transportation stop.

In 2017, after the district court denied the City’s motion for summary judgment

without prejudice, the City Council revised the ordinance. Ordinance 25,777 (“Revised

Ordinance”), amended the section of the City’s municipal code entitled, “Standing,

sitting, or staying on streets, highways, or certain medians.” § 32-458. It outlawed

pedestrian presence on medians in all streets with a speed limit of forty miles per hour or

more, but exempted government employees and people on the median to cross the street,

perform “legally authorized work,” or “respond[] to any emergency situation.” Id.

The Revised Ordinance included findings, with citations to a Centers for Disease

Control and Prevention (“CDC”) report listing higher vehicle speeds among risk factors

5
for auto-pedestrian crashes and a Federal Highway Administration publication with

general statistics regarding the likelihood of fatality for a pedestrian struck by a moving

vehicle. Neither report addressed medians in particular. The findings also noted that in

2015, pedestrian deaths accounted for 15% of all traffic fatalities; 90% of the pedestrian

deaths were from crashes involving a single vehicle; and 19% of pedestrian deaths were

from crashes involving hit-and-run drivers. Without citation, the Revised Ordinance also

concluded that people sitting, standing, or remaining on medians “create additional

distractions for the operators of motor vehicles using such streets and highways.”

Before the Revised Ordinance was passed, an assistant city attorney told the City

Council that the City had conducted further research to determine the highest risk factor

for pedestrians who remained on medians for longer than necessary to cross the street.

Based on National Highway Traffic Safety Administration (“NHTSA”) statistics, the City

determined that vehicles traveling at high speeds caused the most risk. According to the

NHTSA, the pedestrian fatality rate in accidents with vehicles traveling at forty miles per

hour is 85%, compared to 45% for vehicles traveling at thirty miles per hour and 5% for

vehicles traveling at twenty miles per hour. The CDC similarly reported that vehicle

speeds increased both the likelihood of pedestrians being struck by a motor vehicle and

the severity of injury.

The City solicited the opinion of Master Sergeant Brian Fowler, a fatality

investigator for the Oklahoma City Police Department, who observed that in 2015 the

Insurance Institute of Highway Safety reported that 54% of pedestrian deaths occurred on

large, arterial roadways. He testified that median curbs offer “very minimal” protection

6
when vehicles are traveling at higher speeds, that driver distractions cause pedestrians to

be at higher risk, and that pedestrians on medians cause distractions to drivers. Fowler

and Chief Citty set forth a descending hierarchy of the riskiest places for pedestrians to

be: (1) traffic lanes, (2) medians, and (3) roadsides or sidewalks. Fowler also testified

that he believed that the longer a pedestrian remained on a median, the greater the risk.

He was unable, however, to quantify the risk or provide any support in safety literature

for his opinion.

In response to a request from plaintiffs for all accident reports involving medians

or pedestrians, the City produced 504 reports dating from 2012 to 2017. No report

involved a pedestrian struck on any median. Out of 39,833 accidents reported from 2010

to 2015, none involved pedestrians on medians. Further, at trial, the City could not

identify anyone injured on a median in Oklahoma City or any accident caused by

pedestrian activity on a median. Moreover, Fowler admitted that he did not have any

research or data to support his conclusion that pedestrians remaining on medians in

Oklahoma City are exposed to more risk.

The Revised Ordinance prohibits pedestrians from being on approximately four

hundred medians across Oklahoma City.1 The City asserts that there are at least 103

medians unaffected by the ordinance because they are on roads with speed limits lower

1
The parties stipulated that a legal intern personally verified the physical
existence and location of each of the 406 medians on the list plaintiffs introduced at
trial.

7
than forty miles per hour.2 Plaintiffs respond that at least 27 of these 103 medians are

unavailable to panhandlers and solicitors under the Aggressive Panhandling Ordinance

because they are within fifty feet of a bus stop.3

B

Plaintiffs are individuals and organizations whose use of the medians has been

barred by the Revised Ordinance. Mark Faulk, the chair of the Oklahoma County

Democratic Party and a former state legislative candidate, has held campaign signs and

taken part in political protests on affected medians. The Oklahoma Libertarian Party has

used medians to garner signatures for petitions and to spread its message. Red Dirt

Report is a central Oklahoma online daily periodical that uses medians to cover breaking

news. Calvin McCraw and G. Wayne Marshall panhandle on medians to pay for food,

shelter, medicine, and other necessities. McCraw has also stood on medians to distribute

The Curbside Chronicle, a street newspaper.

2
The parties dispute the application of the Revised Ordinance to the medians
along Lincoln Boulevard, which is also State Highway 0. The district court
concluded that because the Lincoln Boulevard medians were under the direct control
of the State, not the City, they were unaffected by the ordinance. Plaintiffs argue that
the City maintains control of the Lincoln Boulevard medians because it has municipal
jurisdiction to “[r]egulate and control the use of streets, roads and other public ways
within the limits of the municipality,” Okla. Stat. tit. 11, art. XXXVI, § 36-101(1),
further evidenced by the fact that it grants a right-of-way along Lincoln Boulevard
and its medians for the annual Oklahoma City Memorial Marathon. Because it is not
dispositive of our analysis, we note the dispute but do not decide whether the Revised
Ordinance applies to Lincoln Boulevard’s medians.
3
Plaintiffs do not address whether any of the remaining 76 unaffected medians
would be unavailable to panhandlers under other provisions of the Aggressive
Panhandling Ordinance.
8
 Trista Wilson and Neal Schindler use the medians when they run. Wilson stops on

medians to converse with her jogging companions. Schindler celebrates life and honors

those who died in the Oklahoma City bombing by running on medians while training for

the Oklahoma City Memorial Marathon.

After passage of the Revised Ordinance, plaintiffs added claims alleging the new

law violated their constitutional rights. The district court dismissed their claims

challenging the Original Ordinance as moot. It also dismissed Wilson’s First

Amendment claim, concluding that she had not alleged that the Revised Ordinance

impinged on protected expression. On summary judgment, the court rejected plaintiffs’

Fourteenth Amendment vagueness challenge, holding that the Revised Ordinance’s

definition of “emergency” did not render the ordinance unconstitutionally vague.

Following a bench trial with live witnesses and deposition designations, the district court

rejected the remaining claims. It concluded that “a substantial number of medians subject

to the Ordinance qualify as traditional public fora,” but the Revised Ordinance was a

valid time, place, and manner restriction under the First Amendment because it was

narrowly tailored and provided ample alternative channels of communication. It also

rejected plaintiffs’ Fourteenth Amendment due process claim, concluding that there was

no fundamental right to intrastate freedom of movement and that the Revised Ordinance

passed rational basis review. Plaintiffs appealed.

II

“In a First Amendment case, we have an obligation to make an independent

examination of the whole record in order to make sure that the judgment does not

9
constitute a forbidden intrusion on the field of free expression.” Citizens for Peace in

Space v. City of Colo. Springs, 477 F.3d 1212, 1219 (10th Cir. 2007) (quotation

omitted).4 We therefore “review the district court’s findings of fact and its conclusions of

law de novo” and “without deference to the trial court.” Id. (quotation omitted). We also

review de novo the court’s legal conclusions regarding plaintiffs’ due process claims.

See McClure v. Ind. Sch. Dist. No. 16, 228 F.3d 1205, 1212 (10th Cir. 2000).

A

Because it concluded the Revised Ordinance was a constitutionally permitted time,

place, and manner restriction, the district court entered judgment for the City on

plaintiffs’ First Amendment claims. The First Amendment, applicable to the States under

the Due Process Clause of the Fourteenth Amendment, provides that “Congress shall

make no law . . . abridging the freedom of speech, . . . or the right of the people peaceably

to assemble, and to petition the Government for a redress of grievances.” U.S. Const.

amend. I; see also iMatter Utah v. Njord, 774 F.3d 1258, 1263 (10th Cir. 2014). “To

demonstrate a violation of their First Amendment rights, Plaintiffs must first establish

that their activities are protected by the First Amendment. If so, a court must identify

whether the challenged restrictions affect a public or nonpublic forum; that determination

dictates the extent to which the government can restrict First Amendment activities

4
Pursuant to our obligation, we have reviewed the entire record and base our
analysis and conclusions on our review. However, we note that this obligation does
not excuse the parties from their requirement under Federal Rule of Appellate
Procedure 28 to cite to the “parts of the record on which [they] rel[y] . . . .” Fed. R.
App. P. 28(a)(8)(A), (b).
10
within the forum. Finally, courts must determine whether the proffered justifications for

prohibiting speech in the forum satisfy the requisite standard of review.” Verlo v.

