Clicking Your Speech Rights Away: The Constitutionality of Gagwrap Licenses
I find it indefensible that customers are asked—no, required—to give up
their free speech for the privilege of buying software. Your basic rights
are not something that you can trade away so easily.
The average user who purchases a new software package is usually eager to
install it and begin using it right away. The user tears off the shrinkwrap
and pulls out the disks, ignoring, forgetting about, or throwing away the
included paperwork that usually contains very small print written in dense
legalese. And more likely than not, the user installs the software clicking
through the licensing agreement as quickly as possible, figuring that it
just prohibits things like software piracy. What the user has just done is
to agree to a number of licensing restrictions without even reading
them—restrictions that can be broad and insidious.
"Shrinkwrap" and "clickwrap" licenses have become very common vehicles for
Dictionary defines "shrinkwrap" as "a printed license that is displayed on
the outside of a software package and that advises the buyer that by
opening the package, the buyer becomes legally obligated to abide by the
terms of the license." These license agreements derive their names
either from the shrinkwrap around the package that the user must break to
install the software or from the fact that the user must click through a
Because the details of these licenses usually do not appear on the outside
of software packages, users generally are not able to see the terms of the
license agreement before they purchase the product. As we will see below,
courts have not found this to be problematic in terms of license
enforcement. However, sometimes terms lurk within these agreements that
some users would find troubling or objectionable. This paper addresses one
such license agreement and examines its enforceability in light of both
First Amendment doctrine and contract law. Dubbed "gagwrap" licenses by the
Consumer Project on Technology (CPT), a non-profit organization that
focuses on intellectual property rights and health care, electronic
commerce and competition policy, these licenses contain contract terms
that restrict users' ability to engage in certain kinds of speech.
According to CPT's definition,
Gagwrap Clauses purportedly ban disparagement of a website, service,
product or company through barring use of graphics, logos, trademarks and
other intellectual property to criticize the website, service, product or
company. The gagwrap clauses also occasionally condition certification and
affiliate programs. The conditions may include a stated duty not only to
forbear from disparagement but an affirmative duty to report disparagement.
After a brief look at the license in question, we will turn to an
examination of the enforceability of shrinkwrap licenses by the courts and
of the proposed Uniform Computer Information Transactions Act. We will then
subject the license to First Amendment and contract law analysis.
The License in Question: Microsoft FrontPage 2002
In September 2001, several websites reported a gagwrap clause in
Microsoft's license agreement for FrontPage 2002; other sites later picked
up the story. The end-user license agreement, or EULA, was alleged to
contain the following section:
You may not use the Software in connection with any site that disparages
Microsoft, MSN, MSNBC, Expedia, or their products or services, infringe any
intellectual property or other rights of these parties, violate any state
federal or international law, or promote racism, hatred or pornography.
According to at least some reports, after the uproar these articles caused,
the clause disappeared from the EULA.
Clearly there are implications for free speech in this EULA. After an
examination of the law in this area and some commentators' thoughts on
contracts of silence, this paper will return to an examination of this
particular license agreement in light of both First Amendment and contract
The Constitutionality of Shrinkwrap Licenses: The ProCD Case
The courts were first invited to examine the constitutionality of
shrinkwrap licenses in 1996 in ProCD v. Zeidenberg. Matthew Zeidenberg
purchased a consumer copy of ProCD's product Select Phone, an aggregation
of over 95,000,000 residential and commercial listings. He then made those
listings available over the Internet for commercial purposes, in violation
of the product's shrinkwrap license, which read, in part, "You will not
make the Software or the Listings in whole or in part available to any
other user in any networked or time-shared environment, or transfer the
Listings in whole or in part to any other computer other than the computer
used to access the Listings." ProCD filed suit. The district court held
that because Zeidenberg was not able to read the license before he
purchased the software, let alone bargain or object to its terms, he was
not bound by it.
Upon appeal, the Seventh Circuit reversed. Shrinkwrap licenses, where the
license is contained inside the box, are no different from many other
license arrangements. Drawing analogies to a number of these arrangements,
from radios to drugs, the court found that interpreting shrinkwrap licenses
in the way that Zeidenberg argued would "return transactions to the
horse-and-buggy age." The Seventh Circuit was clear: "Shrinkwrap
licenses are enforceable unless their terms are objectionable on grounds
applicable to contracts in general (for example, if they violate a rule of
positive law, or if they are unconscionable)."
