Probability thresholds

Iowa Law Review

Volumn 92, Issue 4, 2007, Pages 1293-1357

  Masur, Jonathan S ^a  

^a UNIVERSITY OF CHICAGO   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 34548238156     PISSN: 00210552     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Review

References (402)

1. The Judge's Only Obligation Is to the Rule of Law, N.Y. TIMES, Jan. 10, 2006, at A1 (quoting the opening statement at now Supreme Court Justice Alito's Senate confirmation hearing).
2. RON SUSKIND, THE ONE PERCENT DOCTRINE 61-62 (2006) (quoting Vice President Cheney's observations during a CIA briefing). My thanks to Cass Sunstein for bringing this quote to my attention.
3. U.S. CONST, amend. I.
4. See, e.g., N.Y. Times Co. v. United States, 403 U.S. 713, 717-18 (1971) (Black, J., concurring) ("[N]o law means no law.").
5. Kennedy v. Mendoza-Martinez, 372 U.S. 144, 160 (1963).
6. Near v. Minnesota, 283 U.S. 697, 716 (1931).
7. See generally GEOFFREY R. STONE ET AL., THE FIRST AMENDMENT (2d ed. 2003);
8. LAURENCE TRIBE, AMERICAN C ONSTITUTIONAL LAW §§ 12-1 to 12-39 (2d ed. 1988). Indeed, it is the taxonomy of these cases - their categorization as one type or another - that is most frequendy decisive to the analysis. STONE ET AL., supra, at 18.
9. See infra notes 13-20 and accompanying text.
10. See infra notes 11, 13-16 and accompanying text.
11. Some commentators reject this essentially consequentialist view in favor of an approach that draws more substantially upon deontological moral and philosophical norms. See, e.g., LARRY ALEXANDER, IS THERE A RIGHT OF FREEDOM OF EXPRESSION? 55-81 (2005);
12. David A. Strauss, Persuasion, Autonomy, and Freedom of Expression, 91 COLUM. L. REV. 334, 352-60 (1991).
13. See, e.g., Center for Nat'l Sec. Studies v. U.S. Dep't of Justice, 331 F.3d 918, 925-27 (D.C. Cir. 2003);
14. N.Jersey Media Group, Inc. v. Ashcroft, 308 F.3d 198, 208-09 (3d Cir. 2002);
15. Detroit Free Press v. Ashcroft, 303 F.3d 681, 705 (6th Cir. 2002);
16. United States v. Progressive, Inc., 467 F. Supp. 990, 993 (W.D. Wis. 1979).
17. See infra notes 271-72.
18. See generally RICHARD A. EPSTEIN, CASES AND MATERIALS ON TORTS (7th ed. 2003);
19. RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW (6th ed. 2003).
20. See generally Cass R. Sunstein, The Arithmetic of Arsenic, 90 GEO. L.J. 2255 (2002) (discussing government adoption of public regulation of arsenic in drinking water).
21. Exec. Order No. 12,866, 58 Fed. Reg. 51,735 (Sept. 30, 1993);
22. Exec. Order No. 12,498, 50 Fed. Reg. 1036 (Jan. 4, 1985);
23. Exec. Order No. 12,291, 46 Fed. Reg. 13,193 (Feb. 17, 1981);
24. see also Lisa Schultz Bressman, Beyond Accountability: Arbitrariness and Legitimacy in the Administrative State, 78 N.Y.U. L. REV. 461, 487-88 (2003) (discussing President Reagan's deployment of cost-benefit analysis to support a "primarily antiregulatory agenda").
25. See Richard H. Fallon, Jr., The Supreme Court, 1996 Term - Foreword: Implementing the Constitution, 111 HARV. L. REV. 54, 78-79 (1997).
26. See, e.g., Bartnicki v. Vopper, 532 U.S. 514, 526-35 (2001) (explicidy employing a balancing approach in deciding a First Amendment issue);
27. Buckley v. Valeo, 424 U.S. 1, 64-68 (1976) (same);
28. Miller v. City of S. Bend, 904 F.2d 1081, 1088 (7th Cir. 1990) (Posner, J., concurring) (same), rev'd sub nom., Barnes v. Glen Theatre, Inc., 501 U.S. 560 (1991).
