Michigan Law Review
Copyright (c) 2000 Michigan Law Review
May, 2000; 98 Mich. L. Rev. 1916
CLIMBING THE WALLS OF YOUR ELECTRONIC CAGE
[Code: And Other Laws of Cyberspace. By Lawrence Lessig.
New York: Basic Books. 1999. Pp. xii, 297. Cloth, $ 30; paper, $ 15.]
(Assistant Professor of Law, Vanderbilt University
School of Law. J.D., Yale; M.A. (Public Policy), University of Chicago; Ph.D.
(Philosophy), University of Illinois at Chicago - Ed. I am grateful to Lisa
Bressman, Robert Brewer, John Goldberg, Ryan Raforth, Bob Rasmussen, Don
Welch, Chris Yoo, and Nick Zeppos for comments on an earlier draft and grateful
to Robert Brewer and Janet Hirt for their expert research assistance.)
[*1916] Space. The final frontier. Not so, say the doyennes of the
first-generation Internet community, who view themselves as the new frontiersmen
and women staking out a previously unexplored territory - cyberspace. Numerous
metaphors in the Internet literature picture cyberspace as a new, previously
unexplored domain. Parallels are frequently drawn to the American colonies,
the Western frontier, or outer space. In Code: And Other Laws of Cyberspace,
Lawrence Lessig n1 says, "Cyberspace is a place. People live there."
n2 In this place, we will build a "new society" (p. 4). A sense of this background
is helpful in appraising Lessig's claims. [*1917] He argues that "we"
need a "constitution" for cyberspace. n3 This seems reasonable, a new
social compact for a new society. n4
While Lessig has his legal training in the U.S. system, as a former law clerk
to Justice Antonin Scalia and a recognized American constitutional law scholar,
in Code, he uses the word "constitution" in its British rather than its American
sense. n5 For the British, a constitution is an unwritten common understanding
about fundamental social values and social practices that merits institutional
protection from the vicissitudes of ordinary politics. n6 The purview
of Lessig's project, then, is constitutional theory understood as the theory
of social order, a broader inquiry than the top-down, text-based American
constitutional theory. n7
[*1918] Lessig implores us to begin the search for a way of life in
cyberspace that protects "fundamental values" (p. 6). Either we do so very
soon, he insists, or we risk locking ourselves into an architecture of computer
code that will destroy liberty, n8 as a by-product of promoting the
interests of global electronic commerce. n9 Given the libertarian leanings
of the Internet community, n10 it is ironic, Lessig observes, that
the forces of the market, Adam Smith's invisible hand, will wield the hammer.
n11 After all, it is the libertarian creed that markets create liberty, not
destroy it. According to Lessig, the overly zealous commitment to libertarianism
on the part of the first-generation community blinds them to this threat,
While Lessig paints a foreboding picture of the dark clouds of oppression
gathering on the online horizon, he notes as well that a brighter future
is possible. n12 Code is by nature mutable and may be used to secure
a constitutional structure for cyberspace that promotes political freedom.
Lessig contends that we as a society have yet to realize that a choice must
be made with regard to the degree of liberty we want in cyberspace (pp. 6-7).
The goal is to choose - from among all the possible cyberspaces - the one
with an architectural code that best promises to support liberty and other
fundamental values we choose to import into cyberspace (p. 6).
Lessig goes so far as to say "code is law" (pp. 6, 59). Taken at face value,
this is an extraordinary claim, given the dominance of pos [*1919]
tivism in modern jurisprudence. n13 Whatever Lessig's overall jurisprudence
of cyberspace, one thing is certain; conceiving of code as law makes the
choice of code political. Indeed, for Lessig, as for his intellectual forebears,
code is quintessentially political. n14 One of the book's most important
contributions is that it raises the basic and crucial proposition regarding
the normativity of code to a new level of sophistication, demonstrating the
applicability of the thesis to issues of privacy, speech, and other core
constitutional values (pp. 109-209).
Lessig has written the first book devoted to the political theory of computer
code. "Code," as the term is used by Lessig, refers, however, both to computer
code - the code written by programmers - and to legal code - the code written
by legislators. One of the book's leitmotifs is a comparison between the
properties of computer code and those of legal code, or, as Lessig quips,
"West Coast code" versus "East Coast code" (p. 53). The overarching similarity
is that both regulate human behavior.
The regulation of human behavior, "regulability," is a second key topic of
the book. Regulability refers to the "capacity of a government to regulate
behavior within its proper reach." n15 On the account Lessig develops,
there are four important regulators of behavior: law, norms, architecture,
and markets (pp. 87-89). Lessig argues that in cyberspace, network computer
architecture is a "newly powerful regulator" of human behavior (p. 86). Nevertheless,
a full account of the social order of cyberspace requires an examination
of the interplay of all four regulatory forces.
Code's main normative thesis is that we must resist the migration toward
a more regulable Internet. More specifically, Lessig argues for a "commons"
in the key architectural code of cyberspace (p. 8). This commons will result
if the application layer of the Internet is dominated by open source code
(p. 100). Open source code is to be understood by contrast with closed source
code. Simply understood, open code reveals its source and closed code does
not. Because open code [*1920] carries its source code along
with the object code, modifications are more easily possible. Lessig is an
unabashed proponent of the open source code movement, which he sees as revolutionary
(p. 8). Open code will make cyberspace less regulable because it puts coding
in the hands of more people, and, thus, code will be less subject to centralized
control. n16 Lessig conceptualizes the situation in constitutional
terms. The citizenry will be safe from the tyranny of government only if
the awesome potential for power held by code is distributed broadly among
the programming world rather than concentrated in a small number of hands.
Lessig's overall argument is spread throughout the book's seventeen chapters,
which are divided into four parts. The main arguments of Code, which will
be the focus of this Review, are set out in Parts 1 and 2, entitled "Regulability,"
and "Code and Other Regulators." Part 3, entitled "Applications," applies
the arguments of Parts 1 and 2 to four important issues: intellectual property,
privacy, free speech, and sovereignty. Part 4, entitled "Responses," outlines
responses to actual and hypothetical objections to the book's main arguments.
Chapter One (as well as the Preface) provides a straightforward overview
of the book. In Chapter Two, Lessig introduces the book's four themes and
illustrates them by means of four stories; the themes are 1) Regulability,
2) Regulation by Code, 3) Competing Sovereigns, and 4) Latent Ambiguity.
n18 Lessig explains, "my aim in the balance of this book is to work through
the issues raised by these four themes" (p. 19). As Lessig says, he uses
the four themes to "understand cyberspace as it is" and as "[he] believes
it is becoming" (p. 23).
Below, I first set out and evaluate the thirteen propositions (positive and
normative) which I argue provide a logical foundation to Lessig's main themes.
Since propositions are more specific than Lessig's "themes" (propositions
have truth values, themes do not), this analytic device will facilitate clarity
in the evaluation of the themes.
[*1921] Basic Propositional Structure of Lessig's Argument
The following set of propositions is meant to capture the interconnecting
structure of the core arguments implicit in Lessig's four themes. In reading
the propositions in the following summary form, note that they fall into
a logical sequence. This sequence is implicit in the structure of Code. Lessig's
themes "describe," rather than state propositions. Hence, they do not have
truth values and consequentially lack logical connectivity. n19
1. Net95 was unregulable. n20
2. Libertarians believe the Internet is unregulable by nature.
3. The Internet is regulable.
4. Libertarians hold a faulty conception regarding the nature of the Internet.
5. Code is the most important regulator of the Internet.
6. The Internet embodies values.
7. Liberty is an important value that ought to be respected and promoted.
8. There is an inverse correlation between regulability and liberty.
9. The Internet is diminishing in its capacity to promote liberty.
10. The trend toward increased regulability of the Internet ought to be reversed.
11. Open source code is less regulable than closed (proprietary) source code.
12. Open source code ought to be promoted.
13. We Need An Internet Constitution
The three statements above that are italicized (numbers 7, 10, and 12) are
"normative" - that is, "ought" statements, not "is" statements. Hume's Law
is sometimes stated as: An ought cannot be derived from [*1922]
an is. n21 The proper conception of Hume's Law, however, is that an
ought statement cannot be derived merely from an is statement. Lessig does
not explicitly characterize his leading assertions as either positive or
normative. As we work through the set of propositions that embody his argument,
however, it will be important to pay attention to the connections he draws
between the "ises" and the "oughts."
