Explaining the Persistence of the “Ample Alternative Channels” Test
Explaining the Persistence of the “Ample Alternative Channels” Test
Thomas Berry 0 1 2
0 Thi s Response is brought to you for free and open access by the Law School Journals at Washington & Lee University School of Law Scholarly Commons. It has been accepted for inclusion in Washington and Lee Law Review Online by an authorized editor of Washington & Lee University School of Law Scholarly Commons. For more information , please contact , USA
1 Thom as Berry, Explaining the Persistence of the “Ample Alternative Channels” Test, 73 Wash. & Lee L. Rev. Online 786 , 2017
2 Cato Institute Center for Constitutional Studies
Follow this and additional works at: http://scholarlycommons.law.wlu.edu/wlul-ronline Part of theLaw Commons Recommended Citation
Explaining the Persistence of the
“Ample Alternative Channels” Test
I. Introduction ....................................................................786
II. The Three Meanings of “Channel” ..................................789
III. Restrictions on Audience: The Quintessential
“Ample Alternative Channels” Case ...............................791
IV. Toward a Better Version of the Marketplace Theory .....793
A. The Marketplace Theory as Currently Applied ........793
B. The Self-Autonomy Theory........................................794
C. The Marketplace Theory as It Should Be Applied....797
V. Why Courts Have Upheld Category III Restrictions ......798
Professor Enrique Armijo’s excellent article on the history of
the “Ample Alternative Channels” Doctrine1 provides plausible
answers to two of the most important questions—“what” and
“how”—raised by this problematic rule. What have courts done
when presented with speech restrictions that supposedly leave
“ample alternative channels” for speech? Armijo provides a
thorough catalog of the major cases that have invoked ample
alternative channels, both in upholding speech restrictions and
(occasionally) in striking them down.2 And how can our First
* Legal Associate, Cato Institute Center for Constitutional Studies.
1. Enrique Armijo, The “Ample Alternative Channels” Flaw in First
Amendment Doctrine, 73 WASH. & LEE L. REV. 1657, 1659 (2016).
2. See id. at 1705–22 (discussing the conflict between the ample
alternative channels doctrine and free speech zone cases, adult entertainment
Amendment jurisprudence be improved so that such unjustified
restrictions on speech are not upheld in the future? Armijo gives
a convincing answer, namely that courts must require the
government to show that the use of a particular channel of speech
is incompatible with a government interest.3
Rather than attempting to improve on Armijo’s work on
these two questions, this response will focus instead on a third
question: Why. Why has the Ample Alternative Channels
Doctrine (AAC) so firmly taken hold in our judicial canon?
Professor Armijo provides the first half of an answer to this
question with his engrossing and revelatory behind-the-scenes
account of Justice Harlan’s O’Brien concurrence.4 The history
behind Harlan’s opinion tells the AAC origin story. But this still
leaves unanswered why courts have accepted, applied, and even
expanded the AAC doctrine for over forty years since O’Brien,
seldom noting dissent or even concern with the implications of
This response will examine several factors that have
contributed to the AAC doctrine’s resilience, and will also suggest
possible solutions to counteract these factors and thereby move
away from the AAC doctrine. The most basic cause of the
doctrine’s longevity, I propose, is linguistic. The concept of a
“channel” of speech may seem simple enough. Yet, in fact, courts
have used the word “channel” to describe three distinct elements
of speech transmission.6 The failure to distinguish between these
meanings has led to a blurring of the lines between regulations
that truly affect speech and those that do not.7
cases, prohibited means cases, and abortion clinic protest cases ).
3. See id. at 1728–38 (providing as a solution the “incompatibility test”).
4. See id. at 1670 (“Alternative channels analysis was born . . . in Justice
Harlan’s concurrence. And an earlier version of that latter opinion had much
bigger game in its sights: what Harlan viewed as the dangerous logical fallacy of
the Court’s distinguishing between speech and content in First Amendment
cases.”); id. at 1671–89 (providing the background of Justice Harlan’s
concurrence in O’Brien).