Martinez, 820 F.3d 1113, 1128 (10th Cir. 2016) (citations omitted).

B

We agree with the district court that all plaintiffs whose claims proceeded to trial

engaged in protected speech. See McCullen v. Coakley, 573 U.S. 464, 488-89 (2014)

(leafletting and communicating ideas in normal conversation protected First Amendment

activity); Morse v. Frederick, 551 U.S. 393, 403 (2007) (political speech “at the core of”

the First Amendment); Riley v. Nat’l Fed’n of the Blind of N.C., Inc., 487 U.S. 781, 789

(1988) (solicitation of charitable contributions is protected speech); Edenfield v. Fane,

507 U.S. 761, 767 (1993) (commercial speech protected under First Amendment);

Heffron v. Int’l Soc’y for Krishna Consciousness, Inc., 452 U.S. 640, 647 (1981) (First

Amendment protection of newspaper distribution not lost merely because the paper being

distributed is sold, rather than given away); Branzburg v. Hayes, 408 U.S. 665, 681-82

(1972) (news gathering protected by First Amendment); Lovell v. City of Griffin, 303

U.S. 444, 452 (1938) (First Amendment protects distribution and publication of

newspapers); Speet v. Schuette, 726 F.3d 867, 878 (6th Cir. 2013) (holding “that

begging, or the soliciting of alms, is a form of solicitation that the First Amendment

protects”); Smith v. City of Fort Lauderdale, 177 F.3d 954, 956 (11th Cir. 1999) (“Like

other charitable solicitation, begging is speech entitled to First Amendment protection.”);

Loper v. N.Y.C. Police Dep’t, 999 F.2d 699, 704 (2d Cir. 1993) (“Begging frequently is

accompanied by speech indicating the need for food, shelter, clothing, medical care or

11
transportation. Even without particularized speech, however, the presence of an unkempt

and disheveled person holding out his or her hand or a cup to receive a donation itself

conveys a message of need for support and assistance.”); Clark v. Comm’y for Creative

Non-Violence, 468 U.S. 288, 304 (1984) (“In a long line of cases, this Court has afforded

First Amendment protection to expressive conduct that qualifies as symbolic speech.”).

We turn to Wilson, whose claim the district court dismissed as lying outside First

Amendment protection. Wilson alleged that she was an “avid jogger” who ran

throughout Oklahoma City, including on medians covered by the Original and Revised

Ordinances. Because it determined that Wilson had not alleged facts demonstrating that

her jogging was expressive activity, the district court concluded she had not alleged that

she was engaged in any protected speech as contemplated by the First Amendment. We

agree that Wilson has not alleged facts demonstrating that her jogging was expressive

activity.

However, Wilson also alleged that she engaged in communicative activities while

out on a run—allegations the district court appears to have disregarded. Specifically, she

described stopping on medians to have personal conversations with her jogging

companions. Even though these conversations may not amount to grand rhetoric or

political soapbox oratory, they are nonetheless protected by the First Amendment. “Most

of what we say to one another lacks religious, political, scientific, educational,

journalistic, historical, or artistic value (let alone serious value), but it is still sheltered

from government regulation.” United States v. Stevens, 559 U.S. 460, 479 (2010)

(quotation and alteration omitted); see also Capitol Square Review & Advisory Bd. v.

12
Pinette, 515 U.S. 753, 760 (1995) (private expression “fully protected under the Free

Speech Clause”); Connick v. Myers, 461 U.S. 138, 147 (1983) (speech on private matters

not within “one of the narrow and well-defined classes of expression which carries so

little social value, such as obscenity, that the State can prohibit and punish such

expression by all persons in its jurisdiction”); Givhan v. W. Line Consol. Sch. Dist., 439

U.S. 410, 413 (1979) (“We are unable to agree that private expression of one’s views is

beyond constitutional protection . . . .”); Eichenlaub v. Twp. of Indiana, 385 F.3d 274,

284 (3d Cir. 2004) (“[W]hile speech on topics of public concern may stand on the

‘highest rung’ on the ladder of the First Amendment, private speech (unless obscene or

fighting words or the like) is still protected on the First Amendment ladder.”).

The City’s argument that Wilson’s communicative activities are “incidental” to

her jogging and do not merit First Amendment protection is merely an attempt to

minimize Wilson’s protected speech to such a degree that it is extinguished. Wilson’s

communicative activities are distinct from her jogging and therefore subject to the normal

constitutional inquiry. Her speech does not lose protection either because she is

simultaneously engaged in non-expressive activity or because the City has deemed

Wilson’s speech valueless. After all, it would be a boring day if runners would be denied

the lingua franca of athletes in training. Accordingly, we reverse the dismissal of

Wilson’s First Amendment claim.5

5
Although the district court dismissed Wilson’s claim prior to summary
judgment and trial, Wilson submitted an affidavit and designated deposition in
support of the remaining plaintiffs’ case. She thus had a full and fair opportunity to

13
 C

Under the First Amendment, the extent to which the government may regulate

access to public property depends on the category of forum into which the property falls:

“the traditional public forum, the designated public forum, and the nonpublic forum.”

Verlo, 820 F.3d at 1129. Traditional public fora are those that “by long tradition or by

government fiat have been devoted to assembly and debate . . . .” Perry Educ. Ass’n v.

Perry Local Educators’ Ass’n, 460 U.S. 37, 45 (1983); McCullen, 573 U.S. at 476

(traditional public fora are areas that “have been used for purposes of assembly,

communicating thoughts between citizens, and discussing public questions”). In contrast,

designated public fora “are not generally open to the public for First Amendment activity

and are created by purposeful governmental action to allow speech activity.” Evans v.

Sandy City, 944 F.3d 847, 853 (10th Cir. 2019) (quotation omitted). All other fora are

nonpublic. Id.

To determine whether a particular property is a traditional public forum, we look

at “the objective characteristics of the property, such as whether, by long tradition or by

government fiat, the property has been devoted to assembly and debate.” Ark. Educ.

Television Comm’n v. Forbes, 523 U.S. 666, 677 (1998) (quotation omitted). We

conclude that Oklahoma City’s medians fall within this category. Objectively, medians

share fundamental characteristics with public streets, sidewalks, and parks, which are

quintessential public fora. See Cornelius v. NAACP Legal Def. & Educ. Fund, Inc., 473

present her claim and we do not have any concerns in adjudging her claim along with
those of the remaining plaintiffs.
14
U.S. 788, 802 (1985); Hague v. Comm. for Indus. Org., 307 U.S. 496, 515 (1939)

(Roberts, J., concurring) (“Wherever the title of streets and parks may rest, they have

immemorially been held in trust for the use of the public and, time out of mind, have

been used for purposes of assembly, communicating thoughts between citizens, and

discussing public questions . . . .”); United States v. Grace, 461 U.S. 171, 179 (1983)

(“Sidewalks, of course, are among those areas of public property that traditionally have

been held open to the public for expressive activities and are clearly within those areas of

public property that may be considered, generally without further inquiry, to be public

forum property.”). Many medians in Oklahoma City, particularly those at intersections,

contain sidewalks. “The typical traditional public forum is property which has the

physical characteristics of a public thoroughfare, . . . [and] which has the objective use

and purpose of open public access or some other objective use and purpose inherently

compatible with expressive conduct.” Warren v. Fairfax Cty., 196 F.3d 186, 191 (4th

Cir. 1999) (en banc) (addendum to majority opinion) (citation omitted). Medians share

with streets and sidewalks the physical characteristics of public thoroughfares.