29. See, e.g., Republican Party of Minn. v. White, 536 U.S. 765, 777-88 (2002);
30. Burson v. Freeman, 504 U.S. 191, 196-211 (1992).
31. See infra Part II.A-B.
32. The second and third of these doctrinal incorporations are best conceived as variants of the first; in each case, the value of the constitutional right - which is often difficult or impossible to measure on a case-by-case basis - is set at a fixed level, and the court measures the expected harm against that baseline.
33. The typical example is of a speech act triggering a thermonuclear war that in turn destroys all civilization and life on the planet. Richard Posner conservatively estimates the value of such an "extinction event" at $600 trillion. RICHARD A. POSNER, CATASTROPHE: RISK AND RESPONSE 170 (2004).
34. For a small sampling of this literature, see, for example, Neil K. Komesar, Taking Institutions Seriously: Introduction to a Strategy for Constitutional Analysis, 51 U. CHI. L. REV. 366, 408-09 (1984);
35. Richard A. Posner, Pragmatism Versus Purposivism in First Amendment Analysis, 54 STAN. L. REV. 737, 738-42 (2002) ;
36. Mark C. Rahdert, The First Amendment and Media Rights During Wartime: Some Thoughts After Operation Desert Storm, 36 VILL. L. REV. 1513, 1517-18 (1991);
37. Shira J. Schlaff, Using an Eruv to Untangle the Boundaries of the Supreme Court's Religion Clause Jurisprudence, 5 U. PA. J. CONST. L. 831, 877-78 (2003) ;
38. Rodney A. Smolla, Free the Fortune 500! The Debate over Corporate Speech and the First Amendment, 54 CASE W. RES. L. REV. 1277, 1278 n.11 (2004);
39. cf. GEOFFREY R. STONE, PERILOUS TIMES: FREE SPEECH IN WARTIME FROM THE SEDITION ACT OF 1798 TO THE WAR ON TERRORISM 546 (2004). This is the case even among those commentators who advocate a "thumb on the scale" in favor of speech on account of behavioral tendencies to overestimate harm.
40. See, e.g., CASS R. SUNSTEIN, LAWS OF FEAR 218-23 (2005);
41. Paul Horwitz, Free Speech as Risk Analysis: Heuristics, Biases, and Institutions in the First Amendment, 76 TEMP. L. REV. 1, 27-49 (2003).
42. For a small sampling of this literature, see generally Matthew D. Adler & Eric A. Posner, Rethinking Cost-Benefit Analysis, 109 YALE L.J. 165 (1999).
43. See also James A. Goldston et al., A Nation Less Secure: Diminished Public Access to Information, 21 HARV. C.R.-C.L. L. REV. 409, 444 (1986);
44. Cass R. Sunstein, Cost-Benefit Default Principles, 99 MICH. L. REV. 1651, 1654 (1999);
45. n factor represents the temporal discounting. The model employed in this Article subsumes O and A within the harm calculation and brackets the issue of discounting; in all other respects, it is identical to Posner's.
46. W. Kip Viscusi & Richard J. Zeckhauser, Sacrificing Civil Liberties to Reduce Terrorism Risks, 26 J. RISK & UNCERTAINTY 99, 101 (2003). Note that there is no similarly long tail of low-magnitude, high-probability events, as the probability of any occurrence is bounded at "1."
47. See infra Part III.
48. See infra Part III.
49. Part III, infra, examines the rhetoric defining how low a probability must be for it to fall beneath the threshold.
50. See generally FRANK H. KNIGHT, RISK, UNCERTAINTY, AND PROFIT (1921). The importance of this assumption will become clear in Part III. In short, the likely harm from an event of known probability is easily quantifiable and could be measured against any desired baseline without fear of systematic bias.
