Skip to main content

Legal Conventionalism in the U.S. Constitutional Law of Privacy*

  • Mark Tushnet (a1)

Drawing on themes important in moral and political philosophy, much of the scholarship on the constitutional law of privacy in the United States distinguishes between privacy understood as a person's control over information and privacy understood as a person's ability to make autonomous decisions. For example, Katz v. United States (1967) established the framework for analyzing whether police activity constituted a “search” subject to the Fourth Amendment's requirement that the police either obtain a warrant before conducting a search or otherwise act reasonably. The defendant was a professional gambler who knew enough about police techniques to use a public telephone to make his business calls. Police agents attached a listening device to the outside of the phone booth, and sought to use the recordings against the defendant. The Supreme Court agreed with the defendant that the Fourth Amendment had been violated. Justice John Marshall Harlan's influential concurring opinion asserted that a person's privacy, in the sense of control over information, depended on two factors: “that a person have exhibited an actual (subjective) expectation of privacy and, second, that the expectation be one that society is prepared to recognize as ‘reasonable.’” Fourth Amendment cases like Katz involve informational control; they define the circumstances under which the government may acquire information from or about a person without first obtaining the person's agreement. In contrast, cases like Griswold v. Connecticut (1965), which barred the state from making it a criminal offense to use contraceptives, and Roe v. Wade (1973), which restricted the state's power to prohibit or regulate abortions, used the language of privacy rights to protect a much broader interest in autonomous decision-making. Seeing these cases and related ones through lenseees provided by moral and political philosophy, scholars have attempted to describe what a morally sound constitutional law of privacy would be, and the broadest sense, dworkinian. That is, they seek to provide an account of privacy with two characteristics: it is broadly consistent with the relevant constitutional decisions, and it is the most morally attractive account possible that satisfies the requirement of consistency with the decisions.

Hide All

1 These concepts may be connected: control over information shields decisions from public scrutiny, and thereby enables truly autonomous decision-making.

2 Katz v. United States, 389 U.S. 347 (1967). The text's statement of the Fourth Amendment's requirements is a gross oversimplification; for example, it elides a significant controversy over whether the probable cause requirement applies to searches conducted without a warrant. The Fourth Amendment states: “The right of the people to be secure in their persons, houses, papers, and effects against unreasonable searches and seizures, shall not be violated.…”

3 Ibid., at 360 (Harlan, J., concurring).

4 In this essay I consider the main lines of discussion of the relevant constitutional doctrines. William Stuntz and L. Michael Seidman have offered alternative accounts of the Fourth Amendment's normative basis. Stuntz stresses the inadequacy of doctrine focused on privacy-as-information-control in a world where the government's substantive regulatory power is unchecked; Seidman stresses the role of privacy in protecting against embarrassment and shame (as distinct from the more antiseptic release of information), and in ensuring that government investigative techniques intrude as little as possible and cause no collateral damage. See Stuntz William J., “Privacy's Problem and the Law of Criminal Procedure,” Michigan Law Review 93, no. 5 (03 1995): 1016–78; and Seidman Louis Michael, “The Problems with Privacy's Problem,” Michigan Law Review 93, no. 5 (03 1995): 1079–101.

5 Griswold v. Connecticut, 381 U.S. 479 (1965).

6 Roe v. Wade, 410 U.S. 113 (1973).

7 Constitutional questions regarding privacy can arise in other settings, which I do not examine here. For example, there may be questions about the constitutionality of legislation authorizing private entities to inquire into, or make use of, certain types of personal information. As a general matter, these questions would face the initial hurdle of showing that the legislation made the inquiries subject to constitutional restriction. Ordinarily the constitutional challenge would not get over that hurdle because of the “state-action” doctrine, according to which constitutional restrictions apply only to actions taken by the government. The legislation at issue would be subject to constitutional restriction, but the inquiries themselves would not. Ordinarily the inquiries would be exercises of nongovernmental power, derived from the private entity's general property rights, and typically, the fact that a private party derives its power from the general law of property is insufficient to trigger constitutional scrutiny of its actions. (The foregoing brutally simplifies the state-action doctrine, but for the present inquiry, the benefits of a more extensive explanation would be too small to be justifiable.)

8 For examples of works using Supreme Court cases and other cases to illustrate moral theories of privacy, see Allen Anita L., Uneasy Access: Privacy for Women in a Free Society (Totowa, NJ: Rowman & Littlefield, 1987); and DeCew Judith Wagner, In Pursuit of Privacy: Law, Ethics, and the Rise of Technology (Ithaca, NY: Cornell University Press, 1997).