5. See id. at 1668 (noting that although the “ample alternative channels
analysis was in its incipiency a misguided afterthought . . . the concept now
carries dispositive force in First Amendment doctrine”).
6. See infra Part II (discussing the three meanings or “categories” of the
But a simple linguistic misunderstanding cannot be the full
explanation. An additional problem arises from the
misapplication of what Armijo calls the “marketplace theory” of
the First Amendment.8 Courts have overwhelmingly applied the
marketplace theory when analyzing speech restrictions, justifying
the value of speech based on its contribution to the marketplace
of ideas.9 This has been to the detriment of speakers challenging
speech restrictions, who have been forced to prove that their
speech (in its intended form) is unique and irreplaceable. If a
regulation seems to remove easily replaceable speech from the
marketplace, then that regulation is frequently upheld as having
no real harm.10
The answer to these errors, I propose, is not to move to the
alternative “self-autonomy” theory,11 which focuses on the rights
of speakers rather than the effects on listeners.12 That theory is
implausible as a definition of speech rights since it can protect
choices unrelated to the perception of an audience. Instead, the
answer is to move to a modified version of the marketplace
theory, one that is still listener-focused but not open to judicial
balancing. Under this view, any regulation that affects the
experience of listeners is a violation of free speech rights. The job
of courts should be limited to determining only whether a
regulation has altered the experience of listeners. Once this
objective line is crossed, the regulation qualifies as a speech
restriction and should only be upheld if it passes a strict test such
as the one Armijo proposes. 13
8. See Armijo, supra note 1, at 1705 (“Ample alternative channels
respects marketplace theory in those cases where alternatives are found to be
poor substitutes, but disrespects marketplace theory in those cases where
alternatives are found to be proper substitutes.”).
9. See id. at 1696 (describing marketplace theory as “the dominant theory
of the First Amendment”).
10. See id. at 1665 (“The restrained expression can thus still contribute to
the search for truth, so the harm the restriction causes, both to the speaker and
to listeners participating in the broader speech market, is minimal.”).
11. See generally id. at 1689–95 (discussing the self-autonomy theory).
12. See id. at 1690–91 (“Accordingly, for self-autonomy theory, it is the
speaker’s choice of expression that is . . . ‘the crucial factor in justifying
protection’ of that expression.” (quoting C. EDWIN BAKER, HUMAN LIBERTY AND
FREEDOM OF SPEECH 52 (1989))).
13. See infra Part IV (elaborating on this solution).
Finally, I will consider why courts have been willing to use
the AAC doctrine to uphold regulations that manifestly interfere
with the content of a speaker’s message, such as the regulation in
O’Brien. One reason, I will argue, is that courts wish to avoid an
uncomfortable fact: if breaking a law is a uniquely powerful form
of expression against that law, then enforcing the law inherently
places an asymmetrical burden on critics of the status quo. For a
legal regime that prides itself on striving for viewpoint neutrality,
this asymmetry is not only an unfortunate fact, but also an
unavoidable one. Courts should acknowledge it, so as to be
clear-eyed in measuring the burden that a regulation places on
II. The Three Meanings of “Channel”
Much of the confusion surrounding the “ample alternative
channel” doctrine arises from linguistic imprecision. Laying out
the three possible meanings of the word “channel” at the outset
will help clarify the distinctions drawn later in this response.
First, there are cases where two different “channels” can be
used to carry the same message to the same set of recipients. For
example, suppose that a letter could be carried by truck to your
city and then placed in your mailbox on Wednesday morning.
Alternatively, that same letter could be carried to your city by
plane and then placed in your mailbox on Wednesday morning.
The truck and the plane are two alternative “channels” by which
the message might travel. But, regardless of which channel is
chosen, your perception, as the recipient of the message, is
identical in every way. No matter what, you will find the same
letter in your mailbox on Wednesday morning. I will call this type
of channel a “Category I” channel.
Second, there are cases where two different “channels” can be
used to carry the same message to two different sets of recipients.