While the Supreme Court has not yet ruled on the constitutionality of
shrinkwrap licenses, the Seventh Circuit's opinion in ProCD has not been
seriously challenged. We now turn to an attempt by the National Conference
of Commissioners on Uniform State Laws (NCCUSL) to address the
enforceability of shrinkwrap licenses on a national, uniform basis.
The Uniform Computer Information Transactions Act (UCITA)
UCITA began as a joint project between NCCUSL and the American Law
Institute (ALI) as an addition to the Uniform Commercial Code, Article 2B,
but after heavy criticism from the government, legal and academic
communities, ALI withdrew its support. However, NCCUSL continued to
support Article 2B; the organization renamed the proposal UCITA and
retained the text from Article 2B almost verbatim.
UCITA is intended to be a uniform approach to software licensing. The scope
and intention of UCITA is outlined in its Prefatory Note:
UCITA is the first uniform contract law designed to deal specifically with
the new information economy. Transactions in computer information involve
different expectations, different industry practices, and different
policies from transactions in goods. For example, in a sale of goods, the
buyer owns what it buys and has exclusive rights in that subject matter
(e.g., the toaster that has been purchased). In contrast, someone that
acquires a copy of computer information may or may not own that copy, but
in any case rarely obtains all rights associated with the information.
…What rights are acquired or withheld depends on what the contract says.
So far only two states, Virginia and Maryland, have adopted UCITA, and it
has been introduced in Delaware, the District of Columbia, and Oklahoma. As
did its predecessor, Article 2B, UCITA has come under heavy criticism. The
American Bar Association failed to approve UCITA in February 2003, and
UCITA has come under fire from consumer, library and technology groups.
However, not surprisingly, software corporations such as Microsoft and IBM
support UCITA as a way to reduce vendor liability costs.
UCITA is interesting because it is one of the early attempts to make
software licensure enforcement uniform. But because UCITA has only been
adopted by two states thus far, we do not need to consider it at length in
this paper. However, if and when more states adopt versions of UCITA, it
will require analysis on its effect on shrinkwrap licenses. We next examine
what other commentators have written on "selling one's silence."
Review of the Literature
Surprisingly little has been written recently on the broad topic of
confidentiality agreements, and much of it has been on the topic made
newsworthy by Jeffrey Wigand: whistleblowing. This section will contain a
brief review of some of the ideas and balancing tests advanced in these
Terry Morehead Dworkin and Elletta Sangrey Callahan examine the
proliferation of secrecy clauses over the past decade and balance them
against the concurrent rise in state whistleblower legislation. However,
they approach the question of the enforceability from the employer's point
of view, suggesting ways in which employers can protect themselves; for
example, employers could craft agreements in which employees would be
required to disclose any damning information internally first. Dworkin
and Callahan note that approaches like this may well limit the amount of
information that gets to the public, but that they are "consistent with the
attempts at the federal level to encourage internal whistleblowing as a
means to curb corporate wrongdoing."
Carol M. Bast, also examining confidentiality agreements in the
whistleblowing context, comes up with a test based on her readings of the
caselaw and contract law that is based on public policy concerns.
The six-part test would require a court to consider the parties' reasonable
expectations, the potential loss to the employer if the confidentiality
agreement is not enforced, the protectability of the information as a trade
secret or proprietary information, substantial adverse effects on third
parties, exacerbation of the adverse effects if the confidentiality
agreement is enforced, and the possibility of limited disclosure.
This test contains a number of elements that are fairly strictly defined
and seem to favor the employer rather than the potential whistleblower.
Several of the elements are concerned with the possible effect on the
employer (potential loss, trade secrets), while none address the effects on
the whistleblower or the public directly.
However, such balancing tests are not uncommon in the literature. Alan E.
Garfield offers a test that focuses more on the free speech elements of a
confidentiality agreement. After examining two extreme cases (a contract to
conceal a crime and a contract to withhold trade secrets), he suggests the
following straightforward guideline:
A court must compare the strength of the public and private interests in
enforcing a contract that suppresses speech, the "confidentiality
interest," with the competing public interest in not having the threat of
contractual liability inhibit speech, the "disclosure interest." If the
disclosure interest clearly outweighs the interest in confidentiality, then
the court should not enforce the contract. The overriding public interest
in the speech in this situation renders the right to sell one's silence
"inalienable." Conversely, if the disclosure interests do not sufficiently
outweigh the confidentiality interests, then the court should enforce the
contract. Here, the right to sell one's silence is "alienable."