Moreover, medians are sandwiched by the uncontested public fora of streets and

sidewalks. In Grace, the Supreme Court found persuasive the lack of a demarcation

between areas traditionally perceived as traditional public fora and those the government

sought to treat as non-public fora. 461 U.S. at 179-80 (“There is no separation, no fence,

and no indication whatever to persons stepping from the street to the curb and sidewalks

that serve as the perimeter of the Court grounds that they have entered some special type

of enclave.”). We similarly decline to carve out a distinction between public streets—

15
“the archetype of a traditional public forum,” Frisby v. Schultz, 487 U.S. 474, 480-81

(1988)—and the medians that lie in the middle of and are surrounded by those streets. If

the road that abuts a median on both sides is a public forum, the median itself also

qualifies.

The City highlights what it characterizes as differences between these

quintessential public fora and medians, citing the speed and volume of passing cars,

among other characteristics.6 These assertions may support the argument that a time,

place, and manner restriction is constitutional. See, e.g., Cox v. Louisiana, 379 U.S. 536,

554, (1965) (rejecting contention that someone could, “contrary to traffic regulations,

insist upon a street meeting in the middle of Times Square at the rush hour as a form of

freedom of speech or assembly”). But ample precedent holds that these characteristics do

not deprive public streets of their status as public fora. See Frisby, 487 U.S. at 481

(noting that the “character of those streets may well inform the application of the relevant

test, but it does not lead to a different test”); Warren, 196 F.3d at 195 (addendum to

majority opinion) (“The test is not whether the property was designed for expressive

activity, but whether the objective uses and purposes of the property are compatible with

the wide measure of expressive conduct characterizing public fora.”). Moreover, “[n]o

particularized inquiry into the precise nature of a specific street is necessary; all public

6
The City also argues that its medians lack benches, sidewalks, trails, or other
amenities that would render them similar to parks. Although the record shows many
medians with those amenities, albeit on streets with lower speed limits, the City’s
distinction does not account for the similarities between medians and the streets in
which they lie.
16
streets are held in the public trust and are properly considered traditional public fora.”

Frisby, 487 U.S. at 481.

Because the proximity, speed, and volume of passing cars does not deprive streets

of their status as public fora, they similarly fail to strip medians of that status—after all,

streets are also not intended to have people in them most of the time. We reiterate: if the

street in which those cars are moving is a traditional public forum, so too is the median in

the center of that street. See Satawa v. Macomb Cty. Rd. Comm’n, 689 F.3d 506, 520

(6th Cir. 2012) (holding that given public use for expressive purposes, even a median “in

the middle of a busy eight-lane road, with a fifty mile-per-hour speed limit . . . [o]n

balance, . . . [was] a traditional public forum”).

Perhaps more significantly, the record demonstrates a “long tradition” of

expressive activity occurring on Oklahoma City’s medians. The record is replete with

examples of speech occurring on medians, from firefighter charity drives to protests to

political campaign signs. Testimony demonstrated that these activities have occurred for

a long time, with plaintiffs stating that the firefighters used medians “[a]s long as I can

remember;” that political signs were erected on medians for “probably 40 years” and had

“gone on forever;” and that people had stood on medians on Election Day since the early

1970s. The City attempts to minimize this tradition of expressive activity by

distinguishing between use for expressive activity “for years,” and use “for time out of

mind.” We decline to specify the precise number of years it takes to create a “long

tradition” of expressive activity. Suffice it to say that testimony that such expression has

occurred for as long as witnesses can remember is enough evidence of tradition.

17
 The City responds that it neither intended to create a public forum for expression,

nor invited the public to use the medians for expression. Although these considerations

are not irrelevant, we note that the question put to us is “whether, by long tradition or by

government fiat, the property has been devoted to assembly and debate.” Ark. Educ.

Television Comm’n, 523 U.S. at 677 (quotation omitted and emphasis added). “It is only

with respect to designated fora that the Supreme Court’s forum analysis has focused on

whether there has been purposeful government action creating a forum in a place not

traditionally open to assembly and debate.” First Unitarian Church of Salt Lake City v.

Salt Lake City Corp., 308 F.3d 1114, 1124 (10th Cir. 2002) (quotation omitted). For

property that has traditionally been open to the public, a lack of government fiat is not

dispositive. Rather, “objective characteristics are more important and can override

express government intent to limit speech.” Id. at 1125.

Further, the Oklahoma City Municipal Code itself defines streets to include

medians. Okla. City, Okla. Code ch. 32, art. I §§ 32-1(29), (59). And under Oklahoma

law, “[t]he title to streets, roads and public ways within the limits of a municipality which

have been dedicated and accepted by the municipal governing body is held by the

municipality in trust for public use and enjoyment.” Okla. Stat. tit. 11, art. XXXVI, § 36-

101. The City cannot now remove medians’ public fora status by fiat. See First

Unitarian Church, 308 F.3d at 1124 (“The government cannot simply declare the First

Amendment status of property regardless of its nature and its public use.”). Just as the

government may not “transform the character of the property by the expedient of

including it within the statutory definition of what might be considered a non-public

18
forum parcel of property,” it similarly may not ignore an existing statutory provision

defining medians as streets, which are considered a public forum. See Grace, 461 U.S. at

180 (government’s “ipse dixit” does not determine the First Amendment status of

property). The evidence overwhelmingly demonstrates that in Oklahoma City, medians

have traditionally been used for expressive activity.

We hold that Oklahoma City’s medians are traditional public fora.7 See Reynolds,

779 F.3d at 225 (“There is . . . no question that public streets and medians qualify as

traditional public for[a].”); see also Cutting v. City of Portland, 802 F.3d 79, 83 (1st Cir.

2015) (holding that Portland’s medians were traditional public fora “on the understanding

that . . . the people of Portland have used median strips for expressive purposes in much

the same way that they have used parks and sidewalks”); Ater v. Armstrong, 961 F.2d

1224, 1226-27 (6th Cir. 1992) (in a challenge to a restriction from distributing literature

on medians and streets, analyzing medians and streets together to hold that the county’s

streets were traditional public fora).

D

Having concluded that medians are public fora, we analyze the Revised

Ordinance’s validity under the time, place, and manner framework. “It is well-settled

that even in a public forum the government may impose reasonable restrictions on the

7
We declined to answer this question in Evans because it was not
determinative. Sandy City’s ordinance was a valid time, place, or manner regulation
and therefore met the requirements for a law restricting speech even in a public
forum. 944 F.3d at 853-54. In this case, however, we squarely address the question
because the character of the forum affects the remainder of our analysis.
19
time, place, and manner of protected speech, provided the restrictions are justified

without reference to the content of the regulated speech, that they are narrowly tailored to

serve a significant governmental interest, and that they leave open ample alternative

channels for communication of information.” Evans, 944 F.3d at 854 (quotation

omitted).

Plaintiffs argue that we should apply strict scrutiny because the Revised Ordinance

discriminates based on content. We need not reach this argument. As discussed below,

we ultimately conclude the Revised Ordinance fails even intermediate scrutiny. Because

it would necessarily also fail strict scrutiny, we assume for the purposes of our analysis

that the Revised Ordinance is content-neutral.8 See Reed, 135 S. Ct. at 2229-31

(describing the differences in standards applied to content-based and content-neutral

regulations).