For example, suppose that an open letter could be publically
disseminated either via flyers posted around a town or via a post
on an internet blog. The letter consists entirely of text, and the
speaker intends the text to be the full extent of the letter’s
14. See infra Part V (exploring why courts have upheld restrictions on
speech where two different channels can be used to carry different messages).
message. In this case, everyone who sees the letter receives the
same message, whether they read it on a flyer or on a blog. But
the number of people who see the message, and the identity of
those people, may well differ depending on which of these two
channels is used.15 I will call this type of channel a “Category II”
Third, and finally, there are cases where two different
“channels” can be used to carry different messages.17 For example,
suppose that an anti-materialist message could be delivered by
means of either burning a real $100 bill or burning a fake $100
bill. The real bill and the fake bill are two different “channels” by
which—through the means of a dramatic demonstration—an
anti-materialist message could be delivered. But because they are
two different objects being burned (only one of them representing
an actual sacrifice on the part of the burner), the message
received by viewers of the two potential demonstrations would be
different. I will call this type of channel a “Category III” channel.
With this terminology laid out, the next section will explore
the category where courts have most frequently used the AAC
doctrine to uphold speech restrictions: Category II cases.
15. For example, the flyers may be read only by the residents of a
particular locality, while the blog post may be read by a much more
geographically dispersed audience (which, if the issue is one of local politics,
may be less desired by the speaker).
16. In these Category II cases, speakers are harmed by being deprived of
what Armijo calls a “technique” of disseminating their messages. See Armijo,
supra note 1, at 1684 (arguing that “the availability of alternative means seems
much more relevant to a law that incidentally burdens speech by banning
‘techniques’ for expression than to a law that bans the expressive act itself”). As
Armijo explains, the term “technique” was first used by Justice Black. Id. at
1684 n.121. In Martin v. Struthers, Black wrote that “door to door campaigning
is one of the most accepted techniques of seeking popular support.” Martin v.
City of Struthers, 319 U.S. 141, 146 (1943) (emphasis added). A Category II ban
is “a restriction on one ‘technique’ by which a message can be expressed.”
Armijo, supra note 1, at 1684.
17. These are the situations where, as Ed Baker puts it, “the intended
meaning of people’s expression relates to the time or the place or the manner of
the expression.” C. Edwin Baker, Unreasoned Reasonableness: Mandatory
Parade Permits and Time, Place, and Manner Restrictions, 78 NW. U. L. REV.
937, 946 (1984). This category corresponds to the “associative conduct” category
of cases that Armijo describes—cases where “the relevant speech’s intended
message and effects, along with its particular audience, are inextricably
associated with the message’s mode, time, and place.” Armijo, supra note 1, at
III. Restrictions on Audience: The Quintessential “Ample
Alternative Channels” Case
The most common “Category II” speech restrictions may be
regulations that prevent people from speaking in certain physical
locations. Frequently, such regulations come in the form of a
so-called “buffer zone,” a rule preventing speakers from coming
within a certain distance of a sensitive area.18 Buffer zones affect
who will receive a speaker’s intended message. For example, a
buffer that prevented protesters from approaching the site of a
World Trade Organization (WTO) conference “unquestionably
limited the barred speakers from reaching the speech
marketplace of their choice—those individuals the protestors
most sought to persuade, i.e., the WTO delegates themselves, as
well as others closely following the conference.”19
How have courts ruled in these cases? Most often, as
Professor Armijo catalogs, the regulations have survived thanks
to the AAC doctrine.20 But it may be those rare cases in which
regulations have been struck down that are the most telling. In
McCullen v. Coakley,21 the Supreme Court struck down a law that
banned approaching persons to engage them in conversation
within 35 feet of an abortion clinic.22 Why did this buffer zone fail
18. See Armijo, supra note 1, at 1718 (“A typical example is from Clift v.
City of Burlington, Vermont, in which Burlington adopted a 35-foot radius
around reproductive health care facilities in the city. Burlington’s ordinance
decreed that ‘no person or persons shall knowingly congregate, patrol, picket, or
demonstrate in the buffer zone.’” (citing Clift v. City of Burlington, Vt., 925 F.