51. See id. at 197-232 (describing the operation of true uncertainty).
52. Posner, supra note 22, at 740.
53. Schenck v. United States, 249 U.S. 47, 52 (1919).
54. Brandenburg v. Ohio, 395 U.S. 444, 448-49 (1969) (per curiam).
55. Chaplinsky v. New Hampshire, 315 U.S. 568, 573 (1942).
56. See, e.g., Brandenburg, 395 U.S. at 444;
57. see also infra Part III.A.
58. See, e.g., N.Y. Times Co. v. United States, 403 U.S. 713, 726-27 (1971) (Brennan, J., concurring);
59. see also infra Part III.C (discussing probability thresholds within the doctrine of prior restraints).
60. Brandenburg, 395 U.S. 444.
61. Schenck v. United States, 249 U.S. 47 (1919).
62. Another First Amendment doctrine used to negotiate attempts to ban speech because of its harmful consequences - and another potential locus of a probability threshold - is the doctrine of secondary effects, which governs cases in which the government wishes to prohibit an adult business from locating in a particular area due to the crime it may induce.
63. See City of Erie v. Pap's A.M., 529 U.S. 277, 314 (2000) (Souter, J., concurring in part and dissenting in part) ("What is clear is that the evidence . . . must be a matter of demonstrated fact, not speculative supposition.").
64. Brandenburg, 395 U.S. at 446. A speech at an immediately prior rally included such comments as "Send the Jews back to Israel," and "Bury the n_. We intend to do our part." Id. (alteration added).
65. Id. at 447.
66. Id.
67. See Jed Rubenfeld, A Reply to Posner, 54 STAN. L. REV. 753, 758 n.23 (2002) ("In theory, a cost-benefit analysis should be concerned, on the cost side, only with the magnitude of the harm (with which the Brandenburg doctrine is not concerned at all) and its probability.");
68. Richard A. Posner, Free Speech in an Economic Perspective, 20 SUFFOLK U. L. REV. 1, 37 (1986) (offering a similar description).
69. Brandenburg, 395 U.S. at 447.
70. See Bernard Schwartz, Holmes Versus Hand: Clear and Present Danger or Advocacy of Unlawful Action?, 1994 SUP. CT. REV. 209, 240-41 (noting these two requirements and adding a third: that the law sanction only "express advocacy"). Schwartz's third requirement is of limited relevance to the analysis here.
71. Brandenburg, 395 U.S. at 447.
72. For discussions on the importance of accurate discounting and the issues involved therein, see W. Kip Viscusi & Joseph E. Aldy, The Value of a Statistical Life: A Critical Review of Market Estimates Throughout the World, 27 J. RISK & UNCERTAINTY 5, 57 (2003) ;
73. Dexter Samida & David A. Weisbach, Paretian Intergenerational Discounting 28-29 (Univ. of Chi. Law School, John M. Olin Program in Law & Economics Working Papers, No. 255, 2005), available at http:// www.law.uchicago.edu/law-pdf/law-econ/255.pdf.
74. See STONE, supra note 22, at 409.
75. Brandenburg, 395 U.S. at 445-46.
76. See William Cheek & Aimee Lee Cheek, John Mercer Langston and the Cincinnati Riot of 1841, in RACE AND THE CITY 29-69 (Henry Louis Taylor, Jr., ed., 1993) (describing the Cincinnati race riots of 1967);
77. Henry Louis Taylor, Jr., Introduction: Race and the City, 1820-1970, in RACE AND THE CITY, supra, at 19-21 (same).
78. Schenck v. United States, 249 U.S. 47, 52 (1919).
79. Id.
80. Id.
81. Id.
82. TRIBE, supra note 7, § 12-9, at 841-49.
83. Id.
84. RICHARD A. POSNER, LAW, PRAGMATISM, AND DEMOCRACY 360 (2003). Posner uses danger to mean expected harm, i.e., the probabilistic calculation of the expected outcome.
85. Id. at 361;
86. see also Posner, supra note 23, at 124-25;
87. Posner, supra note 42, at 29.
88. POSNER, supra note 56, at 361.
89. Dennis v. United States, 341 U.S. 494, 508-11 (1951).
90. Id. at 496 (quoting the Smith Act, 18 U.S.C. §§ 10-11 (1946)).
91. Id. at 508 ("In this case we are squarely presented with the application of the 'clear and present danger' test, and must decide what that phrase imports.").