8
Our independent examination of the record reveals troubling evidence of
animus against panhandlers in the passage of the Original and Revised Ordinances.
But because we conclude that the City’s Revised Ordinance fails even intermediate
scrutiny, we are not required to delve into whether its ostensible content-neutrality is
instead camouflage for the City’s desire to sacrifice speech in order to ban unpopular
panhandling. See McCullen, 573 U.S. at 486 (“Where certain speech is associated
with particular problems, silencing the speech is sometimes the path of least
resistance. But by demanding a close fit between ends and means, the tailoring
requirement prevents the government from too readily sacrificing speech for
efficiency.” (quotation and alteration omitted)); Reed v. Town of Gilbert, 135 S. Ct.
2218, 2227 (2015) (facially content-neutral restrictions “will be considered content-
based regulations of speech” if they “were adopted by the government because of
disagreement with the message the speech conveys” (quotation and alteration
omitted)); see also Elena Kagan, Private Speech, Public Purpose: The Role of
Governmental Motive in First Amendment Doctrine, 63 U. Chi. L. Rev. 413, 451-52
(1996).
20
 1

“For a content-neutral time, place, or manner regulation to be narrowly tailored, it

must not burden substantially more speech than is necessary to further the government’s

legitimate interests.” McCullen, 573 U.S. at 486 (quotation omitted). Although the City

is not required to show that its regulation is the least restrictive means of promoting its

interest, see Ward v. Rock Against Racism, 491 U.S. 781, 798 (1989), this burden

ensures that restrictions on speech are not permitted when either the harms or the

remedial effects of the government’s restrictions are supported only by speculation or

conjecture, or when the regulation burdens substantially more speech than is necessary to

further the government’s legitimate interests. See Edenfield, 507 U.S. at 770-71.9

In order to assess whether the Revised Ordinance is narrowly tailored, we must

measure it against the City’s asserted interest. The City claims that it passed the Revised

Ordinance to “protect pedestrians on medians from encroaching traffic, and drivers from

distractions caused by pedestrians on medians.”10 When we apply “our special standard

9
Although Edenfield, as well as Aptive Environmental, LLC v. Town of
Castle Rock, __ F.3d __, 2020 WL 2503912 (10th Cir. May 15, 2020), which we
address infra, dealt with restrictions on commercial speech, we have recognized that
“[t]he validity of time, place, and manner restrictions is determined under a standard
essentially identical to that governing the regulation of commercial speech.” Citizens
for Peace in Space, 477 F.3d at 1220 n.3; see Lorillard Tobacco Co. v. Reilly, 533
U.S. 525, 554 (2001) (recognizing “substantial similar[ity]” of the two tests). As a
result, we have applied commercial speech precedent when analyzing time, place,
and manner restrictions. See, e.g., Citizens for Peace in Space, 477 F.3d at 1220.
10
Although a government’s interest in public safety is clearly significant, see
McCullen, 573 U.S. at 486, it is not enough for the City to use broad safety
justifications. Rather, “to assess whether a restriction is an appropriate ‘fit’ to some

21
of de novo review” and evaluate the record as a whole to ensure that a “forbidden

intrusion on the field of free expression” has not occurred, Citizens for Peace in Space,

477 F.3d at 1219-20, we conclude that the City has not met its burden to show that its

recited harms are real, see Aptive Envtl., 2020 WL 2503912 at *19.

When we examine the evidence the City offers in support of the Revised

Ordinance, we conclude that the City’s evidence is insufficient to demonstrate that the

City’s “recited harms are real” or that the Revised Ordinance “will in fact alleviate these

harms in a direct and material way.” Citizens for Peace in Space, 477 F.3d at 1221.11

Critically, this record is devoid of evidence that accidents involving vehicles and

important government interest, it is necessary that the government interest be
specifically defined.” Citizens for Peace in Space, 477 F.3d at 1221, 1223.
“Otherwise, the narrowly tailored analysis more closely resembles the ‘reasonably
necessary’ standard used in reviewing restrictions on speech in areas that are not
public forums.” Id. “[T]he question of narrow tailoring must be decided against the
backdrop of the harms that a particular set of [responsive] measures are designed to
forfend.” Id. As a result, “the burden falls on the City to show that its ‘recited harms
are real . . . and that the regulation will in fact alleviate these harms in a direct and
material way.’” Id. at 1221 (quoting Turner Broad. Sys., Inc. v. FCC, 512 U.S. 622,
664 (1994)). As we explained in Aptive Environmental, these requirements are
critical to prevent restrictions on speech designed to advance other interests that
would not on their own justify the burden on expression. 2020 WL 2503912, at *18
(citing Rubin v. Coors Brewing Co., 514 U.S. 476, 487 (1995)).
11
The City has provided ample evidence of citizen complaints about
panhandling across the City, including instances of panhandlers stepping into the
streets, but these complaints do not meet the City’s burden. The City must
demonstrate that its articulated interest of “protect[ing] pedestrians on medians from
encroaching traffic, and drivers from distractions caused by pedestrians on medians,”
is concrete and non-speculative. See id. It cannot meet this burden by proffering
only evidence that there are panhandlers on medians or that those panhandlers are
unpopular without showing that panhandlers in medians create a public safety risk.

22
pedestrians on medians in Oklahoma City is an actual issue, as opposed to a hypothetical

concern. There is neither evidence of any accident involving a pedestrian on a median,

fatal or not, nor evidence that a pedestrian on a median caused an accident or distracted a

driver enough to compromise the safety of the pedestrian or the driver.12

Further, although city officials identified pedestrian presence on medians as one of

their highest concerns, they were unable to identify any accidents in which a pedestrian

on a median was involved. Fowler testified that in his career, he had seen “a couple

hundred” vehicles on medians, although he could neither identify any data, reports, or

other evidence to support that estimate, nor describe any involvement of pedestrians in

these anecdotes. Even if, as the City asserts, there is an increasing number of pedestrians

on medians, there is no objective evidence that these pedestrians are getting hurt or

hurting others. If medians present the danger that the City argues they do, we are baffled

as to why there is no “impersonal hard evidence” of harm arising from their presence.

See Aptive Envtl., 2020 WL 2503912, at *27 (Hartz, J., concurring) (recognizing that

“when a law is justified as a protection of health or safety,” it can “be measured by

impersonal ‘hard’ evidence”).

In contrast, plaintiffs, who apparently spend significantly more time on medians

12
The City points to testimony by Chief Citty and Fowler that they have seen
car tracks on medians, but these marks do not demonstrate that a pedestrian was
present when the marks were created. Further, even evidence of one or a very limited
number of such accidents over an extended period of time might not rise to the level
of evidence necessary to substantiate the government’s interest in enacting this
particular ordinance. Because there is no evidence of any such accident, we need
not, and do not, speculate where the threshold of materiality might lie.
23
than do any city officials, presented evidence demonstrating that they feel safer on

medians than on sidewalks. They testified that cars move slowly at the portions of

medians on which they stand. They also testified that their chosen medians are wide and

protect them from encroaching cars. Drivers testified that they were more aware and

drove more safely both when approaching intersections—where plaintiffs are more likely

to stand—and when around pedestrians on medians. And the City’s chief traffic engineer

testified that there is a “safety zone” for pedestrians eighteen inches beyond a curb that

gives a driver adequate time to regain control of a vehicle after hitting the curb. There is

also evidence that most pedestrian fatalities occur at “mid-block” locations—but again,

not on medians—whereas plaintiffs testified they prefer to stand at the ends of medians,

close to intersections.

The City contends that the government is entitled to prevent anticipated harms and

that its predictive judgments are entitled to substantial deference. It further argues that it

may rely on reasonable inferences drawn from substantial evidence to support its

legislative conclusions and that it may rely on common sense rather than empirical

studies or data to support its assessment of the harm. It is true that municipalities remain

free to determine what type of evidence they will use to support proposed remedial

regulations, and there is no constitutional requirement that governments “compile data or

statistics” in particular. Evans, 944 F.3d at 858; see also Aptive Envtl., 2020 WL

2503912, at *19 (recognizing that the Supreme Court has not “require[d] that empirical

data come accompanied by a surfeit of background information,” but rather has allowed

reference to studies, anecdotes, history, consensus, and common sense to support a

24
municipality’s regulation (quotation and alteration omitted)). We further acknowledge

that a government need not wait for accidents or fatalities to address its interest through

safety regulations. See Evans, 944 F.3d at 858.

Nevertheless, the City’s prerogative to determine how to support a regulation does

not extinguish its burden “to show that its recited harms are real.” See Citizens for Peace

in Space, 477 F.3d at 1221. The evidence in this record does not meet that burden.

Moreover, when evaluating the sufficiency of the municipality’s evidence, regardless of

form, we evaluate that evidence “in light of the cases where those categories of evidence

have previously been invoked.” See, e.g., Aptive Envtl., 2020 WL 2503912, at *19, *21

(holding Castle Rock’s anecdotal and “common sense” evidence to be “woefully

insufficient” when compared to similar evidence held sufficient in Florida Bar v. Went

For It, Inc., 515 U.S. 618 (1995)). The City’s evidence does not stand up to this review.