Supp. 2d 614 (D. Vt. 2013))). Another case involving abortion clinic buffer zones
is Hill v. Colorado, 530 U.S. 703 (2000). Cases involving protesters of
conferences and events include Menotti v. City of Seattle, 409 F.3d 1113 (9th
Cir. 2005); Bl(a)ck Tea Society v. City of Boston, 378 F.3d 8 (1st Cir. 2004);
Coalition to March on the RNC and Stop the War v. City of St. Paul, Minn., 557
F. Supp. 2d 1014 (D. Minn. 2008); and American Civil Liberties Union of
Colorado v. City and County of Denver, 569 F. Supp. 2d 1142 (D. Col. 2008).
20. See id. at 1718 (“The question then becomes whether that physical
separation impermissibly impedes on the speakers’ expression. And through the
use of ample alternative channels analysis, courts consistently hold that it does
22. See id. at 2541 (concluding that “the extreme step of closing a
substantial portion of a traditional public forum to all speakers” is not
where so many others succeeded? Perhaps the most important
clue is that the Court described the ban as interfering with “the
close, personal conversations that [the petitioners] view as
essential to ‘sidewalk counseling.’”23 In other words, the speakers
were helped by having a unique and specific method of
communication, one that the Court could easily grasp as being
connected in a special way to the area in which they wished to
Similarly, in Weinberg v. City of Chicago,24 the U.S. Court of
Appeals for the Seventh Circuit easily understood why the author
of an anti-Chicago Blackhawks book would particularly wish to
reach attendees at Blackhawks games.25 As a result, a restriction
banning sales of the book near the Blackhawks arena on game
days was struck down.26
In considering other Category II restrictions besides “buffer
zones,” courts have likewise only protected the most specifically
targeted speech. One such case was the ban on “signs ‘affixed to
any wooden, plastic, or other type of support’ during parades and
public assemblies.”27 In that case, the U.S. Court of Appeals for
the Ninth Circuit correctly recognized that a sign held aloft
during a Nazi parade would reach a particular set of people (the
attendees), a set that would otherwise have been impossible to
reach as a group.28 The court thus readily understood why a
“consistent with the First Amendment”).
23. Id. at 2535.
24. 310 F.3d 1029 (7th Cir. 2002).
25. See id. at 1042 (“His intended audience is Chicago Blackhawks fans.
The most opportune time and place to reach this audience is outside the United
Center, before and after Blackhawks home games.”).
26. See id. (holding that “the peddling ordinance is not a reasonable time,
place, and manner restriction”).
27. Armijo, supra note 1, at 1704 (quoting Edwards v. City of Coeur
D’Alene, 262 F.3d 856, 859–60 (9th Cir. 2001)).
28. See Edwards, 262 F.3d at 867
Because there is no other effective and economical way for an
individual to communicate his or her message to a broad audience
during a parade or public assembly than to attach a handle to his
sign to hoist it high in the air, Section 1(D) of Ordinance 2920
prevents Edwards from reaching his intended audience. We conclude,
therefore, that Ordinance 2920(1)(D) also does not comport with the
third prong of the time, place, and manner test because it does not
allow for ample alternative means of communication.
protester would want his anti-Nazi message to be visible to those
attending the Nazi rally.
In contrast, the U.S. Court of Appeals for the Third Circuit
upheld a ban on Sunday openings of adult stores, on the grounds
that “the statute allows those who choose to hear, view, or
participate publicly in sexually explicit expressive activity more
than thirty-six hundred hours per year to do so.”29 An adult store
almost surely would have reached more total listeners in the
additional 728 hours per year it wished to remain open on
Sundays than Weinberg reached by standing outside Chicago
Blackhawks games. But because no particular group of relevant
people was likely to visit an adult store on Sunday, the store’s
claim was unsuccessful.
What should we take away from these decisions? Courts in
category II cases can certainly be sympathetic to the plight of
those who are prevented from reaching their intended audience.
But for such sympathy to be determinative, courts have
demanded an explanation for why that intended audience is
particularly relevant to the message being expressed. The AAC
doctrine is the vehicle by which courts have upheld Category II
restrictions that do not—in the judge’s view—involve bans on
reaching uniquely relevant audiences. And so to answer the
question of why courts have found the AAC doctrine attractive,
we must understand why they have placed such an emphasis on
the fit between speech and audience.