The allure of such a simple test is also its potential problem: While
Bast's test attempts to formalize the elements that would make up the
balance between the need for confidentiality and the public right to know,
Garfield's test is more open-ended. Such a test provides more possibilities
for liberal interpretation, but also provides an equal number of
opportunities for restrictive interpretation. The authors of this paper
favor an approach that could be characterized as striking a middle ground
between the tests offered by Bast and Garfield: an attempt to
apply contract law with a heavy dose of First Amendment jurisprudence as a
public policy concern.
Analysis of the FrontPage EULA
As some commentators do above, this section will subject gagwrap EULAs,
such as the Microsoft FrontPage EULA, to a contract analysis. However, a
First Amendment analysis must be added to the contract law to produce a
test that is sufficiently respectful of expressive rights in this context.
The Restatement (Second) of Contracts states in relevant part:
§ 178: When a Term Is Unenforceable on Grounds of Public Policy
(1) A promise or other term of an agreement is unenforceable on grounds of
public policy if legislation provides that it is unenforceable or the
interest in its enforcement is clearly outweighed in the circumstances by a
public policy against the enforcement of such terms.
(2) In weighing the interest in the enforcement of a term, account is taken of
(a) the parties' justified expectations,
(b) any forfeiture that would result if enforcement were denied, and
(c) any special public interest in the enforcement of the particular term.
(3) In weighing a public policy against enforcement of a term, account is
(a) the strength of that policy as manifested by legislation or judicial
(b) the likelihood that a refusal to enforce the term will further that policy,
(c) the seriousness of any misconduct involved and the extent to which it
was deliberate, and
(d) the directness of the connection between that misconduct and the term.
In addition to this test, the Restatement (Second) suggests a way in which
courts can interpret "public policies" when applying the above test.
§ 179: Bases of Public Policies Against Enforcement
A public policy against the enforcement of promises or other terms may be
derived by the court from
(a) legislation relevant to such a policy, or
(b) the need to protect some aspect of the public welfare, as is the case
for the judicial policies against, for example,
(i) restraint of trade (§§ 186-188),
(ii) impairment of family relations (§§ 189-191), and
(iii) interference with other protected interests (§§ 192-196, 356).
An application of the test in Section 178 (3) to the Microsoft FrontPage
EULA might suggest that the contract entered into by the end user is
unenforceable. Let us begin with the question of whether freedom of
expression is an appropriate public policy concern.
Section 179 (b) (iii), which discusses how a public policy should be
interpreted, is a catch-all prohibition against "interference with other
protected issues." One need look no further than the famous "Footnote Four"
to find a protected status for speech, as well as for other rights
contained within the Bill of Rights:
There may be narrower scope for operation of the presumption of
constitutionality when legislation appears on its face to be within a
specific prohibition of the Constitution, such as those of the first ten
Amendments, which are deemed equally specific when held to be embraced
within the Fourteenth.
Many cases adopt a preferred position for freedom of expression concerns,
particularly when the speech is content-based, as is the case in the
FrontPage EULA. In R.A.V. v. City of St. Paul, for instance, a city
ordinance that proscribed certain kinds of hate speech but not others was
held to be constitutionally invalid. The courts must apply strict
scrutiny to any restriction on speech that is content-based. To withstand a
strict scrutiny analysis, the government must show that: (1) it has a
compelling interest; (2) the restriction is necessary to serve that
interest; and (3) the restriction is narrowly tailored to serve that interest.
FrontPage's EULA would be subject to strict scrutiny were it a governmental
regulation. However, it is important to note that not all government
enforcements of contracts implicate state action. This makes sense from
a practical standpoint, since parties are free to contract with each other
to abridge their rights, and they do so all the time. For example, two
friends might sign a contract that if one stops smoking for a period of a
year, the other will pay him $500. If the first party fulfills his part of
the contract (stops smoking for a year), and the second party refuses to
pay him the money, government enforcement of this contract would not
necessarily involve state action because the agreement is between two
private parties and does not involve anyone else.
Nevertheless, the Supreme Court has made it clear that enforcement of some
private contracts does constitute state action, particularly when the
agreement affects not only the private parties involved in the contract.
