2012] MUST LICENSES BE CONTRACTS? 109
shrink-wrap, click-wrap, and browse-wrap cases generally involve,
and rely on, some act by the licensee that can be deemed to constitute
assent to the contract.
17
Some of the commentary has also touched on
the property-contract issue that is central to the discussion here,
18
but that issue has not been the primary focus.
This Article argues that the weight of judicial authority and sound
policy support a contractual approach to license formation.
19
In the
courts, despite the uncertainty just described, no case has enforced a
clearly noncontractual intellectual property license restriction.
20
From a policy perspective, a requirement that license restrictions be
of some sort is identified, the next stage of the analysis focuses on whether
contractual enforcement of the obligation thus undertaken is or should be
preempted by federal law.
Madison, supra note 4, at 296.
17. Where that is not true, as in some browse-wrap cases, the courts have been more
cautious in their acceptance of the licenses as contracts. See Lemley, Terms of Use, supra
note 15, at 472-77.
18. See id. at 470-73; Madison, supra note 4, at 279-80.
19. The focus here is license formation, not enforcement. That is, the argument is that
restrictive licenses must be formed with contractual formalities. Once formed, however, the
breach of a license could give rise not only to contractual remedies, but also to property
remedies provided by patent and copyright law. This Article does not enter into the debate
over which set of remedies is more appropriate for intellectual property. Cf., e.g., Mark A.
Lemley, Contracting Around Liability Rules, 100 C
ALIF.L.REV. 463 (2012); Mark A.
Lemley & Philip J. Weiser, Should Property or Liability Rules Govern Information?, 85
T
EX.L.REV. 783 (2007); David McGowan, The Tory Anarchism of F/OSS Licensing, 78 U.
C
HI.L.REV. 207 (2011); Robert P. Merges, Contracting into Liability Rules: Intellectual
Property Rights and Collective Rights Organizations, 84 C
ALIF.L.REV. 1293 (1996).
20. Even where courts have enforced license restrictions without explicitly requiring
the existence of a contract, they have pointed to contract-like considerations, suggesting
that they believed the existence of a contract was important. See infra text accompanying
notes 139-43.
In a recent trademark case, the Supreme Court described a unilateral “Covenant
Not to Sue” as “unconditional and irrevocable.” Already, LLC v. Nike, Inc., No. 11-982, slip
op. at 1-2, 6 (U.S. Jan. 9, 2013). The Court did not make clear, though, how such a
covenant becomes binding. The Court simply cited a case involving judicial estoppel in
stating that Nike, “having taken the position in court that there is no prospect of such [an
infringing product], would be hard pressed to assert the contrary down the road.” Id. at 7
(citing New Hampshire v. Maine, 532 U.S. 742, 749 (2001)). Judicial estoppel might indeed
have such an effect in the courtroom, but the Court went further, saying that “[b]eyond
simply prohibiting Nike from filing suit, [the covenant] prohibits Nike from making any
claim or any demand.” Id. at 6. Despite this observation, it is not at all clear how, or indeed
whether, any such extra-judicial effect is created by a unilateral covenant. Cf. Already,
LLC v. Nike, Inc., No. 11-982, slip op. at 4 (U.S. Jan. 9, 2013) (Kennedy, J., concurring)
(referring to the future effect of the covenant, but only in the context of “any future
trademark proceeding”).
In any event, the covenant in Already v. Nike, even if unilateral, was an
undertaking by Nike, the party whose conduct was purportedly restricted by it. The focus
of this Article is on claims that intellectual property owners’ unilateral acts have binding
effects on others. Treating a non-contractual commitment as binding on the one who makes
it, through estoppel or a similar theory, is quite different from allowing one party’s
unilateral acts to bind others. See infra text accompanying notes 128-29 (describing
revocability of non-contractual copyright licenses).