In the next section, I will focus on that explanation: courts
have implicitly applied a stunted version of what Armijo calls the
“marketplace theory” of speech rights.
IV. Toward a Better Version of the Marketplace Theory
A. The Marketplace Theory as Currently Applied
As Armijo explains, “Marketplace theory defines the First
Amendment’s primary function as facilitating a process by which
truth can be reached.”30 For our purposes, the most important
29. Ben Rich Trading, Inc. v. City of Vineland, 126 F.3d 155, 163 (3d Cir.
30. Armijo, supra note 1, at 1696.
attribute of marketplace theory is that it “is listener-based in its
orientation; it is listeners who are witnesses to the truth-finding
function taking place within the marketplace of ideas.”31 The
marketplace theory is thus fundamentally instrumentalist; it
favors the protection of speech only as a means to the end of a
more informed populace.
From this listener-focused perspective, the actions of judges
in Category II cases makes more sense. If speech is useful
because of its effects on listeners, then we can understand why
judges would have more sympathy for speech that is specifically
targeted to particular listeners (and often urging those listeners
to take a particular action, such as asking Blackhawks fans to
lobby for a change in the team’s mascot).
But this listener-focused approach also has a troubling
consequence. Speech that is targeted to a particular audience has
a better chance of overcoming a Category II restriction than
speech that a speaker simply wishes to broadcast to as many
people as possible. This means that the content of speech is
influencing the level of protection it receives. Such favoritism,
though unintended, is a serious blow to the principle of
contentneutrality toward which First Amendment jurisprudence strives.
B. The Self-Autonomy Theory
Given the negative consequences of the marketplace theory
as currently applied, what should be done to help lead courts in a
better direction? One obvious possibility is to urge a rejection of
the listener-focused marketplace theory entirely.
As Armijo explains, the main alternative to the marketplace
theory is the self-autonomy theory, which focuses on the liberty of
the speaker.32 The self-autonomy theory values freedom of speech
because that freedom allows each individual to exercise “the
ability to think on one’s own, to choose one’s audience, to speak
with that audience, and to express and receive ideas, so as to
31. Id. at 1698.
32. See id. at 1696 (noting “that much of First Amendment scholarship
undertakes as its primary task the decoupling of self-autonomy theory from
marketplace theory, and then arguing over which supplies a better justification
for supporting the freedom of speech”).
achieve that best version of oneself through reason, reflection,
and exchange.”33 Whether the speech convinces an audience—or
even whether it is relevant to an audience—has no bearing on the
question of whether that speech merits protection under the
self-autonomy theory. Instead, “it is the speaker’s choice of
expression that is . . . ‘the crucial factor in justifying protection’ of
Does the self-autonomy theory represent the solution to the
overly deferential approach judges have taken in Category II
cases? Should judges switch to a pure speaker-focused
selfautonomy theory, rejecting all attempts at line-drawing based on
the effects on the listener? Many have made a strong case for
moving to the self-autonomy theory, including Charles Fried, on
whom Armijo principally relies in expounding the case for the
theory.35 Fried declares the self-autonomy theory to be superior to
the instrumentalist marketplace theory by suggesting that
several aspects of our current First Amendment doctrine would
have to be discarded if we truly accepted the marketplace theory.
I do not believe, however, that Fried’s arguments are
determinative in the debate.
First, Fried suggests that under the marketplace theory we
would lose the right not to speak a message with which we
disagree, a right currently protected by West Virginia Board of
Education v. Barnette36 and subsequent cases.37 Fried argues that
there is no logical reason to protect such a right under the
marketplace theory. If we protect speech rights only so that more
ideas can be added to the marketplace, Fried reasons, then there
is no justification for a constitutional right to withhold an idea
from the marketplace.