In the following discussion, I evaluate, independently and then in conjunction
with each other, each of the thirteen propositions that together constitute
the core arguments of the book.
1. Net95 was Unregulable
Lessig discusses what he refers to as "Net95," which is what the Internet
was like, circa 1995. n22 The most significant feature of Net95 was
that it was a world that could not be controlled. It is this Internet that
libertarians have in mind. Lessig sets out three structural elements that
were conspicuously absent in Net95. These are "credentials," "labels" and
"zones" (p. 28). As Lessig notes, these elements can either be seen as "features"
or "imperfections," depending on whether one favors increased regulability
of the Internet (p. 27).
Net95 lacked information about users' identities (p. 28). By identification,
Lessig means all the true facts about a person, such as "your name, your
sex, where you live, what your education is, your driver's license number..."
(pp. 30-31). "Authentication" is the process by which aspects of your identity
become known (p. 31). Net95 did not have an architecture that allowed people
to authenticate personal facts that are not self-authenticating. n23
Some personal facts are self-authenticating. Lessig argues that in the physical
world, many facts are self-authenticating or easily authenticated. In cyberspace,
however, at least with regard to Net95, personal facts are not self-authenticating
(pp. 32-33). For facts that are not self-authenticating, [*1923]
Lessig argues that new architectures are quickly evolving on the Internet
that will make them capable of authentication. These are "credentials." Lessig
examines three forms of credentialing currently found on the Internet, which
are "passwords," "cookies," and "digital certificates." n24
A second feature that defined Net95 was an absence of labels. While it is
true that the packets of data that sail across the Internet are labeled in
the sense of having an Internet Protocol (IP) address, n25 beyond that,
they could contain anything at all. Under Net95, there was no system for
obtaining verifiable information about the data on the Net because such data
traveled unlabeled. n26
The third core feature of Net95 - the absence of zones - tied the first two
together. Because there was no simple way either to know who someone was
or to classify data, there was no simple way to make access to data depend
on who the user was, or on the data to which she or he wanted access. As
Lessig says, under Net95, there was "no simple way to zone cyberspace."
These three features in combination made Net95 largely unregulable, a dream
telecosm by libertarian lights. Lessig accounts for this lack of regulability
as an unintended consequence of the early exigencies [*1924]
of the largely research-oriented academic community that developed the Internet.
2. Libertarians Believe the Internet is Unregulable by Nature
Chapter Three is entitled, "Is-isms," which, according to Lessig, is the
fallacy of thinking that because something is a certain way, it must stay
that way (p. 25). According to Lessig, this fallacy is widespread among,
indeed characteristic of, the Internet community when it comes to their belief
in the unregulability of the Internet (p. 4). Lessig compares what he calls
the "libertarian utopianism" of the Net to what he found in Eastern Europe
in the early 1990s. n29 This community sees the Internet as an unregulated
space, and implicitly thinks its destiny is to remain that way.
3. The Internet is Regulable
As mentioned earlier, Lessig's first theme is "regulability." As the form
of cognate chosen suggests, the book provides a discussion of "regulability"
understood so as to be applicable to systems. With regard to the Internet,
Lessig argues that while the Internet was unregulable in 1995, it is advancing
rapidly toward a state of greater regulability. Lessig explores in detail
the great extent to which each of the three features missing from Net95 can
be laid onto the architecture of the Internet as it currently exists. Lessig
argues convincingly that it will be in the interest of electronic commerce
to code in all three of these features, credentials, labels, and zones, in
order to create more sophisticated marketplaces in cyberspace (p. 42).
For the lay libertarians who peopled early cyberspace, the absence of these
three features made for a free and open electronic space in which people
could quietly come and go as they pleased. In this early period, the Internet
was noncommercial and without markets. For the assembling titans of wired
capitalism, however, the lack of credentials, labels, and zones are serious
obstacles to their business models and accordingly must be overcome. Lessig
argues that it is through technologies of identification and authentication
that users will be credentialed, and data credentialed and labeled, such
that cyberspace will be [*1925] capable of being effectively
zoned (p. 57). This process is already under way such that cyberspace becomes
more regulable every day. n30
Computer code provides the means to make the Internet more regulable, and
electronic commerce provides the incentive (p. 42). Lessig argues that the
commercial development of the Internet is creating the main pressure pushing
toward greater regulability. The above three features of Net95 each facilitate
electronic commerce. n31
4. Libertarians Hold a Faulty Conception Regarding the Nature of the Internet
As just discussed, Lessig demonstrates that while it is true that the Internet
was unregulable in 1995, by the time he was completing the book in 1999,
the Net had become substantially more regulable (pp. 43-60). Libertarians,
however, think that the Net is unregulable by nature (p. 5). If Net95 was
unregulable and Net99 is becoming regulable, then obviously the Net does
not have an essential nature when it comes to regulability. Libertarians
who think otherwise are simply wrong.
To evaluate the conclusion that Lessig draws in Proposition Four (that the
libertarians are wrong regarding regulability), it is necessary to evaluate
each of the premises - propositions one through three. Proposition one, that
Net95 was unregulable, is in need of qualification. Lessig defines regulability
as the "capacity of a government to regulate behavior within its proper reach"
(p. 19). Note that this definition makes no requirement that regulability
come by means of computer code for advanced methods of identification and
authentication of the sort described by Lessig. In other words, regulability,
as generically understood, allows scope for low-tech means of regulability
as well as high-tech means of regulability.
Title II of the Digital Millennium Copyright Act ("DMCA") provides a low
tech means of increasing regulability. n32 It creates safe harbors
from copyright liability for Internet Service Providers (ISPs). Because ISPs
play a leading role with regard to the Internet, creating safe harbors for
many of their activities is an important instance of statutory regulation
of the Internet. Yet, the means by which the safe [*1926] harbor
provisions work is low-tech, requiring none of the sophisticated architectural
devices discussed by Lessig. Thus, low-tech means of Internet regulation
may produce dramatic results in terms of regulability. n33 Low-tech
means of regulability of this sort were available under Net95. Hence, it
is not the case that Net95 was unregulable. Proposition one is incorrect.
Proposition two, which holds that libertarians think that the Net is unregulable,
is also in need of qualification. When Lessig talks about libertarians, he
is generally not referring to academic libertarians but rather to the large
and influential cohort within the Internet community who are libertarian
in their political views regarding the Internet. n34 With regard to
this lay libertarian political community, Proposition two is correct. Lessig
is right that it has been, and remains, common for Internet cognoscente to
state that the Internet is not capable of regulation. And many others, who
do not go quite this far, nevertheless maintain that government regulation
of the Internet, while perhaps not impossible, is nevertheless a bad idea.