92. United States v. Dennis, 183 F.2d 201, 212 (2d Cir. 1950).
93. Kent Greenawalt, Speech and Crime, 1980 AM. B. FOUND. RES. J. 645, 717 (1980) (describing Dennis in cost-benefit terms);
94. Posner, supra note 42, at 8 (same);
95. cf. United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir. 1947) (introducing the classic B/PL formulation).
96. Dennis, 341 U.S. at 510.
97. See JOHN HART ELY, DEMOCRACY AND DISTRUST: A THEORY OF JUDICIAL REVIEW 115 (1980) (arguing that Brandenburg would "have demanded the contrary result in the early Espionage Act cases and the later Communist cases" (emphasis added));
98. STONE ET AL., supra note 7, at 61;
99. Gerald Gunther, Learned Hand and the Origins of Modern First Amendment Doctrine: Some Fragments of History, 27 STAN. L. REV. 719, 754 (1975) (describing Brandenburg as the Supreme Court's "clearest and most protective standard under the first amendment" while noting that the crucial innovation in that case was "hardly established" in prior law).
100. Neb. Press Ass'n v. Stuart, 427 U.S. 539, 562 (1976).
101. Bailey v. Systems Innovation, Inc., 852 F.2d 93, 98 (3d Cir. 1988) ;
102. Culver v. Sec'y of the Air Force, 559 F.2d 622 (D.C. Cir. 1977);
103. Gay Lib v. Univ. of Mo., 558 F.2d 848 (8th Cir. 1977);
104. United States v. Smith, 555 F.2d 249, 253 n.3 (9th Cir. 1977);
105. Sherling v. Townley, 464 F.2d 587 (5th Cir. 1972).
106. NAACP v. Claiborne Hardware Co., 458 U.S. 886 (1982).
107. Id. at 902.
108. Id. at 904.
109. Id. at 928.
110. The Court's estimation of harm may well have been biased improperly by hindsight. Even if direct harm did not result from Evers's speech, it does not follow that harm was unlikely to occur; in the context in which Evers gave the speech, it might have been quite evident to any bystander that Evers had whipped the crowd into a violent frenzy. Of course, this issue - post hoc examination of causation and probability - is yet another indication of the Court's heavy emphasis on probability thresholds. See infra Part III.C (examining probability thresholds within the doctrine of prior restraints).
111. Hess v. Indiana, 414 U.S. 105, 106 (1973).
112. Id. at 107.
113. Id. at 107-08 (emphasis added).
114. Id. at 109 (citing Brandenburg v. Ohio, 395 U.S. 444, 447 (1969) (per curiam)).
115. See, e.g., N.Y. Times Co. v. United States, 403 U.S. 713, 726 (1971) (Brennan, J., concurring) (explaining that the framework would not be used only in an "extremely narrow class of cases");
116. TRIBE, supra note 7, § 12-9, at 841-48.
117. See, e.g., Virginia v. Black, 538 U.S. 343, 359 (2003) (concerning cross-burning);
118. Ashcroft v. Free Speech Coal., 535 U.S. 234, 236 (2002) (concerning indecent speech harmful to children).
119. Chaplinsky v. New Hampshire, 315 U.S. 568 (1942).
120. Id. at 572-73.
121. Id. at 572.
122. Cohen v. California, 403 U.S. 15 (1971).
123. Id. at 20.
124. Id.
125. Id. at 25.
126. I take no position here on whether the First Amendment, under any sort of balancing scheme, ought to include these types of potentialities in the cost-benefit equation. I mean only to indicate that the Cohen Court might have decided to do so.