The lead case in this area is Shelley v. Kraemer, in which property owners
in St. Louis signed an agreement that stipulated that for 50 years, their
properties must be owned only by Caucasians. In striking down this
agreement, the Court said:
These are not cases, as has been suggested, in which the States have merely
abstained from action, leaving private individuals free to impose such
discriminations as they see fit. Rather, these are cases in which the
States have made available to such individuals the full coercive power of
government to deny to petitioners, on the grounds of race or color, the
enjoyment of property rights in premises which petitioners are willing and
financially able to acquire and which the grantors are willing to sell. The
difference between judicial enforcement and non-enforcement of the
restrictive covenants is the difference to petitioners between being denied
rights of property available to other members of the community and being
accorded full enjoyment of those rights on an equal footing.
The Court suggests that rather than the state merely enforcing a contract
voluntarily entered into by private parties, in this case the contract
affects outside parties as well. Non-Caucasians who might want to purchase
property in that area and had the resources to do so were forbidden to do
so without consent of all the existing Caucasian property owners.
We can extend this line of argument into the gagwrap area: it is
reasonable to suggest that even if an end user voluntarily contracted with
Microsoft by accepting the EULA, the rights of the public to hear whatever
Microsoft-restricted speech that user might have wanted to express using
FrontPage have been abridged without its consent.
As far-fetched as this may seem at first glance, there is legislative and
judicial precedent that follows this line of reasoning. The most notable
example is legislative action that protects the activities of
whistleblowers. Jeffrey Wigand's now-famous disclosures against the tobacco
company Brown & Williamson brought whistleblowing into the public eye.
Wigand, a top executive at B&W for four years, was interviewed by Mike
Wallace for 60 Minutes about the dangers he knew that smoking posed. CBS
refused to air the interview after being threatened by B&W for allegedly
interfering with confidentiality agreements that Wigand had signed upon his
According to the California State Personnel Board, whistleblowing is
"disclosing information that a state or community college employee, or
applicant for state or community college employment, reasonably believes is
evidence of an improper governmental activity, or refusing to obey an
illegal order or directive." Forty-one states have passed
whistleblowing statutes. The protections in these statutes vary, as
might be expected. However, the sheer number of these statutes strongly
suggests that states have an interest in protecting those who wish to
expose corporate wrongdoing in violation of their confidentiality
agreements. It should also be noted that whistleblowing has First Amendment
implications for the whistleblower as well. As one commentator writes,
Whistleblowing is a compelling form of self-expression because it is an act
that is critical to the employee-speaker's identity as citizen. Deciding to
speak out for the public's benefit, despite possible recriminations not
only from one's employer but also from co-workers, neighbors, and family
members, presumably for most whistleblowers is among the most intense and
indeed the most politically charged choices the employee is ever likely to
make. An individual who decides to blow the whistle becomes personally
involved in a public issue in a way that finds few counterparts for the
ordinary citizen. This whistleblowing is a form of self-expression that our
constitutional system should not lightly permit to be bargained away.
It is clear from the above analysis that freedom of expression should be
classified as a public policy that could deny the enforceability of a
contract that abridges it. When a user, voluntarily or not, enters into the
EULA for FrontPage 2002, she is limiting her own First Amendment rights to
speak, perhaps in violation of public policy.
The common law requires us to inquire further before declaring such
agreements void on the grounds of public policy. Let us now turn to the
application of the test provided in the Statement (Resecond) of Contracts,
Section 178 (3) (a) and (b) are easy to address from a First Amendment
perspective. First Amendment jurisprudence, as noted above, is clear in its
positioning of freedom of expression rights as superior within the context
of other rights. Similarly, it stands to reason that if a contract against
freedom of expression were unenforceable, more speech could occur.
Subparts (c) and (d) provide more of a challenge. In the case of the
Microsoft FrontPage EULA, no misconduct is alleged, either on the part of
Microsoft or anyone else. We propose to extend the reach of the test to
cover not only misconduct, but other activities as well.