Compare Evans, 944 F.3d at 854-55 (noting that the ordinance was supported by “several

close calls where accidents involving pedestrians and vehicles could have been

devastating” and city officials had personally surveyed the relative safety of medians

(quotation and alteration omitted)) with Aptive Envtl., 2020 WL 2503912, at *21

(deeming Castle Rock’s evidence “woefully insufficient” because “Castle Rock has

provided us no studies, no supportive evidence-based findings, and no survey results” and

“there [wa]s no evidence that commercial solicitors are the source of any public-safety

problems”). We therefore conclude that the City has not met its burden to demonstrate

that its interest is based on a concrete, non-speculative harm. See Citizens for Peace in

Space, 477 F.3d at 1221; Golan v. Holder, 609 F.3d 1076, 1084 (10th Cir. 2010)

25
(Regulations “must be directed at a real, and not merely conjectural, harm.”).

2

For a regulation to be narrowly tailored, it must not only promote “a substantial

government interest,” but that interest must “be achieved less effectively absent the

regulation, and . . . not burden substantially more speech than is necessary to further the

government’s legitimate interests.” Verlo, 820 F.3d at 1134 (quotation omitted).

Although a regulation “need not be the least restrictive or least intrusive means” of

furthering this interest, the First Amendment requires a “‘close fit between ends and

means’ to ensure speech is not sacrificed for efficiency.” Evans, 944 F.3d at 856

(quoting McCullen, 573 U.S. at 486). “[T]he government may not forgo options that

could serve its interests just as well, if those options would avoid substantially burdening

the kind of speech in which Plaintiffs wish to engage.” Verlo, 820 F.2d at 1135

(quotation and alterations omitted).

To evaluate this fit, we begin by reviewing the evidence the City uses to support

the “means,” i.e., the restrictions imposed by the Revised Ordinance. Our evaluation of

this evidence reveals many of the same weaknesses we identified when analyzing

whether the City’s evidence met its burden to show the existence of a real, non-

conjectural harm. The fundamental problem is that the City has presented no evidence of

concrete harm arising from the presence of pedestrians on its medians. This failure

infects our analysis of both the “ends” and the “means.”

Chief Citty and Fowler testified that in their opinion, pedestrians are in danger

near roadsides, including on medians. They identified medians as the second most

26
dangerous place for pedestrians, after traffic lanes, because medians expose pedestrians

to traffic on both sides. To justify the Revised Ordinance’s applicability to medians on

streets with speed limits above forty miles per hour, the City presented information about

the relative risk of fatalities in auto-pedestrian accidents given different vehicle speeds.13

But not only did this generic “speed kills” evidence not address medians, it also did not

address any other factor, including the relationship between fatalities and the width or

composition of medians, which we held relevant in Evans. 944 F.2d at 858 (“The

Ordinance only prohibits sitting or standing on narrow or unpaved medians where it

would be dangerous to do so. This is the sort of close fit the narrow tailoring requires.”).

The Revised Ordinance places a severe burden on plaintiffs’ speech. In Evans, we

concluded that the ordinance’s burden on speech was minimal because had Evans—the

plaintiff in that case—stood ten feet farther down the same median, he would have been

in compliance with the ordinance. Id. at 857. Under those circumstances, we held that

Evans had not shown that a ten-foot difference substantially burdened his speech. Id.

There is no similarly simple solution for plaintiffs in this case. The Revised Ordinance

entirely prohibits plaintiffs’ presence on the more than four hundred affected medians.

They cannot walk mere feet down a median to reach a legal standing spot. Instead, they

13
As part of its argument, the City occasionally refers to alleged distractions
created by pedestrians on medians. But it identifies no evidence in the record
supporting the proposition that pedestrians on medians provide such distractions that
they affect drivers’ ability to drive safely. Moreover, drivers testified that
pedestrians are used to seeing speakers on medians and that pedestrian presence
makes them more aware of their driving.

27
must leave the median and either stand on a roadside or sidewalk or travel to an

unaffected median on a different block.

Moreover, the fact that plaintiffs may still engage in their speech on roadsides,

sidewalks, or other medians does not mean that their speech is not burdened by the

Revised Ordinance.14 See McCullen, 573 U.S. at 487 (concluding that the regulation

substantially burdened McCullen’s speech despite her persuasion of eighty women not to

terminate their pregnancies because she “‘reache[d] far fewer people’ than she did before

the amendment” (citations and quotation omitted)). The Revised Ordinance is solely

responsible for plaintiffs’ inability to stand on these medians, the most effective place for

their communication.

3

In light of the severity of this burden, the City has failed to demonstrate that less

burdensome alternatives would not achieve its interest in median safety. As the City

acknowledges in its brief, under McCullen, “the government must demonstrate that

alternative measures that burden substantially less speech would fail to achieve the

government’s interests, not simply that the chosen route is easier.” 573 U.S. at 495; see

also ACLU of Colo. v. City & Cty. of Denver, 569 F. Supp. 2d 1142, 1176 (D. Colo.

2008) (“[T]he more extensive the restrictions, the more precise the justifications for that

restriction must be.”). The City asserts that “narrow tailoring does not require [it] to

14
We also note that the passage of the Revised Ordinance was quickly
followed by the expansion of the existing Aggressive Panhandling Ordinance, which
further restricted these alternative locations.
28
undertake the futile task of identifying, reviewing[,] and rejecting alternatives that could

not possibly protect pedestrians on medians from vehicles, or drivers of vehicles from

distractions caused by pedestrians.” But the only way for the City to evaluate alternatives

is to consider them—precisely the burden articulated in McCullen. Because the City

presents us with no evidence that it contemplated the relative efficacy or burden on

speech of any alternatives, we conclude it has not met its burden.

Plaintiffs propose several alternatives that would be less burdensome on speech

but would still advance the City’s asserted interest in median safety. These include

specifying times during which pedestrians can stand on medians, limiting the Revised

Ordinance’s application to the most dangerous intersections, requiring pedestrians to

stand more than eighteen inches back from the curb, or applying the ban to careless or

negligent behavior.

The City dismisses each option out of hand. It asserts it cannot limit the Revised

Ordinance’s application to certain times because “pedestrian accidents occur at all times

of day.” Although this may be true, the record evidence demonstrates that these

accidents certainly are not equally distributed throughout the day. For example, the

City’s evidence demonstrates that “pedestrian collisions are highest in hours where the

sun has set or is setting.”15 Again, in addition to the fact that these collisions are not

15
We note also that the report presenting this data proposes increased
pedestrian infrastructure, education campaigns for drivers and pedestrians about the
most dangerous times of the year for pedestrians, and “traffic-demand management
strategies” to increase safety. It does not propose limiting or banning pedestrian
presence in any manner.
29
related to medians, a cursory statement implying that a time-based restriction would not

stop all pedestrian injuries does not sufficiently demonstrate that the City “seriously

undertook to address the problem with less intrusive tools readily available to it.”

McCullen, 573 U.S. at 494.

The City further states it cannot apply the Revised Ordinance only to dangerous

intersections because six months of data—in contrast to the twelve-plus years of data in

the record—demonstrate that the dangerousness of intersections can vary over time. But

the City has already identified the intersections where fatalities have occurred since at

least 2003, and it provides no argument as to why targeting only these intersections

would fail to achieve its interest. Similarly, a city planning report, adopted by the City

Council and introduced by plaintiffs at trial, identifies the intersections at which auto-

pedestrian accidents most frequently occurred from 2003 to 2015, and the times of day—

early to mid-evening—when accidents were most frequent. Perhaps this data correlates

to a street’s speed limit; perhaps not. But we can say for certain there is no evidence in

the record that the City considered any such correlation in creating its median ban. Given

that the City has at its disposal information regarding the relative safety of its medians at

different times and in different locations, its failure to consider alternatives is especially

harmful to its argument.16 The data supports numerous alternatives to a total ban on

16
We also note that the City employs a chief traffic engineer, who did not
participate in or contribute to the development of the Original or Revised Ordinances.
He was also not requested to conduct any studies, evaluations, or surveys of the
City’s medians to assess their relative safety, or to provide any recommendations
regarding safety measures for pedestrians on medians.

