Internet exceptionalism: an overview from general constitutional law.

• Author: Tushnet, Mark V.

TABLE OF CONTENTS INTRODUCTION I. WHY--OR WHY NOT--INTERNET EXCEPTIONALISM: SOME PRELIMINARY OBSERVATIONS II. STRATEGIES FOR DEVELOPING A "LAW OF THE INTERNET UNTO ITSELF" A. Allowing a Period of Policy Experimentation B. Drawing Inferences from the History of First Amendment Treatment of Technological Innovations III. The Internet's Distinctive "Natures, Values, [and] Abuses" A. Internet Norms Are Fluid or Nonexistent B. The Internet Is a Bigger and Better System for Amplifying "Sound" C. Disseminating Information over the Internet Is Dramatically Less Costly than Other Modes of Dissemination D. The Putative Anonymity of the Internet IV. SOME ADDITIONAL QUALIFICATIONS A. Doctrinal Structure B. Business Models and the First Amendment C. The Lurking Problem of State Action CONCLUSION INTRODUCTION

This Article considers First Amendment Internet exceptionalism. I use that term in what I think is a reasonably standard way to refer to the question of whether the technological characteristics of the Internet (and, more generally, twenty-first-century information technologies) justify treating regulation of information dissemination through the Internet differently from regulation of such dissemination through nineteenth- and twentieth-century media, such as print, radio, and television. My aim here is not to provide an answer to that question, but to identify several subquestions whose answers must be part of the larger answer.

I begin with a disclaimer. After thinking and writing about general constitutional law and theory for many years, I began to think about the First Amendment relatively recently, and about the implications of that Amendment for the Internet even more recently. With so much specialized writing about the Internet and the First Amendment already produced, I should note that my reflections on the possibility of Internet exceptionalism might simply be reinventing the wheel--that is, discussing in quite summary form matters that have been discussed in more detail elsewhere. (1) The fact that the term "Internet exceptionalism" is well known in the field indicates that much has indeed been said about the questions I discuss here. (2) Still, I have not found a crisp statement presenting many of the matters I find of interest in a single place, so the Article may have some value as such a compilation even if it is not all that original.

Coming to the topic from general constitutional theory and law, I believe that I am somewhat more sensitive than most of those who write on the subject to the question that pervades the entire field. That is the question of the appropriate degree or form of judicial deference to legislative regulatory interventions, whether those interventions occur in the material economy or in the information economy. Not surprisingly, scholars who focus almost exclusively on the Internet and the First Amendment, to the exclusion of general constitutional theory and law, simply assume that relatively intrusive judicial supervision of regulatory decisions dealing with the information economy is appropriate. (3) To the extent that the scholarship adverts to the question of judicial deference, I believe that it tends to assume that the question is adequately answered by referring to Footnote Four of United States v. Carotene Products (4) and the scholarship of John Hart Ely. (5) My view is that such an assumption is not warranted. I defend this view only indirectly by attempting to identify why and how the question of judicial deference is a complex one. (6)

After the description in Part I of some general questions about the structure of constitutional doctrine, Part II examines two strategies that courts have used to deal with technological innovations--one allowing legislative experimentation until experience accumulates, the other imposing existing (or what I call "standard") doctrine from the outset. Part III looks at some attributes that are said to distinguish the Internet from traditional media--norms, scope, cost, and anonymity--with the aim of mapping out how or when Internet exceptionalism might be justified. Part IV discusses several general qualifications to the preceding analysis, involving doctrinal structure yet again, the First Amendment's bearing on regulation of business models, and the state action doctrine. Finally in a brief conclusion, I suggest that framing the discussion as one about Internet exceptionalism in a broad sense is misleading.