127. For a canonical statement of the doctrine, see Org. for a Better Austin v. Keefe, 402 U.S. 415, 419-20 (1971).
128. See, e.g., Thomas I. Emerson, The Doctrine of Prior Restraint, 20 LAW & CONTEMP. PROBS. 648, 650-52 (1955) (including sources cited therein);
129. William T. Mayton, Toward a Theory of First Amendment Process: Injunctions of Speech, Subsequent Punishment, and the Costs of the Prior Restraint Doctrine, 67 CORNELL L. REV. 245, 247-49 (1982).
130. John Calvin Jeffries, Jr., Rethinking Prior Restraint, 92 YALE L.J. 409, 412 (1983).
131. WILLIAM BLACKSTONE, 4 COMMENTARIES *152.
132. TRIBE, supra note 7, § 12-34, at 1039;
133. see also Near v. Minnesota ex rel. Olson, 283 U.S. 697, 713 (1931);
134. Emerson, supra note 88, at 650-52;
135. Mayton, supra note 88, at 247.
136. See, e.g., N.Y. Times Co. v. United States, 403 U.S. 713, 717 (1971) (Black, J., concurring). Justice Black argued: In the First Amendment the Founding Fathers gave the free press the protection it must have to fulfill its essential role in our democracy. The press was to serve the governed, not the governors. The Government's power to censor the press was abolished so that the press would remain forever free to censure the Government. The press was protected so that it could bare the secrets of government and inform the people. Only a free and unrestrained press can effectively expose deception in government. Id.
137. See Jeffries, supra note 89, at 421.
138. See id. at 421-26;
139. cf. Richard A. Epstein, The Permit Power Meets the Constitution, 81 IOWA L. REV. 407, 411-14 (1995). The litany of objections to an administrative-preclearance system is well documented, and, for present purposes, there is no need to explore them at length. Suffice it to say that even the harshest critics of broader prior-restraint doctrines accept the validity of the doctrine in that context.
140. See, e.g., Marin Scordato, Distinction Without a Difference: A Reappraisal of the Doctrine of Prior Restraint, 68 N.C. L. REV. 1, 6-7 (1989).
141. For instance, "Judges are not perfect; sometimes they may err on the side of suppression and enjoin speech without sufficient justification. But the fact remains that judges, unlike professional censors, have no vested interest in the suppression of speech." Jeffries, supra note 89, at 426-27.
142. See, e.g., id. at 427-30 (noting that post hoc punishment is very similar in structure to an injunction against speech and, in fact, may chill a broader swath of conduct);
143. Scordato, supra note 95, at 15-16 ("Taken as a whole, this analysis strongly suggests that laws traditionally characterized as prior restraints are no more likely to produce a greater or more effective chilling effect on constitutionally protected speech than laws traditionally characterized as subsequent sanctions.").
144. Jeffries, supra note 89, at 428.
145. The following discussion concerns only injunctive remedies obtained in court and not any types of administrative-licensing schemes, which are objectionable for a multitude of unrelated reasons. See supra notes 88-91 and accompanying text. Accordingly, the term "prior restraint," wherever it is subsequently used in this paper, should be understood to describe only injunctions against speech.
146. See, e.g., Se. Promotions., Ltd. v. Conrad, 420 U.S. 546, 547-48 (1975) (concerning the musical "Hair").
147. See, e.g., N.Y. Times Co. v. United States, 403 U.S. 713, 714 (1971) (per curiam);
148. Near v. Minnesota ex rel. Olson., 283 U.S. 697, 704-05 (1931);
149. United States v. Progressive, Inc., 467 F. Supp. 990, 993 (W.D. Wis. 1979).
150. See, e.g., Near, 283 U.S. at 705 ("The defendants demurred to the complaint upon the ground that it did not state facts sufficient to constitute a cause of action, and on this demurrer challenged the constitutionality of the statute.").
151. See, e.g., Walker v. City of Birmingham, 388 U.S. 307, 319-21 (1967).
152. See TRIBE, supra note 7, § 12-35, at 1042 & n.2.
153. Brandenburg v. Ohio, 395 U.S. 444, 444-45 (1969) (per curiam);
154. Schenck v. United States, 249 U.S. 47, 48-49 (1919).
155. Brandenburg, 395 U.S. at 445.
156. The contempt proceeding is by nature abbreviated, since there remains only one live issue.
157. This means only one opportunity before the trial court. In either case the speaker may appeal the trial court's decision - either the decision to grant the injunction under the priorrestraint regime or the conviction in the post hoc punishment regime-to the court of appeals and eventually, perhaps, to the Supreme Court.