9 For Dworkin's general account, see Dworkin Ronald, Law's Empire (Cambridge, MA: Harvard University Press, 1986).

10 As I discuss below, determining what these settled understandings are is itself an interpretive task, the performance of which may require some nonconventionalist resources.

11 The argument I develop was first made, I believe, by Bobbitt Philip, in Constitutional Fate (New York: Oxford University Press, 1982), esp. at 169–75. Bobbitt called the kind of argument that he and I describe ethical, a somewhat misleading term to the extent that it evokes ideas connected to moral philosophy. Bobbitt defined ethical argument as follows: “constitutional argument whose force relies on a characterization of American institutions and the role within them of the American people. It is the character, or ethos, of the American polity that is advanced in ethical argument as the source from which particular decisions derive” (ibid., 94).

12 In this essay I apply legal conventionalism to the constitutional issue of privacy to identify some characteristics of, and difficulties associated with, the general idea of legal conventionalism. In doing so I necessarily confine the argument to the kinds of conventions associated with privacy—roughly, social conventions about a person's material being in the world. Applying legal conventionalism to conventions about morality itself might raise different issues, although my intuition is that it does not. I do not defend that intuition here, however.

13 In the end, it might be possible to develop a normative defense of legal conventionalism, but I do not attempt to do so here. As noted immediately in the text, such a normative defense would have to confront a large number of issues—for example, questions about moral relativism—about which I can claim no particular insight.

14 My thought here is that the U.S. constitutional concept of privacy may be an aspect of the conventions of Western liberal democracies.

15 Kastigar v. United States, 406 U.S. 441 (1972).

16 A minor problem in attempting to link moral and political theories of privacy and its regulation to U.S. constitutional law is that privacy-as-information-control and privacy-as-decisional-autonomy might sometimes be connected, but U.S. constitutional law does not recognize the connection. Suppose that moral and political philosophy establish that governments may not bar women from obtaining abortions for certain reasons (the list of which may be quite long), but may do so when women seek abortions for other reasons. (The example usually given is abortion for sex-selection reasons.) Determining whether a woman had a morally permissible or a morally improper reason for obtaining an abortion would intrude on her privacy-as-information-control. U.S. constitutional law does not, however, acknowledge that the government may bar abortions when they are sought for specific reasons. Perhaps the U.S. approach may be understood as a way of protecting privacy-as-information-control: The only way to do so is to bar inquiries into reasons for obtaining an abortion, and the only way to do that is to allow women to obtain abortions without regard to their reasons.

17 I must confess, however, a lurking suspicion that any moral theory that does distinguish between sexual practices and business practices will not, ultimately, respect the autonomous decision-making of the person who happens to care much more about the analytically disfavored practices than the analytically favored ones. Consider, for example, how a person whose entire self-conception rests on success at business will fare under theories that make sexuality, but not business, central to their accounts of human flourishing or whatever they take to be morally significant.

18 Unger Roberto Mangabeira, The Critical Legal Studies Movement (Cambridge, MA: Harvard University Press, 1986), 3.

19 Dworkin does require that his account of adjudication fit existing practice. However, his characterization of that practice is interpretively constructed in light of his moral requirements, and so his account need not require that judges self-consciously apply moral/political theories.

20 The case was the last phase in a long effort by Connecticut's Planned Parenthood organization to invalidate Connecticut's statutory ban on using contraceptives. After failing to obtain a judicial ruling in the early 1940s and early in the 1960s without violating the statute, the organization opened a clinic and invited the prosecution.

21 Griswold, 381 U.S. at 484.

22 For a discussion of the metaphor's origins, and a brief criticism of its use, see Greely Henry T., “A Footnote to ‘Penumbra’ in Griswold v. Connecticut,” Constitutional Commentary 6, no. 2 (Summer 1989): 251–65.

23 Pierce v. Society of Sisters, 268 U.S. 510 (1925).

24 The First Amendment does explicitly protect association for a particular purpose when it protects the freedom of assembly for the purposes of petitioning the government for a redress of grievances. However, the amendment does not protect associations—even political associations—that take other forms.