In this case, Microsoft, through the application of its EULA, seeks to
restrict speech activities in certain areas. The Restatement (Second) of
Contracts contains a note that might prove helpful. As a comment to Section
178 (3) (d), the Restatement says, "It is sometimes objectionable to make a
commitment to engage in conduct that is not in itself objectionable. This
is the case, for example, for a promise to vote in a particular way." While
speech rights are not necessarily the equal of voting rights, there is
clearly an expressive element in the right to vote. Freedom of speech is
also accorded an importance in the shaping of America's public and
political discourse that is co-equal with that of the right to vote, or
nearly so. In fact, according to Daniel A. Farber, a well-known commentator
on First Amendment law, political speech is a "double good:" "Information
contained in political speech is one public good, and political
participation, which is often guided by such information, is a second
public good." Farber's broader thesis is that the benefits a community
derives from free speech, not the private motives of speakers, justify
speech's greater protection. This theory fits nicely with the analysis
we offer: the rights of speakers who wish to use FrontPage to criticize
Microsoft are protectable not merely because of the rights of the
individual speaker, but because of the importance of information like that
to a broader community (e.g., the online community).
Accordingly, we can extend the ideas contained in the comment to subparts
(c) and (d). The spirit of the comment suggests that even if there is no
misconduct involved, as with a promise to vote in a particular way, there
might be a case made for limiting the ability to contract away one's right
to speak. If we view subparts (c) and (d) in this light, then the FrontPage
EULA is in danger of unenforceability.
The Seventh Circuit's decision in ProCD, and its general acceptance in the
judicial community, ensures that the central provisions of shrinkwrap and
clickwrap agreements will likely be enforced: Consumers buying software
products such as Microsoft's FrontPage must respect Microsoft's
intellectual property rights; their ability to make and distribute copies
of the software, or to use it on multiple computers, will be strictly
Nevertheless, the enforceability of "gagwrap" licenses is by no means
settled law. The courts recognize certain contractual promises as being
void or unenforceable due to public policy concerns. Because of the
important social value of public speech and criticism, gagwrap licenses may
run afoul of such policies; the treatment of contracts affecting other
important public interests suggests that such provisions may face probing
Enforcement of the software purchaser's promise to forbear from
constitutionally protected speech may also raise concerns under the First
Amendment. This is true particularly where the courts' enforcement affects
the rights of third parties to hear the purchaser's speech.
These views may well require an extension of existing First Amendment and
contract jurisprudence, but gagwrap licenses themselves should be subject
to unrestricted public debate and criticism, especially if public debate
and criticism are what software manufacturers are seeking to avoid.
Garfield, in another article, makes a related point that should be chilling
to those interested in keeping speech free in all forms of media: He
suggests that the case of Cohen v. Cowles Media, in which a law of
general applicability was enforced against the press, has opened the door
for parties to use contract and property law to "cordon off information
from public purview without First Amendment scrutiny." He further warns:
Efforts to control the flow of information through contract and property
rights are likely to become more prevalent in a world in which property
rights in information are paramount and signing onto an adhesion contract
is only a click away. The First Amendment can be either a neutral bystander
to this process or an active regulator. If the marketplace of ideas is to
stay healthy and vibrant, only the latter approach would seem to suffice.
We join Garfield in both his concern and his call for caution. As we move
deeper into an age where information is the commodity of choice, the First
Amendment must be more vigilant than ever to protect the rights of both the
individual and the community against corporate infringement.
Clicking Your Speech Rights Away:
The Constitutionality of Gagwrap Licenses
A paper submitted for consideration for presentation to
the Law Division of AEJMC at its 2003 annual conference
July 30-Aug. 2, 2003, Kansas City, MO
Genelle I. Belmas, Ph.D.
Assistant Professor of Journalism
California State University, Long Beach
1250 Bellflower Blvd., SSPA 024
Long Beach, CA 90840
[log in to unmask]
(Please direct all correspondence to this author)
Brian N. Larson, J.D.
Attorney at Law
2249 East 38th Street
Minneapolis, MN 55407
[log in to unmask]
AV requirements: Overhead projector
Clicking Your Speech Rights Away:
The Constitutionality of Gagwrap Licenses
While the courts have recognized the constitutionality of "shrinkwrap"
licenses to protect the rights of software producers, an insidious clause
has crept into some of these licenses. Dubbed "gagwrap clauses," these
authors believe are invalid. This paper examines one such clause and
subjects it to a combination of contract and First Amendment jurisprudence,
providing a possible test for the constitutionality of gagwrap clauses.
 Carlton Vogt, "Hiding behind EULA's skirts, an ethically challenged
policy." InfoWorld (17 October 2001).
Last visited March 23, 2003.