With regard to academic cyber-libertarianism, however, the picture is more
scrambled. David Post and David Johnson hold the view that Lessig criticizes.
n35 Richard Epstein, however, explicitly distinguishes [*1927]
his flavor of libertarianism from Lessig's characterization. n36 We
see then that Proposition two is true for the lay-libertarian community but
only partially true for the academic-libertarian community. n37
As already stated, Proposition three is correct, the Net is indeed becoming
more regulable. The question then is whether the conclusion stated in Proposition
four follows from Propositions one through three, once Propositions one and
two have been qualified. The answer is yes. While Net95 was not unregulable
in the broad sense indicated by Lessig's definition of regulability, it was
nevertheless difficult to regulate (or regulable only in a perhaps less effective
low-tech sense), so Proposition one is close to true. And most, if not all,
libertarians mistakenly thought the Net's unregulability was an immutable
feature. So Proposition two is substantially true as well. Since Proposition
three is correct in holding that the Internet is regulable, it follows that
libertarians have an incorrect conception regarding the regulability of the
The conclusion Lessig draws in Proposition four is both logically valid and
true. Lessig is not interested in making this point for its own sake. His
larger concern is to argue that the common libertarian misconception is dangerous,
as blindness to the problem entails inattention to the problem. Lessig writes,
"Let the Net take care of itself," is the slogan of our generation - and
the current administration. But if we do, then the Net will become something
very different from what it is just now. That's the argument of my book.
Not an attack on ivory-tower Libertarianism. Nor even an attack [*1928]
on "sensible libertarianism." But an attack on a certain do-nothingness that
pervades our present political culture. n38
Lessig's argument and conclusion here are important.
5. Code is the Most Important Regulator of the Internet
The second "theme" of the book is "regulation by code." Lessig states that
this second theme should be considered in conjunction with Theme one, "regulability."
In Lessig's words, "the regulability described by the first theme depends
on the code described in the second" (p. 20; emphasis omitted). Lessig nestles
this account of regulation by code within a more general account of social
Lessig identifies four main forces affecting social regulation: law, markets,
norms, and architecture (p. 88). Lessig says we should think of each as a
distinct modality of regulation (p. 88). What the modalities have in common
is that each serves as a "constraint" on behavior (p. 88). Laws constrain
behavior by making certain activities illegal (p. 89). Markets constrain
by making certain behaviors more expensive (p. 89). Norms constrain by making
certain behaviors subject to informal yet often potent social sanctions (p.
235). Finally, code (architecture) constrains behavior by creating architectural
structures that constrain behavior (pp. 89-90).
In cyberspace, code is the most important regulator (p. 86). As Lessig says,
in the late twentieth century and into the twenty-first, it is code that
should be our concern (p. 86). While the role of code is salient as never
before, it is not Lessig's argument that we focus exclusively on code. Rather,
we need to implement a more general understanding of how regulation works,
one that accounts for the increasingly important role played by code.
Code is an important regulator in cyberspace for the obvious but profound
reason that code is what provides the raw material out of which cyberspace
is built. Cyberspace simply would not exist without this code. Hume famously
refers to causation as the cement of the universe. n41 Picking up on
this, Jon Elster refers to social norms as the cement of society. n42
For Lessig, code is the cement of cyberspace. In [*1929] fact,
as Lessig reminds us, there is not one cyberspace possible but many, and
the constitutive difference is code. n43
6. The Internet Embodies Values
According to Lessig, not only does code provide the raw materials, it also
determines the normative character of cyberspace. One of the most important
features of architectural code is that it is not value neutral. To the contrary,
different architectures promote different values. Lessig writes, "what distinguishes
different parts of cyberspace are the differences in the regulations effected
through code. In some places life is fairly free, in other places controlled,
and the difference between them is simply a difference in the architectures
of control - that is, a difference in code" (p. 20).
One of the distinctive features of architectural code as a regulator is that
it may promote or stifle values in a relatively invisible manner. Just as
Robert Moses could use the height of highway overpasses to keep city buses
(and, therefore, African Americans) away from the beaches of Long Island,
so too, code writers can create architectures that have significant policy
implications in peoples' lives. n44 Lessig discusses a number of different
sub-regions of cyberspace in order to show how values may be implicit in
different architectural structures.
Lessig compares the computer networks at the University of Chicago and Harvard
University. Chicago's network was explicitly chosen to promote First Amendment
free speech values. The network designers asked the Provost if the network
should build in identification architecture (p. 26). The Provost, First Amendment
scholar Geoffrey Stone, chose instead to promote free speech values by building
anonymity into the system. n45 By contrast, Harvard chose to disallow
anonymity by requiring that members of the university community register
their machines. n46
[*1930] Lessig pays the most attention to the space owned by America
Online ("AOL"). Like Walt Disney before him, AOL CEO, Steve Case, seeks to
provide an environment in which the values that his organization favors may
flourish. AOL has a number of architectural features that are relevant to
the values that adhere in its space. Each AOL account is allowed five screen
names (p. 67). This architectural feature promotes anonymous, indeed pseudonymous,
online activity. As Lessig notes, anonymity may promote political expression
(pp. 70-73). AOL's chat rooms, however, have a maximum of twenty-four participants,
and, thus, it is not possible for members to address other members en masse.
n47 In addition, AOL has a general code of conduct that constrains certain
sorts of behavior such as obscenity (p. 67).
Consider next the online service Counsel Connect. The ninety or so discussion
groups that comprise Counsel Connect each have discussion leaders (p. 72).
By contrast to AOL, however, these discussion leaders do not have the ability
to cancel postings (p. 72). Thus, Counsel Connect is more respectful of free
speech than is AOL. Another distinct feature is that with Counsel Connect,
the lawyer participants must use their real names (p. 73). The use of real
names allows participants to develop reputations, such as the reputation
as a good lawyer to whom another lawyer would refer business (p. 73).
Finally, consider the values that may be instantiated in MUDs and MOOs.
n48 LamdaMOO is one of the older more well-established MOOs. It is a text-based
virtual reality that is linked to over 5000 members from around the world.
Upon entry into this virtual community, one is assigned a character, which
can then build a life in this community. Communities survive particular interactions,
which means that characters may develop reputations through their behavior
in various settings. Thus, in LamdaMOO, people are anonymous (like AOL) but
can develop reputations (like Counsel Connect). n49
[*1931] 7. Liberty is an Important Value that Ought to be Respected
Lessig clearly holds liberty to be an important value. He does not explicitly
provide normative arguments to this effect, however, but rather assumes it
to be so. This assumption is reasonable, given that Lessig is interested
in studying the effects of increased regulability on fundamental values.
Liberty certainly is a fundamental value, the word itself appearing in both
the Declaration of Independence and the Constitution. n50
Nor does Lessig provide explicit arguments to support the normative conclusion
that liberty ought to be respected and promoted. Given the moral logic of
value statements, however, Lessig is justified to make this assumption, for
to hold that a value is a value, is, other things equal, to hold that the
value ought to be respected and promoted. This is simply the meaning of holding
8. There is an Inverse Correlation Between Regulability and Liberty
A related theme of the book is that increased regulability will result in
decreased liberty. Again, Lessig does not say much that directly establishes
this point. Perhaps it simply seems obvious. Alternatively, he may see this
proposition naturally following his conception of regulation, which is defined
in terms of constraints (p. 217). More regulability means more constraints,
which, by definition, means less freedom, as freedom is reasonably construed
as an absence of constraints. n51
9. The Internet is Diminishing in its Capacity to Promote Liberty
The logic of Lessig's argument, then, is that because electronic commerce
is leading to increased regulability, and because increased regulability
causes a loss of liberty, electronic commerce is causing the Internet to
be a less free place. The argument has a proper form. Hence, unless one or
more of the premises is false, Lessig's important conclusion should be accepted.
Let us begin with Proposition five, which holds that code is the most important
regulator of cyberspace. This is an overstatement. [*1932] Lessig
does indeed provide strong reason to believe that code is an important regulator.