158. This assumes, of course, that the speaker can gauge, ex ante, the probability that her speech will be prohibited; errors in this estimation may well lead to over- or underdeterrence. The goal nevertheless remains to achieve perfect deterrence, and the legal sanction is presumably calibrated (to the extent possible) toward this end.
159. See, e.g., N.Y. Times Co. v. United States, 403 U.S. 713, 714 (1971) (per curiam);
160. United States v. Progressive, Inc., 467 F. Supp. 990, 993 (W.D. Wis. 1979).
161. Cf. Guido Calabresi & A. Douglas Melamed, Property Rules, Liability Rules, and Inalienability: One View of the Cathedral, 85 HARV. L. REV. 1089, 1111-12 (1972). Per Calabresi and Melamed's dichotomy, this is to say that a speaker need not bargain with the party who has obtained the injunction for a release of that injunction before speaking. She may simply speak and then absorb the punishment for having done so. Id.
162. N.Y. Times Co., 403 U.S. at 714 (quoting Bantam Books, Inc. v. Sullivan, 372 U.S. 58, 70 (1963) and citing Near v. Minnesota ex rel. Olson, 283 U.S. 697 (1931)).
163. Org. for a Better Austin v. Keefe, 402 U.S. 415, 419 (1971).
164. Vincent Blasi, Toward a Theory of Prior Restraint: The Central Linkage, 66 MINN. L. REV. 11, 11 (1981).
165. See generally id.;
166. Howard O. Hunter, Toward a Better Understanding of the Prior Restraint Doctrine: A Reply to Professor Mayton, 67 CORNELL L. REV. 283 (1982);
167. Jeffries, supra note 89;
168. Martin H. Redish, The Proper Role of the Prior Restraint Doctrine in First Amendment Theory, 70 VA. L. REV. 53 (1984);
169. Scordato, supra note 95.
170. See supra note 115 (cataloging an extensive list of possible explanations, none particularly compelling for reasons stated within the sources themselves).
171. Blasi, supra note 114, at 49-54;
172. see also Jeffries, supra note 89, at 430 n.67 ("Professor Blasi's analysis of these factors is both thoughtful and thought-provoking. . . . In my view, the strongest of these arguments goes under the heading of 'adjudication in the abstract.'");
173. Redish, supra note 115, at 66-69 ("Perhaps the strongest argument against judicial prior restraint is that because such a restraint is imposed prior to the actual dissemination of expression, a court's first amendment ruling will necessarily be made in the abstract without any knowledge of the actual effect of the challenged expression.").
174. See United States v. Dennis, 183 F.2d 201, 212 (2d Cir. 1950) ("We can never forecast with certainty; all prophecy is a guess, but the reliability of a guess decreases with the length of the future which it seeks to penetrate.").
175. See Blasi, supra note 114, at 49-51. The executive certainly need not wait until the harm has actually occurred before punishing someone for a speech act.
176. See generally, e.g., Dennis v. United States, 341 U.S. 494 (1951). But even if it is the case that post hoc trial and punishment will frequendy occur "in the abstract,"
177. see Redish, supra note 115, at 68-69, at least the court will have before it much more information than it did before the speech took place.
178. Blasi, supra note 114, at 49.
179. N.Y. Times Co. v. United States, 403 U.S. 713, 714 (1971) (per curiam).
180. Id. at 714 (internal quotation marks and citation omitted).
181. Id. at 725-26 (Brennan, J. concurring) (emphasis added).
182. Id. at 726-27 (emphasis added).
183. Id. at 730 (Stewart, J., concurring).
184. N.Y. Times Co., 403 U.S. at 730 (Stewart, J., concurring).
185. See, e.g., Amoco Prod. Co. v. Gambel, Alaska, 480 U.S. 531, 540 (1987).
186. N.Y. Times Co., 403 U.S. at 732 (White, J., concurring). For at least Justices White and Stewart, a large part of the weakness of the government's case stemmed from this untenable standard it suggested.