30
presence on affected medians—the “easier” (and far more burdensome to speech)

alternative that was selected by the City.17

As for plaintiffs’ proposal that the City require pedestrians to stay further than

eighteen inches back from the curb, the City summarily responds that vehicles could

travel further than eighteen inches. But it offers no evidence of the frequency with which

vehicles travel further than eighteen inches, does not account for the fact that its own bus

stops are placed at that distance from the curb, and disregards the testimony of its own

chief traffic engineer that the eighteen-inch distance provides a “safety zone” for

pedestrians.18

We also note that as in McCullen, the City has existing laws that could advance its

interest in pedestrian safety on medians. For example, one law prohibits people from

stepping into the street. City officials dismissed this alternative, stating that “it’s very

17
We note that the characteristics of medians that plaintiffs argue make them
“safe,” see supra, also provide information that would help the City craft an
alternative to the Revised Ordinance that on this record would be less burdensome to
plaintiffs’ speech. For example, the City could limit the ordinance’s application to
medians that are both insufficiently wide and located on streets where the actual
speed of passing cars poses a substantial danger to pedestrians. Or it could only
apply the Revised Ordinance to the mid-block portions of medians on a given block.
The City has more than enough information to craft an ordinance that burdens less
speech and is closely aligned with its stated interest of public safety, should it decide
to do so.
18
We are also not convinced by the City’s cursory response that it could not
regulate pedestrian activity on roadsides or sidewalks. It already does. See, e.g.,
§§ 30-428, 30-430 (barring aggressive panhandling within twenty feet of outdoor
restaurant seating, within fifty feet of a mass transportation stop, within fifty feet of
school property, after dark, or of minors); § 30-81(f) (barring obstruction of
pedestrian traffic on sidewalks).
31
hard for the police to be there when [a person] actually step[s] off.” But ease of

application is not a sufficient reason to burden First Amendment rights. See McCullen,

573 U.S. at 495 (stating that “[o]f course” the state’s regulation would make law

enforcement’s “job so much easier,” but that the burden requires “demonstrat[ing] that

alternative measures that burden substantially less speech would fail to achieve the

government’s interests, not simply that the chosen route is easier”).

In addition to our conclusion that the City has not demonstrated that pedestrian

presence on medians is a concrete, non-speculative problem, we also conclude that the

City’s summary dismissal of alternatives is insufficient. “[G]iven the vital First

Amendment interests at stake, it is not enough for [the City] simply to say that other

approaches have not worked.” McCullen, 573 U.S. at 496. This is particularly so when

there is no evidence that the City has tried, or even considered, any less-burdensome

alternatives. Instead, the City relies on unsupported statements that hypothetically these

alternatives could not possibly work. The City “has not shown that it seriously undertook

to address the problem with less intrusive tools readily available to it[, n]or has it shown

that it considered different methods that other jurisdictions have found effective.”

McCullen, 573 U.S. at 494; see also Cutting, 802 F.3d at 91 (“But the City did not try—

or adequately explain why it did not try—other, less speech restrictive means of

addressing the safety concerns it identified.”); Reynolds, 779 F.3d at 231 (“[T]he burden

of proving narrow tailoring requires the County to prove that it actually tried other

methods to address the problem.”).

32
 4

The underinclusive nature of the Revised Ordinance also demonstrates at best a

loose fit between its means and the City’s safety interest. Under the Revised Ordinance’s

exception for legally authorized work, the City permits a non-profit named OKC

Beautiful to landscape its medians. Through OKC Beautiful, volunteers from private

businesses and organizations may stand, sit, or otherwise stay on medians to landscape

the medians for the City and, in return for their services, install signs on the medians

publicizing the entity’s sponsorship.

Therefore, at the same time as OKC Beautiful volunteers of all ages are permitted

to remain on medians for substantial periods of time, the City entirely bars plaintiffs’

presence. Surely if it is safe for volunteers to be on the medians long enough to beautify

them, it is also safe for plaintiffs to be on the medians for similar periods of time. But the

Revised Ordinance only allows for the volunteers’ activities, not for plaintiffs’ protected

expression. The City provides no real answer to this discrepancy, stating only that “[a]

regulation is not otherwise objectionable simply because it doesn’t address all potential

problems.” But on this record, plaintiffs’ expression is protected by the First

Amendment, whereas the volunteers’ beautification efforts are not. See Williams-Yulee

v. Fla. Bar, 575 U.S. 433, 448-49 (2015) (“[U]nderinclusiveness can raise doubts about

whether the government is in fact pursuing the interest it invokes, rather than disfavoring

a particular speaker or viewpoint” and “can also reveal that a law does not actually

advance a compelling interest” (quotation omitted)). The City has thus selectively

criminalized activities protected by the First Amendment while allowing unprotected

33
activities to proceed.19

5

We conclude that the Revised Ordinance is not narrowly tailored to the

problem it purports to address. The City has utterly failed to demonstrate the

requisite “‘close fit between ends and means’ to ensure speech is not sacrificed for

efficiency.” Evans, 944 F.3d at 856 (quoting McCullen, 573 U.S. at 486); cf. id.

(holding that the ordinance was narrowly tailored because the burden on Evans’

speech was insignificant and because the regulation directly related to the exact

information city officials had compiled regarding safety of medians in their city).

The City has taken “the extreme step of closing a substantial portion of a traditional

public forum to all speakers . . . without seriously addressing the problem through

19
The City cites Williams-Yulee for the proposition that underinclusiveness is
only relevant if a regulation is “riddled with exceptions.” In Williams-Yulee, the
Court concluded that the law “raise[d] no fatal underinclusivity concerns” because it
“aim[ed] squarely at the conduct most likely to” advance its interest; it “appli[ed]
evenhandedly to all;” and “unlike some laws that we have found impermissibly
underinclusive, [the law at issue in that case] is not riddled with exceptions” because
it “contains zero exceptions to its ban on personal solicitation.” 575 U.S. at 449.
Certainly, a regulation’s being “riddled with exceptions” demonstrates “red flag[s].”
Id. But nothing in Williams-Yulee suggests underinclusiveness is only relevant when
multiple exemptions are present. And in any event, the exception to the Revised
Ordinance is, on this record, not constitutionally protected activity. Therefore, the
Court’s conclusion in Williams-Yulee that it would “not punish Florida for leaving
open more, rather than fewer, avenues of expression” does not apply. Id. at 452.
Finally, unlike the regulation at issue in that case, the Revised Ordinance is not
“aim[ed] squarely at the conduct most likely to” achieve the City’s interest. Id. The
evidence demonstrates that pedestrians are getting harmed in myriad other locations,
but not on medians. Therefore, Williams-Yulee does not change our conclusion that
the exemption for OKC Beautiful volunteers is pertinent and supports our holding
that the Revised Ordinance is not narrowly tailored.
34
alternatives that leave the forum open for its time-honored purposes. The [City] may

not do [so] consistent with the First Amendment.” McCullen, 573 U.S. at 497; see

also Reynolds, 779 F.3d at 232 (“[T]here is no evidence that the County ever tried to

improve safety by prosecuting any roadway solicitors who actually obstructed traffic,

or that it ever even considered prohibiting roadway solicitation only at those

locations where it could not be done safely. Without such evidence, the County

cannot carry its burden of demonstrating that the Amended Ordinance is narrowly

tailored.”).

E

Although “the First Amendment does not guarantee the right to employ every

conceivable method of communication at all times and in all places, a restriction on

expressive activity may be invalid if the remaining modes of communication are

inadequate.” Evans, 944 F.3d at 860 (quoting City Council of L.A. v. Taxpayers for

Vincent, 466 U.S. 789, 812 (1984)).20 Analysis of alternative channels “must give

practical recognition to the facts giving rise to the restriction on speech.” Citizens for

Peace in Space, 477 F.3d at 1226. Accordingly, we “ask whether, given the particular

[governmental interest], the geography of the area regulated, and the type of speech

desired, there were ample alternative channels of communication.” Id. “To determine

20
Our conclusion that the City has failed to show that the Revised Ordinance
is narrowly tailored means we are not required to address whether the Revised
Ordinance leaves open ample alternative channels of communication. See, e.g.,
McCullen, 573 U.S. at 496 n.9. However, because this issue is likely to arise again,
we address it. See Reynolds, 779 F.3d at 232 n.5.
35
whether alternative channels are adequate, courts assess in part the speaker’s ability to

reach his or her intended audience.” Evans, 944 F.3d at 860 (citing Ward, 491 U.S. at

802).