But this objection is only to one version of the marketplace
theory. Rather than preserving a liberty to add to the
marketplace of ideas, the marketplace theory, more broadly
understood, protects the liberty to affect the marketplace in any
way that we want to. Withholding support for an argument that
one believes to be false can have just as much of a positive effect
on others as adding support to an argument that one believes to
be true. Those who disagree with a view and do not want to be
couriers of that view may feel that the marketplace will be a
better one if fewer join the chorus for that view. Thus, the right to
withhold support for a message is fully compatible with a broad
version of the marketplace theory.
Second, Fried worries that the marketplace theory must lead
to lessened protections for speech that courts deem “low-value,”
such as obscenity.38 But this problem can also be avoided so long
as we take the most expansive view of the marketplace theory. It
is not up to judges but speakers to decide whether their speech
will improve the marketplace of ideas. Thus, all speech should be
protected if that speech is received by anyone else besides the
speaker. It is the transmission of the speech from speaker to
listener that is sufficient to make the speech worthy of protection,
regardless of its content.39
It is not, then, the marketplace theory per se that is
problematic. It is rather a crimped view of the marketplace
theory that puts at risk the protections Fried identifies. Fried’s
critiques do not force us to abandon the marketplace of ideas as
the core justification for protecting free speech.40 Rather, they
38. See id. at 228 (“Civic republicans explain the historic exclusion of
obscenity from constitutional protection on the ground that obscenity does not
contribute to, but rather degrades, public (republican) discourse. Obscenity law
is a puzzle . . . .”).
39. There is an element of self-autonomy in this fuller version of the
marketplace theory in that it can only be the speaker who decides that his
speech will be a meaningful contribution to the marketplace of ideas.
40. There are other reasons to be skeptical of a move to the self-autonomy
theory. Most seriously, the self-autonomy theory, taken to its logical endpoint,
might require First Amendment protection for even Category I speech
restrictions. Suppose a choice of a particular courier service has a sentimental
meaning to a speaker, but no effect whatsoever on a recipient. An
suggest that a broader view of the marketplace theory is
necessary, a view I will explore in the next section.
C. The Marketplace Theory as It Should Be Applied
What is wrong with the marketplace theory as courts have
most often applied it? Courts have confused the justification for
protecting free speech with a judicial test for protecting free
speech. We protect speech because there should be a robust
marketplace of ideas, but that does not mean speech is worth
protecting only if—in the judge’s view—it will make a difference
in the marketplace of ideas.
Courts have provided lessened protections for generalized
speech (such as protests on free trade issues) and perceived
“low-value” speech (such as adult entertainment) based on
whether a judge thinks a regulation seriously alters the
marketplace of ideas. The result has been self-defeating, for the
simple reason that no court can accurately predict how valuable
speech will be, or who will most benefit from hearing it.
The only way to avoid this problem is for courts to treat all
Category II restrictions alike. If a regulation affects the ability of
a speaker “to choose one’s audience, [and] to speak with that
audience,”41 then it is at least a Category II restriction. The
listener-focused theory of the First Amendment is thus fully
compatible with a bright-line, objective judicial test. Rather than
asking how much the audience of a speaker has been affected,
courts must simply shift to asking whether the audience of a
speaker has been affected.
This does not mean, however, that judges will be left with no
meaningful role to play in adjudicating speech restrictions. First,
even if a ban does constitute a Category II (or Category III)
restriction, a court may still uphold that ban if it is absolutely
environmental group, for example, might choose to distribute flyers using only
bicycle couriers. Because such a choice is made in the course of speech and could
be described as “self-actualizing,” the self-autonomy theory might hold that such
