Student Blogger - A Second Look at the Third Party Doctrine

Professor Richard Epstein on the 4th Amendment Third Party Doctrine

Criminals who can't keep secrets have short careers. A big reason for this is the “third-party doctrine” which, as most readers of this blog are probably aware, holds that information shared with others is not generally protected by 4^th Amendment restrictions on search and seizure. Put differently, the law does not recognize any reasonable expectation of privacy in information shared with others. This doctrine has been widely criticized and, most recently and notably in an article by 4^th Amendment scholar Orin Kerr, defended (The Case for the Third-Party Doctrine, 109 Mich. L. Rev. 561, 563 (2009)). Chicago's own Richard Epstein has recently taken interest in this debate, and presented his preliminary thoughts at this week's Works in Progress (WiP) talk. Essentially, Prof. Epstein thinks current constitutional doctrine is generally correct on this point. He agrees with Prof. Kerr and other defenders of the third-party doctrine that it is worth keeping, though his defense is on somewhat different grounds than those of Prof. Kerr. Since Prof. Epstein comes to this issue as an outsider, he also returns to first principles in his work and asks whether we are analyzing these cases consistently and coherently. His conclusion is to suggest a general method for considering 4^th Amendment issues that optimizes the public benefits and private costs of permitting government action in various classes of cases.

The traditional justification for the third-party doctrine is assumption of risk – revelation of information to someone else implies an awareness of and responsibility for any consequences that result, including that the information might be obtained by the government. Prof. Kerr modifies this defense somewhat, claiming that it is consent, rather than assumption of risk, that justifies the doctrine.

Prof. Epstein argues that neither of these justifications is sufficient. If they were, he argues, the government could simply announce in advance any rule governing its ability to get information, and anyone crossing the line set by that rule could then be said to have assumed the risk or consented to disclosure. This can't be right – if it were, the 4^th Amendment would lose its meaning. As Prof. Epstein argues, “monopoly power cannot be used to extract rights from all citizens.”

Prof. Epstein then searches for alternative justifications for the doctrine and settles on one of reasonable expectations, a principle widely used in the privacy context already. He acknowledges that such a justification is somewhat circular, in that expectations are set by the rules or standards themselves. Nevertheless, he appeals to social norms and lessons from other areas of law such as products liability and takings doctrine to illustrate that such an approach is workable. Furthermore, he points out, a reasonableness element of 4^th Amendment inquiries is strongly implied by the language of the Constitution itself, which after all bars only “unreasonable searches and seizures”. In practice, the reasonable expectations principle explains why police surveillance of public spaces (streets, parks, etc.) is permissible while similar surveillance (without a warrant) of private spaces is not. The same logic, Prof. Epstein argues, can explain the lines that are (or at least should be) drawn when information is shared with third parties. If the person sharing the information has a reasonable expectation that the information will remain secret, then the government will not be able to obtain it without a warrant. If there is no such expectation, or it is unreasonable, then police can get and use the information the same way they could if the subject announced it publicly.

As an illustration, Prof. Epstein looks at “pen register” cases, in which government agents track phone calls being made by someone under investigation. The government in these cases is not listening in on the content of the calls, but merely tracking the connections being made. Both the connections information and the content of the calls are revealed to third parties – the recipient of the call and the telephone company. Restriction of the government’s ability to obtain information without a warrant to connection, rather than content, information seems to match up with the subject's reasonable expectations, while an assumption of risk or consent justification for the third-party doctrine requires at least some contortions to do the same. Indeed, in Smith v. Maryland, the Supreme Court deemed pen registers constitutional (the pen register was ruled not to be a search), though Congress has since restricted their use to some degree by statute.

More broadly, Prof. Epstein argues that in all such cases where government access to private information would have some public benefit (reduction in crime) but private cost (chilling effects on communication, violation of privacy, etc.), an “optimization” approach is needed, in which the rule is set at the point at which public benefit most greatly exceeds private cost. This is not easy, of course – as Prof. Epstein identifies, these problems are as common in private law as in public law, and there are no quick solutions in either domain. As he puts it, “we can be quite precise in identifying the relative conflicts, but very cautious in asserting dogmatic solutions.” Instead of offering such solutions, Prof. Epstein instead advocates a category-based approach. Put simply, similar cases should be treated similarly. This allows for some optimization, without getting bogged-down in a case-by-case approach that would result in analysis paralysis of the courts. Putting it more philosophically, Epstein suggested that rule, not act utilitarianism is the proper approach.

Commenters at the WiP questioned Prof. Epstein on some of the assumptions and conclusions made in the paper. Some argued that the social norms on which a reasonable-expectations justification for the third-party doctrine must be based are problematic. They may not be as stable as Epstein seems to suggest or his analysis requires, or they may do nothing to solve the circularity problem. If social norms can change, then legal rules or standards may at least in some part drive them. The result is either unprincipled swings in application of the principle, or ossification at an arbitrary, unprincipled point.

Others questioned Prof. Epstein's broader conclusions about the proper analytical framework for 4^th Amendment questions. While agreeing in principle that a category-based approach might be useful, they questioned whether it would work in practice. It is difficult, they argued, to determine where boundaries between categories lie, how broad the categories should be, and what factors should govern which category a given case falls into. Is answering these questions really any easier than simply taking a case-by-case optimization approach, which Prof. Epstein rejected? This problem, the commenters pointed out, is particularly apparent in information privacy cases where rapid changes in technology make it very difficult to define the limits of government action. It is telling that the pen register cases deal with a technology that simply does not exist anymore. There are analogues, of course – telecom companies can now much more easily track connections – but should these analogues be treated the same way as pen registers? If the government, in principle, can easily find all the telecoms links made by any individual, should we still rely on the public benefit/private cost balance struck when such tracking was much more costly?