In this case, the parties’ dispute regarding this inquiry is one of amount. The City

argues that plaintiffs have not been “completely foreclosed” from either using medians or

engaging in their chosen speech, whereas plaintiffs argue that the lack of viable

communicative spaces means that ample alternative channels do not exist. We agree with

plaintiffs. A valid time, place, and manner restriction “leave[s] open ample alternative

channels of communication.” Frisby, 487 U.S. at 482. The inquiry is not whether the

restriction completely forecloses speech. And not only must there be ample channels;

those channels must also be adequate. See Taxpayers for Vincent, 466 U.S. at 812. For

example, the Supreme Court has held that ample channels existed when an ordinance

allowed picketing in neighborhoods, but barred protesters from picketing a single

residence, see Frisby, 487 U.S. at 483-84, or when the government restriction only

imposed an eight-foot distance between demonstrators and their audience, see Hill v.

Colorado, 530 U.S. 703, 729-30 (2000). And in Evans, we held that ample alternative

channels existed when a pedestrian could stand on the same median ten feet away and

comply with the ordinance. 944 F.3d at 860.

Although we have recognized that “[c]itizens do not have a right to convey their

message in any manner they prefer, . . . they [do] have a right to convey their message in

a manner that is constitutionally adequate.” Citizens for Peace in Space, 477 F.3d at

1226. For plaintiffs such as charitable solicitors, political campaigners, protestors, or

36
panhandlers who engage with passing cars—or, more realistically, with their drivers—the

record does not support the district court’s conclusion that moving to a sidewalk is an

adequate alternative. Record evidence showed that for many of these plaintiffs,

communications from sidewalks and roadsides would not provide adequate alternative

opportunities for communication. Signs and communications from the sidewalk would

not be as visible to those in cars, and plaintiffs would have to compete with a jumble of

other signs and messages from storefronts. And, for those seeking to hand out material or

seeking to solicit funds—all of which we have held to be expressive activity—neither

roadsides nor sidewalks would provide safe and direct access to the driver, who often will

be a car’s only occupant. Rather, solicitors must step into the road to close the distance

between drivers and themselves, making exchanges from roadsides more difficult and

dangerous than those from medians. As one councilman noted while deliberating the

ordinance, “it’s the entry into the street which is where all the violent impact occurs.”

Plaintiffs are also out of the sightline of drivers when on sidewalks. As plaintiff

politician Faulk testified, a sidewalk is “just not as effective” because “if you have eight

to nine lanes of traffic and you’re standing on a street corner, you’re only reaching the

traffic right next to you, and so maybe you see two lanes of traffic and they see what

you’re holding up.” In contrast, “[i]n the median, you catch traffic coming from all four

directions and you catch traffic from every lane in each direction. So you may be

reaching as many as 16 lanes of traffic sometimes.” Just as in real estate, location matters

in some constitutional questions. Cf. McCullen, 573 U.S. at 490 (holding

unconstitutional buffer zone restriction requiring plaintiffs to stand a substantial distance

37
away from an abortion provider’s driveway).

Further, record evidence demonstrates that there are “an extremely limited

number” of medians unaffected by the Revised Ordinance on which plaintiffs can engage

in their expression. According to the City, there are only 103 unaffected medians, in

contrast to approximately four hundred affected ones. And by the district court’s own

estimate, fewer than eighty of those 103 are accessible to plaintiff McCraw, plaintiff

Marshall, and other panhandlers due to the Aggressive Panhandling Ordinance. Plaintiffs

testified that “[i]t’s getting hard to find a spot” because “[t]here’s just not that many

places for panhandlers and Curbside vendors to be.” Competition for the few remaining

unaffected medians is “fierce” and even violent, inhibiting plaintiffs’ ability to use them.

Moreover, even if a plaintiff is lucky enough to secure a spot, unaffected medians can be

significantly less effective at conveying speech because fewer vehicles are present. This

evidence distinguishes this case from our decision in Evans, where we concluded that

Evans did not “distinguish his ability to communicate with his target audience on affected

or unaffected medians” and failed to demonstrate why a legal position ten feet down a

median was constitutionally inadequate.21 944 F.3d at 860. For plaintiffs who must

engage with passing drivers for their expression, the City has also not demonstrated that

21
The City pointed to evidence that the number of Curbside vendors has
increased and that the Oklahoma Libertarian Party was able to obtain the requisite
number of signatures on a ballot initiative petition. But evidence demonstrates that
Curbside’s sales have decreased under the Revised Ordinance, which better reflects
the inadequacy of the unaffected medians, and that the OLP garnered most of its
signatures before the Revised Ordinance took effect.

38
there are ample adequate alternative channels for plaintiffs’ speech.22

F

For the above reasons, we conclude that the City has not met its burden to

demonstrate that the Revised Ordinance is a constitutionally permissible time, place, and

manner restriction. We therefore hold that the Revised Ordinance violates the First

Amendment.

III

Turning to plaintiffs’ due process claims, we first consider the district court’s

dismissal of the claim that the Revised Ordinance violates their fundamental right to

move or linger in traditionally open public places.

In addition to the explicitly recognized right to interstate travel, the Supreme Court

has hinted at a right to freedom of local movement. In City of Chicago v. Morales, 527

U.S. 41 (1999), a three-judge plurality noted that “it is apparent that an individual’s

decision to remain in a public place of his choice is as much a part of his liberty as the

freedom of movement inside frontiers that is a part of our heritage or the right to move to

22
Our concerns regarding the adequacy of alternatives do not extend to
Wilson, Schindler, and Red Dirt Report. Wilson provides no evidence to support a
conclusion that she cannot stop to talk with her jogging companions on sidewalks or
roadsides, rather than medians. Schindler likewise provides no evidence that aside
from his preference to run on grass along the marathon route—which does not pertain
to the effectiveness of his expression—his expressive activity cannot be equally
accomplished on sidewalks or roadsides. Finally, Red Dirt Report does not provide
enough evidence to demonstrate that sidewalks and roadsides across the street from
the event are not adequate alternatives for it to report the news. Although medians
present benefits by being “a little raised above the road” and out of the way of the
first responders, we do not see sufficient evidence in the record to show that a nearby
sidewalk or roadside across the street would not be an adequate alternative.
39
whatsoever place one’s own inclination may direct[,] identified in Blackstone’s

commentaries.” Id. (plurality) at 54 (citations and quotations omitted); see also Kent v.

Dulles, 357 U.S. 116, 126 (1958) (addressing a claim regarding international travel, but

noting that “[f]reedom of movement is basic in our scheme of values”); Papachristou v.

City of Jacksonville, 405 U.S. 156, 164 (1972) (stating that walking, loitering, and

similar activities “are historically part of the amenities of life as we have known them”).

But see Bray v. Alexandria Women’s Health Clinic, 506 U.S. 263, 277 (1993) (stating

that “a purely intrastate restriction does not implicate the right of interstate travel”).

Several of our sibling circuits have explicitly held that a fundamental right to the

freedom of movement exists. See Johnson v. City of Cincinnati, 310 F.3d 484, 498 (6th

Cir. 2002) (“[W]e hold that the Constitution protects a right to travel locally through

public spaces and roadways.”); Nunez by Nunez v. City of San Diego, 114 F.3d 935, 944

(9th Cir. 1997) (“Citizens have a fundamental right of free movement . . . .” (citation

omitted)); Lutz v. City of York, Pa., 899 F.2d 255, 268 (3d Cir. 1990) (“[T]he right to

move freely about one’s neighborhood or town, even by automobile, is indeed implicit in

the concept of ordered liberty and deeply rooted in the Nation’s history,” therefore a

fundamental right.). Other circuits have concluded that individuals hold a liberty interest

in the freedom of movement but declined to deem the right fundamental. See Catron v.