He does not, however, say anything that would establish that it is the most
important regulator. In particular, he does not pay attention to important
non-architectural means of regulating the Internet, such as Title II of the
DMCA and the other Internet-related statutes cited earlier, none of which
regulates computer code. n52 There is no reason to think that traditional
methods of regulation, such as those recently employed by the American and
Chinese governments, will not continue to be the predominant means to regulate
cyberspace. n53 Given this fact, one is not compelled to accept the
claim that code is the most important regulator of cyberspace without more
evidence or argument.
Proposition six holds that the Internet embodies values. This bare proposition
is uncontroversially true. Lessig's subtle discussion of the various contexts
in which this is true serves, however, to highlight the importance of this
claim, as the more we understand how the Internet embodies values, the better
position we will be in to shape these values.
Proposition seven holds that liberty is an important value that ought to
be respected and promoted. While some moral theories would reject this proposition,
most mainstream moral theories indeed recognize the central importance of
liberty, although there will of course be disagreement on whether liberty
is ultimately grounded consequentially or deontologically. n54 Thus,
Proposition seven is not in need of qualification.
Proposition eight holds that there is an inverse correlation between regulability
and liberty. This premise is subject to question. The purported inverse correlation
appears to be based on the claim that, as a general matter, the less e-commerce
entities know about you the better off you will be. n55 While this
claim is not analytically true, there are indeed many instances in which
the possession of your information by [*1933] others can hinder
individual freedom. The most significant single example along these lines
is Julie Cohen's argument that loss of the ability to read anonymously threatens
the core First Amendment value of freedom of conscience and thought.
The relationship between regulability and freedom is more complex, however,
than is suggested by Lessig's analysis. While it is true that you may be
harmed by others' possession of your data, you may also be helped. Code would
benefit from a discussion of the consumer surplus that may accrue to society
due to increased regulability. n57 For example, casual empirical observation
supports the claim that many people are happy to exchange personal information
for benefits received in return. Grocery stores that offer discounts for
card holders have no trouble establishing card-holder relationships despite
the fact that this bargain entails the release of purchasing data to the
store's sophisticated personal data-tracking system. Many people are quite
happy to take part in this particular market in personal data. n58
It would be odd to contend that people who choose to do so are somehow less
free than those who choose not to. n59 Freedom is measured in your
ability to choose, not in what you choose. n60
[*1934] It will not be an adequate response to note that people trade
away their data out of ignorance of the consequences. While this is no doubt
true, it is also true that people not ignorant in this manner trade away
their data. Lessig himself trades away data by his acceptance of cookies
in his web browser. n61 Plausibly, this shows that Lessig prefers the
benefits he gets from unrestricted Internet browsing over the consequent
loss to his informational privacy. While it is true that unless one provides
some data, many interesting websites will be inaccessible, this does not
make the provision of data in these circumstances any more coercive than
when one is required to pay admission to enter a movie or ball game.
Lessig implicitly acknowledges the potential compatibility between liberty
and regulability in the chapter entitled "Privacy" in Part Three of the book,
in which he applies the book's earlier arguments in particular contexts.
Lessig promotes a regime of user self-help to secure Internet privacy, such
as might be supplied by emerging technologies such as P3P (p. 160). P3P allows
client browsers to perform automatic electronic negotiations with websites,
based on the privacy preferences of the user and the privacy policies of
the website. Such transactions have the potential to make both users and
websites better off.
We see, then, that on one plausible conception of freedom, the contractarian
model, consumers may express their free desires when they bargain into a
situation of increased regulability. This means that there is not a simple
inverse correlation between regulability and freedom. Greater regulability
may just mean an architecture that allows for more and better opportunities
for bargaining. n63 Accordingly, the conclusion as stated in Proposition
nine, that cyberspace is diminishing liberty, is not established. Nor has
the negation of this proposition been established. More study into the relationship
between regulability and liberty is needed.
[*1935] 10. The Trend Toward Increased Regulability of the Internet
Ought to be Reversed
From the above set of propositions, Lessig draws a second normative conclusion.
By combining the premise that liberty is a fundamental value that ought to
be respected and promoted with the premise that cyberspace is becoming less
free due to increased regulability, Lessig concludes that the trend toward
increased regulability ought to be reversed (pp. 52, 56, 108). n64
By the lights of the discussion in the previous section, however, it should
be clear that this conclusion does not follow because the premise holding
that increased regulability leads to decreased freedom has not been established.
Thus, it may not be necessary to reverse the trend toward greater regulability
in order to promote liberty. What matters in terms of promoting liberty is
that future states of greater regulability are the result of autonomous processes.
11. Open Source Code is Less Regulable than Closed (Proprietary) Source Code
Lessig argues that open code will be less subject to regulability than proprietary
code. n65 Lessig claims that open source code will make top-down control
harder but will allow for bottom-up control (p. 20). Lessig's argument for
why open code will promote unregulability and hence liberty is simple but
powerful. The basic idea is that important proprietary code will likely be
owned by large commercial entities who will be easily susceptible to governmental
attempts to alter code to serve governmental interests. n66 Lessig
gives the example of [*1936] Communications Assistance for Law
Enforcement Act of 1994 ("CALEA"), n67 the statute that forced phone
companies to build in a backdoor to allow for governmental eavesdropping
of digital telephone lines (pp. 44-45, 106).
Contrast this example of easy regulation of proprietary code with the results
of a government attempt to create a backdoor into open code. Lessig provides
an example involving Netscape (p. 106). The French government tried to get
Netscape to alter its SSL encryption technology to allow a backdoor for police.
n68 Because Netscape had made its source code public, however, it was pointless
to comply with the French request (p. 106). Even if Netscape released a new
module that was compliant, because the source code was open, other vendors
would soon supply SSL without the modifications. In other words, because
SSL is open source code, it is not easily regulable. Accordingly, open code
deters censorship and, thus, promotes freedom. Lessig argues that this example
is indicative of a general fact that it is much more difficult to regulate
open code systems. Lessig's claim here is both significant and plausible.
Lessig sees open code as providing a "structural guarantee of constitutionalized
liberty" (p. 7). According to Lessig, open code "functions as a type of separation
of powers in the American Constitutional tradition" (p. 7). This remark may
at first appear puzzling because code does not perform a separation of powers
in the usual sense. The Constitution is conventionally understood to separate
power in two fundamental respects: first, between the states and the federal
government, and second, between the branches of the federal government.
n69 Apparently, what Lessig has in mind is a separation of powers between
governmental and nongovernmental social forces, in other words, between social
norms and law. We see in this discussion how Lessig's four-part account of
regulation and social order allows him to put meat on the bones of the concept
of a British-style constitution for America.
[*1937] 12. Open Source Code Ought to be Promoted
Because open source code is less regulable than closed source code, and
regulability is inversely related to liberty, and liberty is a value that
ought to be promoted, open source code ought to be promoted. n70 Subject
to the qualifications noted above about the relationship between regulability
and liberty, this is a valid argument, which moves from plausible premises
to a plausible conclusion. And given how high the stakes are - liberty in
cyberspace - an important conclusion as well.
13. We Need An Internet Constitution
Encouraging open source code is Lessig's main concrete prescription for
promoting a constitutional regime for cyberspace that respects and promotes
liberty. n71 Lessig provides little by way of detail as to how a world
of open source code might be accomplished. Unfortunately for his reader,
Lessig may be precluded from doing so by a conflict of interest, as such
a discussion could hardly fail to discuss Microsoft. Lessig does not discuss
the Microsoft antitrust litigation that was ongoing at the time of the publication
of the book. This is appropriate and expected given Lessig's ongoing role
in the litigation. n72
There is a deep tension between open code and Microsoft's business model,
which is that its software runs the world. n73 Microsoft's business
model is precisely that there not be open source code but rather that the
dominant (or exclusive) code of cyberspace be a proprietary Microsoft product.