 Black's Law Dictionary (7th ed. 1999).
 "The Consumer Project on Technology was started by Ralph Nader in
1995. Our work is documented extensively on the CPTech web page. Currently
CPTech is focusing on intellectual property rights and health care,
electronic commerce (very broadly defined) and competition policy."
http://www.cptech.org/about.html, visited March 31, 2003 (bold in original).
 "Gagwrap Clauses in Current Shrinkwrap and Clickwrap Contracts,"
http://www.cptech.org/ecom/ucita/licenses/gagwrap.html, visited Feb. 26,
2003. The site contains a number of examples of gagwrap clauses from
several well-known companies, including Carfax, Macromedia, and Microsoft.
 See Ed Foster, "A punitive puppeteer?" InfoWorld, Sept. 4, 2001,
visited March 23, 2003; and Ed Foster, "Control with fine print,"
InfoWorld, Sept. 28, 2001,
visited March 23, 2003. These seem to be the stories that most other online
articles reference; see also Russell Pavlicek, "EULA, EULA, Where's the
Moola?" http://www.open-mag.com/features/Vol_27/EULA/eula.htm, visited Feb.
26, 2003; Marshall Roch, "Free Speech vs. Microsoft FrontPage," out2teach,
http://www.out2teach.com/articles/article.php?article_id=33, visited Feb.
21, 2003; Donna Longenecker, "Open computing urged," University at Buffalo
Reporter, April 18, 2002,
http://www.buffalo.edu/reporter/vol33/vol33n25/n5.html, visited Feb. 26, 2003.
 This is the verbiage that was reported in several articles, but the
authors were unable to find a copy of the actual FrontPage 2002 agreement.
However, a document entitled "Microsoft Licensing Product Use Rights" dated
April 1, 2001, contains the exact wording. "Microsoft Licensing Product Use
Rights," EMEA English April 1, 2001, p. 2 (in possession of the authors).
 See Pavlicek, supra note 6.
 ProCD v. Zeidenberg, 908 F. Supp. 640, 645 (WD Wisc. 1996).
 Id. at 655.
 ProCD v. Zeidenberg, 86 F. 3d 1447, 1452 (7CA 1996).
 Id. at 1449.
 Cheon-Seok Seo, "Licenses and the Uniform Computer Information
Transactions Act," 1 Buff. Intell. Prop. L.J. 146, 154-155 (2001).
 Uniform Computer Information Transactions Act (UCITA), "Prefatory
Note," http://www.law.upenn.edu/bll/ulc/ucita/ucita01.htm, visited April 1,
 Grant Gross, "UCITA hits snag with lawyer group," ITworld.com,
http://www.itworld.com/Man/2681/030212ucita/, visited Feb. 26, 2003. But
see Sean F. Crotty, "The How and Why of Shrinkwrap License Validation Under
the Uniform Computer Information Transactions Act," 33 Rutgers L. J. 745,
766 (2002): "I think that the objections to this innovation are misguided
and unrealistic because they lose sight of the market realities that serve
as the focus of a proper uniform commercial statute. As a uniform
commercial statute, the UCITA's policies and provisions are meant to
reflect the norms of the community that it will serve—the business community."
 Gross, id.
 Terry Morehead Dworkin and Elletta Sangrey Callahan, "Buying
Silence," 36 Am. Bus. L.J. 151, 190 (1998).
 Id. at 191.
 Carol M. Bast, "At What Price Silence: Are Confidentiality Agreements
Enforceable?" 25 Wm. Mitchell L. Rev. 627, 714 (1999).
 Alan E. Garfield, "Promises of Silence: Contract Law and Freedom of
Speech," 83 Cornell L. Rev. 261, 315 (1998).
 United States v. Carolene Products Co., 304 U.S. 144 (1938), note 4.
 505 U.S. 377 (1992).
 For example, in a Washington Court of Appeals case, State v. Noah,
state enforcement of a settlement agreement between two private parties was
not held to be state action for First Amendment purposes: "For the
existence of a First Amendment violation, state action is required. State
enforcement of a contract between two private parties is not state action,
even where one party's free speech rights are restricted by that
agreement." 103 Wh. App. 29, 50 (2000).
 334 U.S. 1, 4 (1948).
 Id. at 19 (emphasis added).