a choice requires First Amendment protection. But choices that have absolutely
no effect on listeners are implausible candidates for protection under the First
Amendment, and a theory that might lead to such a result should be viewed
41. Armijo, supra note 1, at 1690.
necessary to a government objective, the test that Professor
Second, courts must still engage in an enquiry to determine
whether a facially Category I restriction actually rises, in
practice, to the level of a Category II restriction. For example, to
return to the hypothetical used in Part II, there is no difference in
the experience of a recipient between receiving a letter by air or
by truck. However, if a speaker is forced to use a less-preferred
method, and that method is also more expensive, the speaker may
not have the ability to send as many such letters as he otherwise
would have. In this situation, the set of recipients has changed,
and an apparent Category I restriction has risen to the level of
V. Why Courts Have Upheld Category III Restrictions
Though we now have a plausible account of why the AAC
doctrine is attractive in Category II cases, one important question
remains: why has the AAC doctrine often been applied even in
Category III cases? These cases, in which not just the audience of
a message but the content of a message itself is altered, represent
the most serious abridgment of free speech rights. They are
situations where, as Armijo describes, a speaker is deprived “of
her chosen mode of communication. The result of that restriction
is that the speaker’s message is never subjected to the truth
process at all.”44 For this reason, no version of the marketplace
theory can justify a Category III restriction. To explain why
42. See id. at 1728
One way to achieve this goal is to apply an incompatibility test: when
a speaker’ s expression is infringed by a law or regulation, a
reviewing court should ask whether the infringed speech act—in the
form the speaker intended to express it—is incompatible with the law
and its purpose. The law will survive as applied to the speaker only if
the speaker's mode is incompatible with the governmental interests
asserted in the law’s support.
43. One court, unfortunately, has upheld a ban that had precisely this
effect. Despite acknowledging the burden it was placing on the speaker, the
court wrote that he was not entitled to the “most cost-effective mode of
communication.” Johnson v. City of Philadelphia, 665 F.3d 486, 494 (3d Cir.
44. Armijo, supra note 1, at 1700.
courts have nonetheless used the AAC doctrine to uphold several
such restrictions, we must look for a culprit beyond the
The simplest explanation is that in some cases, it may be
difficult for courts to determine whether a regulation actually
constitutes a Category III restriction. Consider once again the
facts of O’Brien.45 Suppose that there was no way for anyone in
the audience to have known the difference between a real draft
card and an otherwise identical fake draft card. Suppose
O’Brien’s intended performance included no verbal explanation
one way or the other. Suppose (assuming a counterfactual) that a
prominent Supreme Court case had not brought significant
attention to the question of whether O’Brien would be using a
real or fake draft card. In this situation, it could well be that the
message received by the audience would have been exactly the
same whether the burned draft card had been real or fake. In this
and similar cases, judges can disagree in good faith on whether a
regulation will actually affect the perception of an audience, and
thus whether a regulation falls in Category I or III.
But these borderline cases do not fully explain why courts
have often upheld Category III restrictions. Even when courts
have acknowledged that certain prohibited conduct has
expressive content, they have often minimized the importance of
that conduct. In Clark v. Committee for Nonviolence,46 for
example, the Supreme Court assumed that an overnight
campground near the White House would have expressive
content.47 Nonetheless, the Court upheld a denial of an overnight
permit, dismissing the notion that “without overnight sleeping
the plight of the homeless could not be communicated in other
ways.”48 In this and other cases, the AAC doctrine has been a
vehicle to minimize the importance of prohibited conduct to
45. See United States v. O’Brien, 391 U.S. 367, 369 (1968) (“On the
morning of March 31, 1966, David Paul O’Brien and three companions burned
their Selective Service registration certificates on the steps of the South Boston
46. 468 U.S. 288 (1984).
47. See id. at 293 (“We need not differ with the view of the Court of
Appeals that overnight sleeping in connection with the demonstration is
expressive conduct protected to some extent by the First Amendment.”).
48. Id. at 295.
expression. Why are courts so tempted to engage in this
Here is one possible answer: courts are reluctant to admit
that illegal acts have an inherent and unique power to send a
message. They are reluctant because once this is admitted, we
are confronted with an uncomfortable fact: that enforcing the law
unavoidably puts some messages at a disadvantage. Rather than
making this admission, courts may find it easier to believe a
useful fiction—that every message sent by a disruptive and
illegal act could be made just as effectively by a legal act.
Consider this uncomfortable line of reasoning: The most
effective demonstration of opposition to a law is to publicly
engage in an act of disobeying that law.49 The government
nonetheless has an undeniably compelling interest in enforcing
its laws. Thus, even though the government may not explicitly
discriminate against speech on the basis of its content,50 simply
enforcing a law will necessarily eliminate some of the most
effective speech against that law. Since no comparable obstacles
stand in the way of speech supporting current law—publicly and
conspicuously obeying a law is not only allowed but encouraged—
a viewpoint in favor of the current state of the law will
unavoidably be aided by the government.