187. See id. ("Much of the difficulty inheres in the 'grave and irreparable danger' standard suggested by the United States.").
188. It is conceivable that by "grave" the government meant instead a proper cost-benefit analysis of the statistically expected harm, as measured by the product of its magnitude and probability. Yet, were this the case, it would lend further credence to the argument set forth in the text above, as the Court would have rejected not merely an unworkable and shortsighted doctrinal formula that focused exclusively on a harm's magnitude but, instead, a genuine attempt to perform classical cost-benefit analysis.
189. There is no doubt that this choice was squarely before the Justices. N. Y. Times, 403 U.S. at 732 (White, J., concurring). Justice White wrote: The Government's position is simply stated: The responsibility of the Executive for the conduct of the foreign affairs and for the security of the Nation is so basic that the President is entided to an injunction against publication of a newspaper story whenever he can convince a court that the information to be revealed threatens "grave and irreparable" injury to the public interest . . . . Id.
190. In parallel with the violence doctrines, this probability threshold is not a modern innovation. The canonical case on prior restraint - Near v. Minnesota - describes the doctrine in significandy less detail but similarly announces that only high-probability threats will trump the Court's baseline antipathy toward prior restraint. Near v. Minnesota ex rel. Olson, 283 U.S. 697, 716 (1931). The Court wrote in Near, "[T]he protection even as to previous restraint is not absolutely unlimited." Id. But that protection lapses only in the face of concrete, highly probable events: "No one would question but that a government might prevent actual obstruction to its recruiting service or the publication of the sailing dates of transports or the number and location of troops." Id. (emphasis added). Absent is any sense that a low-probability harm, even one of enormous magnitude, would legitimize a prior restraint.
191. See SUNSTEIN, supra note 22, at 220-21;
192. see also infra Part IV.A.
193. See, e.g., Cass R. Sunstein, Terrorism and Probability Neglect, 26 J. RISK Sc UNCERTAINTY 121, 131 (2003);
194. see also infra Part IV.B.
195. See infra Part IV.B.
196. See, e.g., Cass R. Sunstein, Probability Neglect: Emotions, Worst Cases, and Law, 112 YALE L.J. 61, 71 (2002);
197. see also infra Part IV.C.
198. See Sunstein, supra note 135, at 64.
199. See infra Part IV.C.
200. See POSNER, supra note 56, at 316. Judge Posner argues that the President (and presumably the executive-branch national-security apparatus as a whole) is the only actor equipped to evaluate the threat posed by terrorists and civil unrest. Id.
201. See infra Part IV.B.
202. Eric A. Posner & Adrian Vermeule, Accommodating Emergencies, 56 STAN. L. REV. 605, 609 (2003). Posner and Vermeule argue that there is no evidence to suggest that the executive branch does any worse of a job at gauging the correct programs to undertake, or of balancing harms and benefits, during emergencies or wartime than at any other time. Id.
203. Id.
204. Mark Tushnet, Issues of Method in Analyzing the Policy Response to Emergencies, 56 STAN. L. REV. 1581, 1589 (2004).
205. Though, this characterization seems entirely plausible. Tushnet, in a reply to Posner and Vermeule's article on emergencies, supra note 140, posits this type of opportunistic behavior as the causal mechanism behind anti-liberty governmental over-reaching during wartime, Tushnet, supra note 142, at 1590, and his account has a ring of truth to it. Executive actors have preferences of all types that they wish to satisfy, and if the opportunity to satisfy a preference arrives only in the context of other, unrelated events, it is difficult to envision why the executive actor would not avail himself of this opportunity. Posner and Vermeule do not dispute Tushnet's characterization; they add only that the expanded realm of politically feasible actions may include just as many new good governmental programs as bad ones. Eric A. Posner & Adrian Vermeule, Emergencies and Political Change: A Reply to Tushnet, 56 STAN. L. REV. 1593, 1594-95 (2004).