While the remark is often made offhand, it does, indeed, appear that Bill
Gates wants to control cyberspace. n74 Lessig wants no one to control
[*1938] Lessig argues that, in order to systematically and democratically
deal with issues such as open code, we need to enter a period of constitution
building. Lessig recommends that we create a constitution for cyberspace.
n76 Recall that for Lessig, the word "constitution" is to be understood in
the British not the American sense. On this definition, constitution building
involves both governmental and nongovernmental action.
On the broad account of constitution, then, what Lessig is really saying
is that we would do well to implement a justified social order for cyberspace.
Considered first at this general level, this claim is highly plausible. Disagreement
will come when we begin to fill in the detail as to what in particular the
constitutional order should look like. Lessig does not provide details as
to what the overarching features of this order should be. Doing so would
be an entire project in itself. Lessig's main point is preliminary to this
discussion. Before creating a constitution, there must first be the recognition
that one is needed. This is the story of Code.
The reader comes away unclear on whether Lessig is proposing that the constitutional
order grow out of the U.S. legal system or instead that it evolve as a form
of world government. Throughout the book, Lessig uses the term "we." It appears
that "we" means we Americans. Does this mean that we Americans are supposed
to create the constitution for cyberspace? This is problematic, of course,
as the Internet is global.
Lessig suggests that we should choose to develop a new constitution for this
space, just as the American Founders and Russians had to (pp. 4-8). But there
are already constitutions in place that pertain to cyberspace. China just
announced rules to govern its space. n77 Despite the fact that the
official U.S. policy is to support industry self-regulation, the United States
has been regulating the Internet. n78 While early attempts at regulating
the Internet may have been unsuccessful, it is unlikely that this will cause
the United States to cease trying. This may make the issue of founding a
new social order to govern the Internet beside the point. In the United States
alone, regulation of the [*1939] Internet has arguably moved
past the point where a founding of a new order could be possible.
Nor does the broad need for a justified social order for cyberspace entail
the desirability of new constitutional amendments or federal statutes. Indeed,
a grundnorm of the Internet has been self-regulation. Lessig is right that
there is no necessary reason to prefer informal social solutions. Nevertheless,
informal solutions have a significant role to play. Lessig is also right
that anarcho-libertarianism is overly antigovernment in allowing no role
for the government. The government plays a fundamental role in providing
legal order. Legal order, however, is only one aspect of social order.
Lessig argues that the Constitution is thin; it does not adequately stretch
to cyberspace (p. 22). Two brief responses are worth mentioning, although
the topic merits greater discussion than can be pursued here. The first is
that the bare fact that there are examples in which the U.S. Constitution
does not apply snugly to Internet facts is certainly not enough reason to
conclude that the larger U.S. legal system is not capable of dealing with
the Internet. In particular, common law processes have done reasonably well
so far in adapting to the Internet. n79 So has the FTC, which has recently
fostered industry self-regulation regarding website provision of privacy
policies. n80 Second, there has not been enough time to tell. For example,
the FTC had made clear that it thinks more time is needed with respect to
self-regulatory efforts regarding informational privacy online. n81
Lessig does not point out that open source code is currently being driven
by commercial entities. n82 While open code may be highly desirable,
it may still be the case, then, that the best provider of it is the market
and norms, not the state.
Climbing the walls of your electronic cage is what you will be doing if
the near dystopia which Lessig foretells in his important new book,
[*1940] Code, comes to pass. In this world, the Internet is a central
and pervasive fact of people's lives, and it is no longer free. Liberty has
been coded away. Your options, choices, and movement are heavily restricted,
due to unprecedented levels of social regulation, made possible by the awesome
efficiencies of the Internet. In the encoded dystopia that Lessig predicts,
you will be wired, but feeling wired in.
The issues Lessig raises are new and important. Thus, while one can take
issue with Lessig on various points, the book on the whole nevertheless provides
a compelling account that is intentionally scary. Lessig seeks to scare because
he is on a mission to deflect the trajectory of modern techno-society from
its current path. His book demands attention and demands attention now because
the changes Lessig describes are radical and are happening in Internet time.
n1. Currently, Jack N. and Lillian R. Berkman Professor for Entrepreneurial
Legal Studies, Harvard Law School. In the fall of 2000, Professor Lessig
joins the faculty at Stanford Law School.
n2. P. 190; see also Dan Rosen, Surfing the Sento, 12 Berkeley Tech. L.J.
213, 216 (1997) (discussing the possible "founding fathers of a new cybernation");
Luke A. Walker, ICANN's Uniform Domain Name Dispute Resolution Policy, 15
Berkeley Tech. L.J. 289, 289 (2000) (calling cyberspace the latest new frontier).
Elsewhere Lessig has written:
While they are in that place, cyberspace, they are also here. They are at
a terminal screen, eating chips, ignoring the phone. They are downstairs
on the computer, late at night, while their husbands are asleep. They are
at work, or at cyber cafes, or in a computer lab. They live this life there,
while here. And then at some point in the day, they jack out, and are only
here. They step up from the machine, in a bit of a daze; they turn around.
They have returned.
Lawrence Lessig, The Zones of Cyberspace, 48 Stan. L. Rev. 1403, 1403 (1996)
n3. See p. 5 ("We build a world where freedom can flourish not by removing
from society any self-conscious control; we build a world where freedom can
flourish by setting it in a place where a particular kind of self-conscious
control survives. We build liberty, that is, as our founders did, by setting
society upon a certain constitution."). Lessig appears to use "we" to mean
both Americans and some undefined larger group: "We should understand that
we are part of a worldwide political battle; that we have views about what
rights should be guaranteed to all humans, regardless of their nationality;
and that we should be ready to press these views in this new political space
opened up by the Net." P. 205.
n4. I would argue instead that we have a constitution for cyberspace already.
It is the one written on parchment and displayed at the National Archives.
Things are, of course, complicated by the fact that the Internet is a global
phenomenon. But just because a phenomenon is global does not mean that the
Constitution cedes jurisdiction over those elements that have significant
impact within the United States. The Internet is revolutionary, but physical
borders still matter, and will continue to, for the foreseeable future. See
Jack Goldsmith, Regulation of the Internet: Three Persistent Fallacies, 73
Chi.-Kent L. Rev. 1119, 1124 (1998). In general, Lessig agrees that real
space jurisdiction matters for cyberspace regulation. P. 190.
n5. See p. 5 ("But by "constitution' I don't mean a legal text.... Rather,
as the British understand when they speak of their constitution, I mean an
architecture - not just a legal text but a way of life - that structures
and constrains social and legal power, to the end of protecting fundamental
values - principles and ideals that reach beyond the compromises of ordinary
n6. See p. 217 ("The Constitution was drawn at a time when basic architectures
were set. The framers found the laws of nature... they were not made by government
or man."); see also H.L.A. Hart, The Concept of Law 56-57, 88-90, 102-03
(2d ed. 1994); Thomas B. McAffee, Prolegomena to a Meaningful Debate of the
"Unwritten Constitution" Thesis, 61 U. Cin. L. Rev. 107, 166 n.192 (1992)
(citing J.W. Gough, Fundamental Law in English Constitutional History 174-91
n7. Lessig speaks of code and commerce as, "two forces of social order."
P. ix; see also, e.g., Russell Hardin, Trust and Trustworthiness (forthcoming
Jan. 2001); Jon Elster, The Cement of Society: A Study of Social Order (1989).
Lessig has been a contributor to the new social norms legal literature, which
places legal regulation within the broader context of overall social regulation.