 See, e.g., Brian Stryker Weinstein, "In Defense of Jeffrey Wigand: A
First Amendment Challenge to the Enforcement of Employee Confidentiality
Agreements Against Whistleblowers," 49 S.C.L. Rev. 129 (1997).
 California State Personal Board, "Questions and Answers about
Whistleblowing Complaints," http://www.spb.ca.gov/svcgen/wbfaq.htm, visited
March 31, 2003 (emphasis in original).
 State whistleblowing statutes include: Alaska Stat. 39.90.100 to
39.90.150 (Michie 1992 & Supp. 1994); Ariz. Rev. Stat. Ann. 38-531 to
38-534 (West Supp. 1994); Cal. Lab. Code 1102.5 to 1106 (West 1995); Colo.
Rev. Stat. Ann. 24-50.5-101 to 24-50.5-107 (West 1994); Conn. Gen. Stat.
Ann. 31-51m (West Supp. 1994); Del. Code Ann. tit. 29, 5115 (Michie 1994);
Fla. Stat. Ann. 112.3187 to 112.3195 (West 1994); Ga. Code Ann. 45-1-4
(Michie 1994); Haw. Rev. Stat. Ann. 378-61 to 378-69 (Michie 1994 & Supp.
1996); Ill. Stat. Ann. 395/0.01 to 395/1 (Smith-Hurd 1995); Ind. Code Ann.
20-12-1-8, 4-15-10-4 (West 1991 & Supp. 1992); Iowa Code Ann. 79A.28 to
79A.29 (West 1994); Kan. Stat. Ann. 75-2973 (1993); Ky. Rev. Stat. Ann.
61.101 to 61.103 (Michie 1994); La. Rev. Stat. Ann. 30:2027 (West 1994);
Me. Rev. Stat. Ann. tit. 26, 831 to 840 (West 1994 & Supp. 1996); Md. Code
Ann., Com. Law II, 3-301 to 3-310 (Michie 1994); Mass. Gen. Laws Ann. ch.
149, 185 (West 1994); Mich. Comp. Laws Ann. 15.361 to 15.369 (West 1994);
Minn. Stat. 181.931 to 181.937 (1998); Mo. Ann. Stat. 105.055 (West 1994);
Mont. Code Ann. 39-2-901 to 39-2-915 (1993); Nev. Rev. Stat. Ann. 281.611
to 281.671 (Michie 1993); N.H. Rev. Stat. Ann. 275-E:1 to 275-E:7 (Michie
Supp. 1992); N.J. Stat. Ann. 34:19-1 to 34:19-8 (West 1994 & Supp. 1997);
N.M. Stat. Ann. 50-9-25 (Michie 1994); N.Y. Lab. Law 740 (McKinney 1994);
N.C. Gen. Stat. 126-84 to 126-88 (Michie 1992 & Supp. 1994); Ohio Rev. Code
Ann. 4413.51 to 4413.53 (West 1995); Okla. Stat. Ann. tit. 74, 840-1.1 to
840-1.4 (West 1994); Ore. Rev. Stat. 659.505 to 659.545 (1994); Pa. Stat.
Ann. tit. 43, 1421 to 1428 (West 1994); R.I. Gen. Laws 28-50-1 to 28-50-9
(Michie 1994); S.C. Code Ann. 8-27-10 to 8-27-50 (Law. Co-op. 1994); S.D.
Codified Laws Ann. 3-6A-52, 60-11-17 to 60-11-17.1 (Michie 1990 & Supp.
1994); Tenn. Code Ann. 50-1-304 (Michie 1991); Tex. Rev. Civ. Stat. Ann.
art.5, 554.001 to 554.010 (West Supp. 1995); Utah Code Ann. 67-21-1 to
67-21-9 (Michie 1994); Wash. Rev. Code Ann. 42.40.010-.900, 42.41.010 (West
1994); W. Va. Code 6C-1-1 to 6C-1-8 (Michie 1994); Wis. Stat. Ann. 230.80
to 230.89 (West 1994).
 Stryker, supra note 25, at 147-148.
 Daniel A. Farber, "Free Speech Without Romance: Public Choice and the
First Amendment," 105 Harv. L. Rev. 554, 563 (1991).
 Id. at 583.
 501 U.S. 663 (1991).
 Alan E. Garfield, "The Mischief of Cohen v. Cowles Media Co.," 35 Ga.
L. Rev. 1087, 1089 (2001).
 Id. at 1126-27.