City of St. Petersburg, 658 F.3d 1260, 1266 (11th Cir. 2011) (without addressing whether

the right is fundamental, holding that “[p]laintiffs have a constitutionally protected liberty

interest to be in parks or on other city lands of their choosing that are open to the public

generally”); Doe v. City of Lafayette, 377 F.3d 757, 769-70 (7th Cir. 2004) (asserting

40
that the “right to enter the parks to loiter or for other innocent purposes . . . is not

unimportant, [but] we cannot say that existing authority establishes that it is

“fundamental”). Other circuits have held that regardless of a right to intrastate travel,

there is no right to remain on a specific piece of public property. See Williams v. Town

of Greenburgh, 535 F.3d 71, 76 (2d Cir. 2008) (holding that although there is a right to

intrastate movement, “it would distort the right to free travel beyond recognition to

construe it as providing a substantive right to cross a particular parcel of land, enter a

chosen dwelling, or gain admittance to a specific government building” (emphasis

omitted)).

We have previously concluded that the fundamental right to freedom of movement

“appl[ies] only to interstate travel.” D.L. v. Unified Sch. Dist. No. 497, 596 F.3d 768,

776 (10th Cir. 2010). In D.L., plaintiffs’ claim that their fundamental right to travel was

infringed by their expulsion from a public school for non-residency failed because “the

travel that Plaintiffs claim was restricted was intrastate travel.” Id. Under D.L.,

plaintiffs’ argument that they have a fundamental right to remain on Oklahoma City’s

medians—a purely intrastate location—similarly fails. Although plaintiffs raise out-of-

circuit decisions and Supreme Court dicta to support their claim, “[w]e must generally

follow our precedents absent en banc consideration.” United States v. Lira-Ramirez, 951

F.3d 1258, 1260 (10th Cir. 2020).

Concluding that plaintiffs do not have a fundamental right to remain on the City’s

medians, we analyze their due process claim under rational-basis review. To satisfy this

test, the Revised Ordinance “need only be rationally related to a legitimate government

41
purpose.” Powers v. Harris, 379 F.3d 1208, 1215 (10th Cir. 2004). “[W]e look only to

whether a reasonably conceivable rational basis exists [and] . . . are not allowed to second

guess the wisdom of legislative policy-determinations.” Ramsey Winch Inc. v. Henry,

555 F.3d 1199, 1210 (10th Cir. 2009) (citation and quotation omitted). Although we

determined above that the City’s evidence failed intermediate scrutiny in the First

Amendment context, it passes the lower bar of rational-basis review applicable when we

review due process challenges to non-fundamental rights. Public safety is a significant

government interest and, although the Revised Ordinance may not be narrowly tailored or

provide ample adequate alternatives, it is conceivably rationally related to public safety.

IV

Finally, we consider plaintiffs’ claim that the Revised Ordinance is

unconstitutionally vague, on which the court granted summary judgment to the City.

“The void-for-vagueness doctrine requires that a penal statute define the criminal offense

with sufficient definiteness that ordinary people can understand what conduct is

prohibited and in a manner that does not encourage arbitrary and discriminatory

enforcement.” United States v. Hunter, 663 F.3d 1136, 1141 (10th Cir. 2011) (quotation

omitted).

The Revised Ordinance exempts “[i]ndividuals responding to any emergency

situation.” § 32-458(e)(4). We hold that the term “emergency,” as used in the Revised

42
Ordinance, is not unconstitutionally vague.23 An “emergency” is commonly understood

to mean “an unforeseen combination of circumstances or the resulting state that calls for

immediate action;” “an urgent need for assistance or relief,” Emergency, Merriam-

Webster, https://www.merriam-webster.com/dictionary/emergency (last visited June 24,

2020); “[a] juncture that arises or ‘turns up;’” or “a state of things unexpectedly arising,

and urgently demanding immediate action,” Emergency, Oxford Eng. Dict.

https://www.oed.com/view/Entry/61130?redirectedFrom=emergency#eid (last visited

June 24, 2020).

We do not mean to minimize the importance of plaintiffs’ proposed scenarios or

their real need to stand on medians, but we conclude there is a readily discernible

difference between, on the one hand, “a candidate campaigning in an unexpectedly close

election, an activist protesting an unforeseen event, a panhandler soliciting for

unexpected expenses, a reporter covering breaking news, or a jogger responding to a text,

call, cramp, or untied shoelace,” and on the other hand, someone having to stand on a

23
We acknowledge the parties’ dispute regarding the applicable definition.
The Original Ordinance excised most of the existing definition, leaving “emergency”
to mean “an unforeseeable occurrence of temporary duration.” Ordinance 25,283.
The Revised Ordinance kept that definition. Ordinance 25,777. Later, Ordinance
25,709 repealed and replaced the definition passed in the Original Ordinance. Okla.
City, Okla. Ordinance 25,709 (Aug. 30, 2017). Ordinance 25,709’s definition
restored much of the original language defining an “emergency” as “an unforeseeable
occurrence of temporary duration causing or resulting in an abnormal increase in
traffic volume, cessation or stoppage of traffic movement, or creation of conditions
hazardous to normal traffic movement, including fire, storm, accident, riot, or
spontaneous assembly of large numbers of pedestrians in such a manner as to impede
the flow of traffic.” Id. § 32-1(22). But we need not decide what the applicable
definition is—under all three definitions, the term is not unconstitutionally vague.
43
median due to an emergency. Dictionary definitions underscore the immediacy and

urgency of an emergency. And we conclude that an ordinary person would know that the

above scenarios do not require action that is sufficiently immediate and urgent to

constitute an emergency, and therefore, an exception to the Revised Ordinance. The

Revised Ordinance thus “conveys sufficiently definite warning as to the proscribed

conduct when measured by common understanding and practices.” Roth v. United

States, 354 U.S. 476, 491 (1957) (quotation omitted).

V

We REVERSE the district court’s dismissal of Wilson’s First Amendment claim

and its entry of judgment for the City on all plaintiffs’ First Amendment claims, and

REMAND for further proceedings consistent with this opinion. We AFFIRM the

court’s grant of summary judgment to the City on plaintiffs’ vagueness claim and its

entry of judgment for the City on plaintiffs’ freedom of movement claim.

44
19-6008 – McCraw v. Oklahoma City

HARTZ, J., Circuit Judge, concurring
Although I cannot agree with some of the language in the majority opinion and

some of the statements in it that are unnecessary to decide this case, I do agree with the

holding.

First, I agree that as a general rule medians should be treated as public fora for

First Amendment purposes. I would be open to argument that an exception should be

made for the narrowest medians, which are clearly not designed for people to stand on or

congregate. But that issue need not be resolved now.

Second, I agree that the City has failed to show that its ordinance is narrowly

tailored to serve a significant governmental interest. The City has singularly failed to

support its ordinance with either data or expert reasoned opinion. The purported

government interest is public safety. But a number of years of relevant data failed to

support the claimed danger. I am not saying that such data are necessary to support a

claim of danger. Common sense and expert opinion may well suffice. But when there

are data available, and they contradict what common sense and expert opinion may tell

us, courts must be cautious before endorsing a governmental claim of danger.

Exercising that caution, I cannot see that the ordinance is narrowly tailored. I

would be inclined to be quite deferential to an ordinance that prohibits standing or

congregating only on narrow medians, or portions of medians, that bordered

thoroughfares with relatively high speed limits, since the proximity and exposure to the

traffic could lead to tragedy if the pedestrian on the median or a driver on the
thoroughfare is distracted or otherwise negligent. See Evans v. Sandy City, 944 F.3d 847,

860 (10th Cir. 2019) (upholding ordinance prohibiting sitting or standing on median of

less than 36 inches). Such an ordinance would appear to be narrowly tailored. But the

record in this case indicates that the medians in Oklahoma City come in a variety of sizes

and shapes. And the City has given no reasoned argument why safety requires barring

pedestrians from the entirety of all medians at all times of day if the speed limit on the

adjacent road is at least 40 mph. In particular, some of the medians covered by the

ordinance are well over 36 inches wide, so people on the median could be more than 18

inches from traffic on both sides. Yet the city’s own chief traffic engineer testified that

18 inches from a curbed roadway provides a safety zone and that city benches may be

placed as close to the curb as 18 inches.

2