See generally, e.g., Lawrence Lessig, The Law of the Horse: What Cyberlaw
Might Teach, 113 Harv. L. Rev. 501 (1999); Lawrence Lessig, The Limits in
Open Code: Regulatory Standards and the Future of the Net, 14 Berkeley Tech.
L.J. 759 (1999); Lawrence Lessig, Commons And Code, 9 Fordham Intell. Prop.,
Media & Ent. L.J. 405 (1999); Lawrence Lessig, The Zones of Cyberspace,
48 Stan. L. Rev. 1403 (1996); Lawrence Lessig, Intellectual Property And
Code, 11 St. John's J. Legal Comment. 635 (1996); Lawrence Lessig, Reading
the Constitution in Cyberspace, 45 Emory L.J. 869 (1996); Lawrence Lessig,
Post Constitutionalism, 94 Mich. L. Rev. 1422 (1996) (book review).
n8. See p. 6 ("We see that much of the "liberty' present at cyberspace's
founding will vanish in its future.").
n9. See p. x ("[A] future of control in large part exercised by technologies
of commerce, backed by the rule of law.").
n10. Lay-libertarian sentiment is aptly characterized in the following well-known
phrase: "We believe in: rough consensus and running code." P. 4. The phrase
was coined by one of the founders of the Internet, David Clark of MIT, in
describing the philosophy of setting Internet standards. See Gary C. Kessler,
IETF - History, Background, and Role in Today's Internet (visited June 22,
2000) <http://www.vtac.com/Tutorials/ietf<uscore>hx.html>. This
strong libertarian stance has the implication that a position such as Richard
Epstein's would make him a "Red." See Lawrence Lessig, Real-World Libertarians
and the Net (visited June 22, 2000) <http://slate.msn.com/code/BookClub/BookClub.asp?Show=1/17/00&idMessa
ge=4391 &idBio=139> [hereinafter Lessig, Real-World Libertarians and
the Net]; Richard Epstein, Libertarianism is not Anarchism (visited June
22, 2000) <http://slate.msn.com/code/ BookClub/BookClub.asp?Show=1/17/00&idMessage=4391&idBio=139>
n11. See p. 6 ("The argument of this book is that the invisible hand of cyberspace
is building an architecture that is quite the opposite of what it was at
cyberspace's birth.... This book is about that change, and about how we might
prevent it."); see also Russell Hardin, Collective Action 6-15 (1982) (discussing
the "back of the invisible hand").
n12. See p. 6 ("We can build, or architect, or code cyberspace to protect
values that we believe are fundamental, or we can build, or architect, or
code cyberspace to allow those values to disappear.").
n13. See generally Jules L. Coleman, Negative and Positive Positivism, 11
J. Legal Stud. 139 (1982); Anthony J. Sebok, Misunderstanding Positivism,
93 Mich. L. Rev. 2054 (1995). What Lessig appears to mean is that code is
law in the sense of being a law of a nature for cyberspace. P. 70.
n14. See Steven Johnson, Interface Culture: How New Technology Transforms
the Way We Create and Communicate (1997) ("All works of architecture imply
a worldview, which means that all architecture is in some deeper sense political.");
James Boyle, Foucault in Cyberspace: Surveillance, Sovereignty, and Hardwired
Censors, 66 U. Cin. L. Rev. 177 (1997); Mitchell Kapor, The Software Design
Manifesto (visited June 23, 2000) <http://www.kei.com/homepages/mkapor/Software<uscore>Design<uscore>Manife
sto.html>; Pamela Samuelson et al., A Manifesto Concerning the Legal Protection
of Computer Programs, 94 Colum. L. Rev. 2308 (1994).
n15. P. 19. For a similar but non-normative definition, see p. 14 where Lessig
states, "by "regulable' I mean simply that a certain behavior is capable
n16. See infra text accompanying notes 65-74. Though Lessig is understandably
coy, the implication for the Microsoft antitrust litigation is apparent.
See United States v. Microsoft Corp., 87 F. Supp. 2d 30 (D.D.C. 2000).
n17. See p. 7 ("One part of this question of ownership is at the core of
the current debate between open and closed source software. In a way that
the American founders would have instinctively understood, "free software'
or "open source software' - or "open code'... is itself a check on arbitrary
n18. Pp. 19-23. Note that the first two themes, "Regulability," and "Regulation
by code," correspond to the first two parts of the book, "Regulability,"
and "Code and Other Regulators."
n19. Themes "describe." P. 8. Lessig's themes are nevertheless adaptive,
as they make the book more accessible to a wider audience, which is an important
goal of the book. Lessig says that his audience is second-generation netizens.
n20. For a definition and discussion of "Net95," see infra text accompanying
n21. David Hume, A Treatise of Human Nature, 1739-40 (L.A. Selby-Bigge &
P.H. Nidditch eds., 2d ed., Oxford Univ. Press 1978); see also R.M. Hare,
Moral Thinking: Its Levels, Method and Point 16 (1981) (discussing "Hume's
n22. See pp. 25-27. Lessig states that the computer network that previously
existed at the University of Chicago exemplifies Net95. P. 27. He had direct
acquaintance with this system, as he was teaching at the University of Chicago
when this system was in place.
n23. See p. 31.
"Authentication" is the process by which aspects of your identity become
known.... If I walk into a bank, the teller will know a lot about me even
if I don't say a thing: he will know I'm a puffy, middle-aged white guy with
glasses and blondish hair; he will know I'm not big and not strong.... He
will know all this whether I want to tell him or not.... Hiding usually does
not hide itself very well; usually we reveal that we are hiding.
P. 31. Lessig gives the example of the University of Chicago system. The
architecture required no credentials; thus, it was "both easy to hide that
you [were] a dog and hard to prove that you were not." P. 33.
n24. Pp. 34-35. Each of these three credentials provides a means to identify
someone on the Internet. Typically, a "password" is a word or number that
is kept secret. The user enters the password along with her account name
in order to verify that the user is authorized to use the system. Manufacturers
are experimenting with biometric devices such as thumbprint readers and retina
scanners to link particular individuals to particular machines. P. 57. A
"cookie" is a small bit of data entered by your browser to a "cookie file"
on your hard drive. Websites that you visit initiate these cookies so that
when you return to the site, they can recognize you because the cookie is
sent by your browser to the website along with the request for the site.
Like the above two technologies, "digital certificates" serve to identify
you on the Internet. Digital certificates provide and certify more detailed
information, however, such as citizenship, age, occupation, gender, etc.
n25. IP addresses are the bits of data that allow packets of information
to move from one place to another on the Net. They look like this: 3126.96.36.1996.
Basically, IP addresses function like postal addresses, saying from where
the packet is coming and to where the packet is going.
n26. See p. 28 ("Pictures of flesh come across a screen, but the system cannot
tell whether the pictures are medical photos or pornography. Data about bodily
functions come across the wire, but the system cannot tell whether the data
are from medical records or a novel.... Net95 had no requirement that data
n27. P. 28 (emphasis omitted). Lessig argues that zoning techniques will
lead to previously unimagined abilities to regulate the Web. He writes:
The effect, in short, would be to zone cyberspace based on the qualifications
carried by individual users. It would enable a degree of control of cyberspace
that few have ever imagined. Cyberspace would go from being an unregulable
space to, depending on the depth of the certificates in the space, the most
regulable space imaginable.
P. 57 (emphasis omitted).
n28. See p. 33 ("This minimalism in design is intentional. It reflects both
a political decision about disabling control and a technological decision
about the optimal network design. The designers were not interested in advancing
social control; they were concerned with network efficiency.").
n29. See p. 4 ("As in post-Communist Europe, first thoughts about cyberspace
tied freedom to the disappearance of the state. But here the bond was even
stronger than in post-Communist Europe. The claim now was that government
could not regulate cyberspace, that cyberspace was essentially, and unavoidably,
free.") (emphasis omitted).
n30. See Deborah M. Thaw, The Net Makes Notaries More Necessary, Not Less,
Bus. Wk., Dec. 13, 1999, at 18 (arguing that current business dealing demands
n31. See p. 30 ("As the Net is being remade to fit the demands of commerce,
architectures are being added to make it serve commerce more efficiently.