25 This is Justice Douglas's analysis of NAACP v. Alabama, 357 U.S. 449 (1958). I should note that the approach I am imputing to Justice Douglas, while clearly grounded in Griswold, may also represent his effort to work around an approach to constitutional interpretation, the balancing of competing interests, with which he was fundamentally unsympathetic. In this connection it may not be irrelevant that the decision in NAACP v. Alabama was written by Justice Harlan, the Supreme Court's leading proponent of balancing.

26 For a pictorial presentation of this point, see Tushnet Mark, “Two Notes on the Jurisprudence of Privacy,” Constitutional Commentary 8, no. 1 (Winter 1991): 77.

27 Fit between prior decided cases and the concept of privacy is guaranteed by Justice Douglas's construction.

28 Griswold, 381 U.S. at 485–86. Note that Justice Douglas's question rests on the proposition that the Constitution does not protect privacy-as-information-control as such. If the state could make it a criminal offense to use contraceptives, the Fourth Amendment would allow a search to discover evidence of the offense upon a proper showing, either of reasonableness or of probable cause.

29 I acknowledge, however, the Catholic natural law tradition that does give marriage a specific and rather detailed content.

30 Eisenstadt v. Baird, 405 U.S. 438 (1972).

31 Ibid., at 453.

33 Roe, 410 U.S. at 153.

34 Katz, 389 U.S. at 351.

36 Dow Chemical Co. v. United States, 476 U.S. 227 (1986).

37 United States v. Place, 462 U.S. 696 (1983).

38 United States v. Knotts, 460 U.S. 276 (1983); United States v. Karo, 468 U.S. 705 (1984).Karo holds in addition that police need a warrant if they use a beeper to determine that the article containing the beeper remains at the place to which it was delivered.

39 Knotts, 460 U.S. at 281.

40 Ibid., at 282.

41 For a discussion, see Nissenbaum Helen, “Protecting Privacy in an Information Age: The Problem of Privacy in Public,” Law and Philosophy 17, nos. 5 and 6 (11 1998): 559–96.

42 Karo, 468 U.S. at 714.

43 Ibid., at 726 (O'Connor, J., concurring in the judgment).

44 California v. Greenwood, 486 U.S. 35 (1988).

45 Ibid., at 40.

46 Ibid., at 43 n. 5.

47 Ibid., at 51–52 (Brennan, J., dissenting).

48 California v. Ciraolo, 476 U.S. 207 (1986).

49 According to the Chief Justice, the fences would not “shield these plants from the eyes of… a policeman perched on top of a truck or a two-level bus” (ibid., at 211).

50 Ibid., at 215.

51 Ibid., at 223–24 and note 8 (Powell, J., dissenting).

52 I do not contend, however, that the Court's descriptions of the American people's settled understandings are always correct. Decisions allowing searches of areas around houses—technically, their curtilages—sometimes involve intrusions that I think many people would find to be violations of their sense of privacy. See, e.g., Oliver v. United States, 466 U.S. 170 (1984), where the police walked around a locked gate that had a “No Trespassing” sign on it, and Florida v. Riley, 488 U.S. 455 (1989), where the police observed property by hovering in a helicopter at four hundred feet above the property.

53 Johnson v. United States, 333 U.S. 10, 14 (1948) (referring to the fact that police officers are “engaged in the often competitive enterprise of ferreting out crime”).

54 An interesting case from the Netherlands illustrates the argument. The Netherlands Constitution expressly provides that none of its provisions (other than the one requiring compliance with international agreements) are enforceable in Dutch courts (Netherlands Constitution, art. 120). The Netherlands Supreme Court nonetheless invalidated a practice in which jail authorities placed detainees under continuous television surveillance as an infringement of a constitutionally identified right of privacy. (The decision is described in Bulletin on Constitutional Case-law [Strasbourg] 1996, no. 1 [1996]: 54.) The Court held that police authorities had to have specific authorization for a practice that so seriously intruded on privacy interests. (The British quasi-constitutional law of ultra vires operates in much the same way.)

55 Ely John Hart, Democracy and Distrust: A Theory of Judicial Review (Cambridge, MA: Harvard University Press, 1980).

56 Working out a representation-reinforcing account of the Fourth Amendment is, however, quite difficult. Those adversely affected by a legislatively authorized search policy may be in a minority, but as long as they can vote, they are in a position to participate in a pluralist bargaining process that can be expected to take account of their interests. For a discussion of this difficulty, see Tushnet Mark, Red, White, and Blue: A Critical Analysis of Constitutional Law (Cambridge, MA: Harvard University Press, 1988), 9698.