I do not necessarily think there is a solution to this problem;
life is difficult in many ways for those who disagree with current
law, and this is one of them. The marketplace of ideas will never
be an entirely fair fight between those opposed to current
regulations and those supporting them, and that is unfortunate.
49. For a recent example, see Alicia Victoria Lozano, Marijuana Activists
Gather for ‘Smoke-In’ at Rittenhouse Square, NBC 10 PHILADELPHIA (Jan. 20,
http://www.nbcphiladelphia.com/news/local/Rittenhouse-SquareMarijuana-Protest-411332455.html (last visited Apr. 24, 2017) (on file with the
Washington and Lee Law Review).
50. Of course, a public act of lawbreaking may include a message
promoting “imminent lawless action” (e.g., encouraging others to immediately
engage in similar lawbreaking), which is not protected by the First Amendment
under current precedent. Brandenburg v. Ohio, 395 U.S. 444, 447 (1969). But
actions can carry more than one message simultaneously. Even if that portion of
a demonstration advocating imminent lawbreaking is unprotected, that portion
of the demonstration advocating a long-term change in the law itself remains
protected as First Amendment expression.
Acknowledging that there is a problem is preferable to the
alternative, however, which is to force oneself to believe that
there is no problem at all. This is the optimistic view implied by
holding that burning a fake draft card is just as effective in
sending an anti-draft message as burning a real one.51 The result
of this optimistic view, ironically, has been a greater acceptance
of the AAC doctrine, and with it even less protection for
protesters than they would otherwise have received.
In this response, I have explained the distinction between the
three categories of restrictions on “channels” of speech, a
distinction that courts and scholars have not made as explicit as
they should. Further, I have suggested why courts have
frequently applied the ample alternative channels doctrine in
Category II cases, as an attempt to improperly devalue speakers’
choices that do not make “enough” of an effect on the marketplace
of ideas. I have suggested how judges could be more consistent in
their treatment of Category II cases while also distinguishing
them from Category I cases, which are implausible candidates for
First Amendment protection. Finally, I have suggested that
courts must be more realistic about the powerful expressive effect
that many disruptive and illegal acts can have. Accepting this
fact does not mean courts must allow widespread lawlessness
(Armijo’s proposed incompatibility rule would still allow many
necessary regulations to stand in the face of First Amendment
claims), but it does mean courts would give full weight to the
speech interests at stake.
Professor Armijo’s powerful and provocative Article is sure to
provoke many responses, as courts consider whether they have
been following the wrong path for more than 40 years. My hope is
that this response can also play a role in that process, helping
courts to look at their own prior reasoning with a fresh eye and
an open mind.
51. Or, more relevant to today’s political climate, it is the view implied by
believing that a “smoke-in” with fake marijuana would be just as effective as
lighting up the real thing.
19. Armijo , supra note 1, at 1702 ( citing Menotti v . City of Seattle, 409 F.3d 1113 ( 9th Cir . 2005 )).
33. Id . at 1690.
34. Id . at 1690-91 (quoting C. EDWIN BAKER , HUMAN LIBERTY AND FREEDOM OF SPEECH 592 ( 1989 )).
35. See id. at 1689-90 (“ Summarizing the moral case for individual choice, Charles Fried writes that '[t]he capacity for judgment, to make plans, to choose one's good, is what we share with other persons'; indeed, this capacity is 'what makes us persons.'” (quoting CHARLES FRIED , MODERN LIBERTY AND THE LIMITS OF GOVERNMENT 56-57 ( 2007 ))).
36. 319 U.S. 624 ( 1943 ).
37. See Charles Fried , The New First Amendment Jurisprudence: A Threat to Liberty , 59 U. CHI. L. REV . 225 , 227 ( 1992 ) (“The real trouble begins when this conception of the First Amendment is pressed further to deny free speech protection to speakers who wish not to pronounce certain views .”).