Regulability will be a by-product of these changes.").
n32. Pub. L. No. 105-304, 112 Stat. 2860 (1998). Title II of the Act is known
as the Online Copyright Infringement Liability Limitation Act.
n33. In fact, each of the statutes produced by the U.S. Congress thus far
to regulate the Internet are best viewed as low-tech, non-code-based forms
of regulation. See Anticybersquatting Consumer Protection Act, Pub. L. No.
106-113, 113 Stat. 1536, 1501A-545 (1999) (codified as scattered sections
of 15 U.S.C.A. & 16 U.S.C.A. (West Supp. 2000)) [hereinafter Anticybersquatting
Act]; No Electronic Theft ("NET") Act, Pub. L. No. 105-147, 111 Stat. 2678
(1997) (codified as scattered sections of 17 U.S.C., 18 U.S.C., 28 U.S.C.
(1994 & Supp. III 1997)) [hereinafter NET Act]; Communications Decency
Act of 1996, Pub. L. No. 104-104, 509, 110 Stat. 133, 137-139 (1996) (codified
as 47 U.S.C. 230 (1994 & Supp. II 1996)) [hereinafter 230 of the CDA].
This is true as well with regard to recent efforts by other countries, such
as China, to regulate Internet use. The Chinese government recently stepped
up efforts to censor the content that appears on the screens of computers
in China. The government utilizes a variety of means to accomplish this result,
many of which are distinctly low-tech, such as threatening to imprison those
who post proscribed content:
They routinely block Web sites for some international media outlets and are
training special police units to monitor Internet activity. Internet-cafe
operators are charged with making sure users don't "endanger national security"
while online. A final line of defense is the low-tech, tried and true method
of scaring users into policing themselves by making examples of people who
venture into banned territory. In December, a Shanghai court handed down
a two-year jail term to a software vendor, Lin Hai, who sold 30,000 Chinese
e-mail addresses to a U.S.-based online dissident journal.
Susan Lawrence, Beijing Mounts Great Leap Online, Nat'l Post, Mar. 3, 1999,
available in WESTLAW, 1999 WL 13666980.
n34. See p. 85 & n.1 (""Libertarian,' however, has a specific meaning
for us. It associates with arguments against government." "Or more precisely,
against a certain form of government regulation ( in cyberspace).").
n35. Lessig begins Chapter Three with the following epigram, which he says
exemplifies the sort of libertarian view he has in mind: "The rise of an
electronic medium that disregards geographical boundaries throws the law
into disarray by creating entirely new phenomena that need to become the
subject of clear legal rules but that cannot be governed, satisfactorily,
by any current territorially based sovereign." P. 24 (quoting David R. Johnson
& David Post, Law and Borders - The Rise of Law in Cyberspace, 48 Stan.
L. Rev. 1367, 1375 (1996)).
n36. See Epstein, supra note 10, P 5 ("Rather, the libertarian is someone
whose objective function starts with the goal of minimizing the use of force
and fraud in human interactions.... The libertarian is not an anarchist.").
n37. In his subsequent exchange with Richard Epstein in Slate magazine, Lessig
provides the following helpful summary of competing libertarian conceptions.
But there is "libertarianism" in the ivory tower and there is "libertarianism"
on the ground. I recognize the species that Richard describes; I am a permanent
resident of the ivory tower. But my book describes a present political attitude,
not the ivory tower. It is about a present political reality, and a present
rhetorical push. I am describing it because I have been watching it for the
past six years. In that world, if someone argued (as Richard does above)
that a "law of privacy" was needed, as well as law protecting trade secrets;
that laws regulating libel and slander were necessary, as well as a law regulating
blackmail; if one even raised the issue of taxation, or suggested that the
government was needed to "secure the infrastructure," then one would not
be a "libertarian." One would be a Red. The libertarian of the Net has a
simple message, quite different from ivory-tower libertarianism. It is: Keep
the government out.
See Lessig, Real-World Libertarians and the Net, supra note 10, P 7.
n38. Lessig, Real-World Libertarians and the Net, supra note 10, PP 8-9.
n39. Lessig develops this account more fully in the appendix to the book.
See pp. 235-39.
n40. In Chapter 7, in which Lessig lays out his account of social regulation,
he provides a number of diagrams that are meant to capture the individual
persons buffeted about by the four forces impinging from all four directions.
In the center of these forces, the individual is represented by a "dot."
Curiously, Lessig refers to the dots (us) as "pathetic." P. 86.
n41. David Hume, An Abstract of A Treatise of Human Nature 32 (1740) (reprinted
with an introduction by J.M. Keynes and P. Staffa, Cambridge University Press
n42. See generally Elster, supra note 7.
n43. P. 82 ("Cyberspace is not a place; it is many places. Its places don't
have one nature; the places of cyberspaces have many different "natures.'
These natures are not given, they are made.... These architectures are themselves
not given; these architectures of code are set by the architects of cyberspace
- code writers.").
n44. See p. 92 & n.9 (citing Robert A. Caro, The Power Broker: Robert
Moses and the Fall of New York 318 (1974)).
n45. See p. 83 ("At the University of Chicago, if you wanted access to the
Internet, you simply connected your machine to jacks located throughout the
university. Any machine with an Ethernet connection could be plugged into
these jacks. Once connected, your machine had full access to the Internet
- access, that is, that was complete, anonymous, and free." (footnote omitted)).
n46. See p. 26 ("You cannot connect your machine to the net at Harvard unless
the machine is registered - licensed, approved, verified.... Once registered,
all interactions with the network are monitored and identified to a particular
n47. See p. 68 ("There is no town hall or town meeting where people can complain
in public and have their complaints heard by others. There is no space large
enough for citizens to create a riot. The owners of AOL, however, can speak
to all. Steve Case, the "town mayor," writes "chatty" letters to the members.").
n48. A MUD is typically defined as a multi-user dungeon or multi-user domain.
A MOO is an object-oriented MUD. See generally Sherry Turkle, Life on the
Screen: Identity in the Age of the Internet (1995).
n49. See Jennifer L. Mnookin, Virtual(ly) Law: The Emergence of Law in LambdaMOO,
2 J. Computer-Mediated Comm. 1 (1996) <http://www.ascusc.org/jcmc/vol2/issue1/
n50. U.S. Const. preamble & amend. V; The Declaration of Independence
para. 2 (U.S. 1776).
n51. While his focus is on liberty, Lessig contends that increased regulability
will have an adverse impact on other values as well. He writes, "Perfect
authentication would mean that others know for certain all the facts about
you; happiness comes from others knowing a good deal less." P. 31.
n52. See supra text accompanying notes 32-33.
n53. See supra note 33.
n54. See generally John Stuart Mill, On Liberty (Currin V. Shields ed., Bobbs-Merrill
Co. 1956) (1859) (classic consequentialist defense of liberty); Immanuel
Kant, The Metaphysics of Morals (Mary Gregor trans. & ed., Cambridge
Univ. Press 1996) (1797) (classic deontological defense of liberty). Nothing
Lessig says in the book indicates conclusively whether he is a consequentialist
or deontologist. Some of his remarks indicate sensitivity to welfarist concerns.