1. WHY--OR WHY NOT--INTERNET EXCEPTIONALISM: SOME PRELIMINARY OBSERVATIONS

   Considering the use of sound amplifying equipment by trucks cruising city streets to disseminate a political message, Justice Robert Jackson wrote, "The moving picture screen, the radio, the newspaper, the handbill, the sound truck and the street corner orator have differing natures, values, abuses and dangers. Each, in my view, is a law unto itself." (7) To that, we can now add the Internet. So, for example, the cost of distributing information, whatever its nature, over the Internet is much lower than the cost of doing so in other media, particularly when the distributor uses one of the various social networks as an intermediary. It is somewhat easier to distribute information "anonymously" over the Internet in the sense that the steps one must take to identify the speaker may be more complicated or "techy" than the steps one must take to identify the person responsible for a leaflet or television advertisement.

   Justice Jackson's observation rests on a proposition about the form that First Amendment doctrine takes. (8) For him, the constitutionality of specific regulations depends upon an assessment of "values, abuses and dangers"--that is, on what his generation would have called a balancing of interests and what today might be called a determination of the regulation's proportionality. (9) Whether that form was the correct one was contested at the time, with Justice Hugo Black notably asserting that First Amendment doctrine should be more categorical, (10) and is even more contested today. (11) My argument in this Part is that the alternative forms of regulation--categorical rules or balancing tests--can be indistinguishable in practice, at least when each is done carefully. Essentially, a well-designed set of categorical rules will identify a large range of characteristics whose presence in varying degrees triggers the application of discrete rules within the set, and well-performed balancing will take precisely those same characteristics into account and give them appropriate weights in generating outcomes.

   For analytical clarity, we should pry apart the two elements Justice Jackson combined. We might want to develop separate rules for each medium of information dissemination, or we might apply a general balancing or proportionality test to every medium. We might call the rule-based approach one of Internet exceptionalism (12) and the balancing approach a unitary account of the First Amendment.

   Balancing can produce the following outcome: a regulation that would be constitutionally impermissible if invoked against print media would be constitutionally permissible when invoked against Internet dissemination. (13) That might look like Internet exceptionalism on the level of outcomes, but it would result from a unified approach to First Amendment problems.

   One might think that a rule-based analytic approach could not have similar distinct results. Professor Jim Chen provides a useful formulation, in his discussion of regulation of Internet intermediaries, which he calls conduits: "Conduit-based regulation of speech is a constitutional mirage.... Conduit-based regulation raises precisely the same issues as all other decisions reviewable under the First Amendment." (14) Reno v. ACL Uexemplified this approach by applying existing rule-based doctrines in a challenge to the constitutionality of the Communications Decency Act (CDA), which restricted the distribution of some sexually explicit but nonobscene materials via the Internet. (15) According to the Court, "the CDA is a content-based blanket restriction on speech, and, as such, cannot be 'properly analyzed as a form of time, place, and manner regulation."' (16)

   Yet, as Chen observes immediately after stating the general point, "distinct conduits raise distinct regulatory concerns, ranging from strictly physical characteristics to economic predictions regarding markets that exploit that conduit. Persuasive free speech jurisprudence considers differences of this sort." (17) So, we would apply existing, pre-Internet doctrine (no Internet exceptionalism), but with some adjustments, or "tweaks," to take account of the Internet's distinct characteristics (Internet exceptionalism). (18) And, as I suggested at the outset, the choice of doctrinal structure need have no impact on the outcomes generated. Internet exceptionalism and standard doctrine with tweaks could produce the same "rules." (19)

2. STRATEGIES FOR DEVELOPING A "LAW OF THE INTERNET UNTO ITSELF"

   Here I identify two general strategies (20) for following Justice Jackson's advice to think about a law of the Internet unto itself.

   1. Allowing a Period of Policy Experimentation

      In Denver Area Educational Telecommunications Consortium, Inc. v. FCC, Justice John Paul Stevens wrote, "At this early stage in the regulation of this developing industry, Congress should not be put to an all-or-nothing choice." (21) The thought here is that individual technological innovations implicate an array of dangers and constitutional values, and that neither legislatures nor courts have any special insights, relative to the other, about the constitutionally appropriate response when the innovations have just been introduced. This counsels in favor of deference to democratically responsible decision making or, as Justice Stevens put it, deference to congressional choices. But, as experience with the innovation and with policy experiments accumulates...