57 For a description of the political setting in Connecticut, see Garrow David J., Liberty and Sexuality: The Right to Privacy and the Making of Roe v. Wade (New York: Macmillan, 1994), 125–29.

58 Perhaps the legal conventionalist would shift attention to the Constitution itself, and contend that the conventional understanding of the Constitution is that it identifies the situations in which a national solution is appropriate. Even that move seems unhelpful, however, because we can concede its validity and still wonder whether the practice in question is one of the situations the Constitution, conventionally understood, singles out for national resolution.

59 For the most extensive early discussions of the constitutionality of this use of metal detectors, see United States v. Bell, 464 F.2d 667 (2d Cir. 1972); United States v. Davis, 482 F.2d 893 (9th Cir. 1973); and United States v. Edwards, 498 F.2d 496 (2d Cir. 1974).

60 Suppose a new technology became available which allowed users to detect more details about what a person was carrying, and that the technology was subsequently deployed not just in airports, but in other areas as well. The film Total Recall depicts such a technology used at underground transportation systems, such as subways. The conditions under which governments might require using screening devices in subways are not hard to imagine. The privacy-based objection to this use of screening technology would not be frivolous today, though it might be rejected. I note that Justice John Paul Stevens, dissenting in Michigan Department of Public Safety v. Sitz, 496 U.S. 444, 473–74 (1990), wrote: “I would suppose that all subway passengers could be required to pass through metal detectors.”

61 I would not of course overestimate the contribution that court decisions make to changes in the nation's understandings of privacy. My only claim here is that such decisions have some effect.

62 Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833 (1992). Two other justices concurred in the judgment; four justices dissented. The opinion discussed in the text is therefore not accurately described as a plurality opinion, a term that, in cases in which there is no majority opinion, refers to the opinion that received the most votes; the standard designation has become joint opinion.

63 See ibid., at 861 (referring to the “personal reluctance any of us may have” in “affirming Roe's central holding”), 836 (referring to Roe's “error, if error there was”).

64 Ibid., at 856.

67 To the extent one is concerned about the costs of restructuring, one might phase in the legal change, for example, by invalidating legislation that does not contain a sufficiently long transition period.

68 It is probably worth noting here that a later section of the joint opinion, which invalidated a provision requiring a married woman to notify her husband before she obtained an abortion, relied heavily on factual findings about the prevalence of domestic violence (Casey, 505 U.S. at 887–98). I believe that the joint opinion in Casey is the first opinion in any of the Court's abortion decisions to rely explicitly on equality reasons for limiting the government's power to regulate abortion.

69 Analytically, one could go outside legal conventionalism to something other than moral or political theory. For example, one could rely on original understanding. However, the legal conventionalist relies on the people's understandings precisely in those cases where original understanding and other sources of constitutional decision-making are insufficient for the decision-maker's purposes.

70 It upheld the search nonetheless, on the ground that the warrant the police obtained was based on adequate information even when one omitted the information the police provided that was based upon the constitutional violation.

71 Karo, 468 U.S. at 735 (Stevens, J., concurring in part and dissenting in part).

72 The Court remains divided on the question of whether legal conventionalism can appropriately supplement tradition. For an exchange, in which a majority allows supplementation over a vigorous protest by Scalia Justice Antonin, see County of Sacramento v. Lewis, 523 U.S. 833 (1998).

73 Washington v. Glucksberg, 117 S.Ct. 2258 (1997); Vacco v. Quill, 117 S.Ct. 2293 (1997).

74 Glucksberg, 117 S.Ct. at 2268.

75 See, e.g., McConnell Michael, “The Right to Die and the Jurisprudence of Tradition,” Utah Law Review 1997, no. 3 (1997): 665708.

76 See Lewis.

77 Casey, 505 U.S. at 849.

79 I explore this suggestion in “The New Constitutional Order and the Chastening of Constitutional Aspiration,” Harvard Law Review 113, no. 1 (11 1999): 29109.

* As usual, I am indebted to L. Michael Seidman for insights that I have drawn on in this essay.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

Social Philosophy and Policy
  • ISSN: 0265-0525
  • EISSN: 1471-6437
  • URL: /core/journals/social-philosophy-and-policy
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Full text views

Total number of HTML views: 0
Total number of PDF views: 16 *
Loading metrics...

Abstract views

Total abstract views: 92 *
Loading metrics...

* Views captured on Cambridge Core between September 2016 - 17th January 2018. This data will be updated every 24 hours.