See pp. 146-47. Generally speaking, however, the preoccupation with liberty
and freedom sounds in deontology, as these are typically non-instrumentalist
values by the lights of mainstream moral and political theory. It is reasonable
to assume, then, that Lessig is some sort of deontologist who places great
importance on the value of liberty. What is not clear is whether he is the
sort of deontologist who thinks duties and rights can be traded off against
consequentialist considerations, even if the former are not reducible to
n55. See supra note 51.
n56. See Julie E. Cohen, A Right to Read Anonymously: A Closer Look at "Copyright
Management" in Cyberspace, 28 Conn. L. Rev. 981, 1012 (1996) ("The freedom
to read anonymously is just as much a part of our tradition, and the choice
of reading materials just as expressive of identity, as the decision to use
or withhold one's name."). In discussing the "Cohen Theorem," Lessig offers
an example from when he was a student at an English university. While there,
he made several purchases of Scotch as gifts. When he returned to school,
a tutor asked him about his excessive purchasing of alcohol. Pp. 138-39 ("Now
that monitoring can occur, we must ask whether the latent right to read anonymously,
given to us before by imperfections in technologies, should be a legally
n57. Note that the benefits need not be consequentialist goods whose benefit
comes at the expense of liberty. One's access to online media may make one
more enlightened, more autonomous, and, hence, freer.
n58. See Computer-Dating the Customer, The Economist, Mar. 4, 1995 at R7.
The bare fact that this practice exists does not show that it is efficient,
however, as there may be market failure. While there may be market failure,
there also may not be market failure. Lessig does not discuss which of these
possibilities is more likely to be true.
n59. See Susan Rose-Ackerman, Inalienability and the Theory of Property Rights,
85 Colum. L. Rev. 931, 932 (1985) (noting that "unencumbered market trades
are desirable unless we can locate a valid reason for their restriction");
see also Michael A. Heller, The Boundaries of Private Property, 108 Yale
L.J. 1163, 1199-1201 (1999) (discussing the role of restrictions on restraint
Are Americans, in general, freer now than they were a hundred years ago?
Even if this question is meaningful, the answer is certainly unclear. The
answer would not be unclear, however, if there was a direct correlation between
loss of informational privacy and loss of freedom, because we have dramatically
less informational privacy than we had a century ago.
n60. See generally Gerald Dworkin, The Theory and Practice of Autonomy (1988).
n61. See supra note 24; pp. 41-42 ("With one click, you can disable the deposit
of cookies.... but this privacy comes at a cost. Users who choose this option
are either unable to use areas of the Net where cookies are required or forced
constantly to choose whether a cookie will be deposited. Most find the hassle
too great and simply accept cookies on their machine.").
n62. Currently, much personal data gathering by websites is not conducted
pursuant to consensual bargains. Rather, many websites either collect data
without notice or use the data in a manner that is beyond the scope of the
agreement. See Steven Hetcher, The FTC as Privacy Norm Entrepreneur, 53 Vand.
L. Rev. (forthcoming 2000).
n63. In a free and fair market, for example, people may choose to carry digital
credentials in order to conduct the sort of electronic commerce functions
that are facilitated by these credentials.
n64. Depending on the underlying value theory one assumes, the fact that
liberty is an important value will not be enough to draw any necessary practical
inferences solely based on the existence of a serious threat to liberty.
The reason is that by the lights of prominent and respected theories, a diminution
in liberty may be morally permissible if it is the result of some moral or
policy choice that pursues a distinct sort of benefit or sorts of benefits
that, all things considered, balances out the liberty concern. If the proper
normative account requires trade-offs, then Lessig must countenance the possibility
that a diminution in liberty is justified by the fact that the proposed state
of affairs promotes other values. This is more than a theoretical possibility,
if we are to judge based on the current growth in the market in personal
data, a significant and growing part of which is occurring pursuant to consensual
practices. See Hetcher, supra note 62. Lessig has to respect the fact that
mainstream Internet users today may not pine after Net95 the way that first-generation
users did. Freedom may simply mean the autonomy to indulge even more conspicuously
in the fatuous materialism that constitutes much of contemporary American
n65. P. 107 ("To the extent that code is open code, the power of government
is constrained.... When the target of its regulation is plastic, it cannot
rely on its target remaining as it wants.").
n66. P. 52 ("But as code writing becomes commercial - as it becomes the product
of a smaller number of large companies - the government's ability to regulate
it increases."). Lessig also writes "the code is regulable only because the
code writers can be controlled.... An unmovable, and unmoving, target of
regulation, then, is a good start to regulability. And this statement has
an interesting corollary: regulable code is closed code." P. 106.
n67. Communications Assistance for Law Enforcement Act (CALEA), Pub. L. No.
103-414, 108 Stat. 4279 (1994) (codified as amended at 18 U.S.C. 2522 (1994)
and 47 U.S.C. 229, 1001-1010 (1994 & Supp. II 1996)).
n68. SSL is Netscape's protocol for exchanging encrypted data between the
client's browser and the host's server.
n69. See Hayburn's Case, 2 U.S. (2 Dall.) 409 (1792).
n70. See p. 8 ("Guarantee the structural (a space in cyberspace for open
code), and (much of) the substance will take care of itself."). Lessig argues
that we should strive for a "commons" in the core code of the Internet. P.
n71. See p. 7 ("If the code of cyberspace is owned (in a sense that I describe
in the book), it can be controlled; if it is not owned, control is much more
difficult. The lack of ownership, the absence of property, the inability
to direct how ideas will be used - in a word, the presence of a commons -
is key to limiting, or checking, certain forms of government control.").
n72. P. 8. See, e.g., Lawrence Lessig, Amicus Brief on Technological Tying
n73. See p. 105 ("Microsoft may have imagined in 1995 that by 2000 there
would be no other server operating system available except Windows NT, but
when 2000 came around, there was GNU/Linux.").
n74. See Laura Evenson, Gates to the Future: Microsoft Founder Outlines the
World and How He'll Run It, S.F. Chron., Nov. 23, 1995, at C1.
n75. Richard Epstein has argued that open and proprietary codes can easily
coexist. Epstein draws a parallel with the manner in which public highways
easily coexist with gated communities. This is a very important issue. If
a multicode world is the future, then Microsoft presents less of a threat
than if a monocode world is likely instead. Epstein, supra note 10, Jan.
19 posting, PP 2-4.
n76. Lessig thinks it is natural and proper that the state should play an
integral role in the process of constitution building in cyberspace. This
will probably strike the typical reader of this Review as uncontroversial,
but the crowd that Lessig most wants to convince on this particular point
is not constituted of the readers of this Review but rather the influential
libertarian contingent of the Internet policy community. These are libertarians
staunchly opposed to all state involvement. See supra note 34.
n77. See supra note 33.
n78. See, e.g., Anticybersquatting Act, supra note 33; NET Act, supra note
33; 230 of the CDA, supra note 33; DMCA, supra note 32. These statutes are
operative only because they pass muster under the U.S. Constitution.
n79. See, e.g., Compuserve v. Cyberpromotions, 962 F. Supp. 1015 (1997) (applying
trespass theory in context of unsolicited electronic mail).
n80. In its 1999 Report to Congress, the FTC notes that there has been significant
improvement in website provision of privacy policies. Martha Landesberg &
Laura Mazzarella, Federal Trade Comm'n, Self-Regulation and Privacy Online:
A Report to Congress (1999) (visited July 5, 2000) <http://www.ftc.gov/os/1999/9907/
privacy99.pdf>. In its most recent Report to Congress, the FTC recommends
fairly minimal regulations, along with its continued support of industry
n81. See id.
n82. See, e.g., "Red Hat Reinforces Commitment to Open Source, Releases Source
Code for Popular Source-Navigator 10E" (press release) available at <http://www.redhat.com/about/2000/press<uscore>sourcemav.html>.