206. It is costly for the executive to act to curtail speech-whether it is actually suing for an injunction or padlocking a newspaper. Thus, pursuant to this model, it is necessary to assume only that the executive will take action to block speech if, and only if, it genuinely believes that the speech fails the cost-benefit test.
207. Not only will the government undoubtedly choose to make its own case (rather than making the opposition's case in equally strong fashion), but after having already concluded that the speech restriction is justified, it likely will have anchored itself to that conclusion and will view the facts with that perspective in mind. See Andrew J. Wistrich, et al., Can Judges Ignore Inadmissible Information? The Difficulty of Deliberately Disregarding, 153 U. PA. L. REV. 1251, 1286-89 (2005). Countervailing facts that might have led an outside observer to doubt the executive's conclusions will be glossed over; minor facts that do not lend significant support to those conclusions may take on new life. This is the very nature of advocacy; once the course of action has been decided, the advocate becomes convinced of the rectitude of the cause.
208. See generally STONE, supra note 22;
209. David Cole, Enemy Aliens, 54 STAN. L. REV. 953 (2002) (noting that as a historical matter, the United States has frequently overreacted in times of crisis).
210. STONE, supra note 22, at 37 (internal quotation maks and endnote omitted).
211. Id. at 107 (internal quotation marks and endnote omitted).
212. Id. at 171-73.
213. Id. at 334, 365 (internal quotation marks and endnote omitted).
214. Id. at 442 (internal quotation marks and endnote omitted).
215. See generally N.Y. Times Co. v. United States, 403 U.S. 713 (1971).
216. STONE, supra note 22, at 512 n.* (quoting Erwin Griswold, Secrets Not Worth Keeping, WASH. POST, Feb. 15, 1989, at A25).
217. POSNER, supra note 56, at 298.
218. Id. at 298-99. In their review of Judge Posner's book, Michael Sullivan and Daniel Solove point out that none of these examples involves a government attempt to curtail liberty (much less suppress speech).
219. Michael Sullivan & Daniel J. Solove, Can Pragmatism Be Radical? Richard Posner and Legal Pragmatism, 113 YALE L.J. 687, 710-13 (2003)
220. (reviewing POSNER, supra note 56). Indeed, it is unclear how a curb on liberty might have helped avert any of the events Posner discusses.
221. This goes for situations unrelated to speech as well. There is ample indication that the American military drastically inflated the threat of Japanese sabotage on the West Coast during World War II (which led eventually to the internment of Japanese citizens and the Korematsu decision, Korematsu v. United States, 323 U.S. 214 (1944)) because of racial animus rather than any actual evidence.
222. See, e.g., Mark Tushnet, Defending Korematsu?: Reflections on Civil Liberties in Wartime, 2003 WIS. L. REV. 273, 288 (2003).
223. Likewise, the government's reaction to the post-war Red Scare was radically overinclusive and overwrought. See, e.g., id. at 289 ("[T]he Department of Justice actually had very little evidence to support its claims that the aliens it sought to deport were dangerous radicals.").
224. See, e.g., Jonathan Masur, A Hard Look or a Blind Eye: Administrative Law and Military Deference, 56 HASTINGS L.J. 441, 445 (2005) (describing court deference to military decisions despite existing administrative-law principles).
225. See generally, e.g., Cass Sunstein, Administrative Law Goes to War, 118 HARV. L. REV. 2663 (2005) (describing an administrative-law framework for evaluating executive action).
226. See Korematsu, 323 U.S. at 245 (Jackson, J., dissenting) ("In the very nature of things, military decisions are not susceptible of intelligent judicial appraisal.");
227. N. Jersey Media Group, Inc. v. Ashcroft, 308 F.3d 198, 219 (3d Cir. 2002)
228. ("We are quite hesitant to conduct a judicial inquiry into the credibility of these security concerns, as national security is an area where courts have traditionally extended great deference to Executive expertise." (citing Zadvydas v. Davis, 533 U.S. 678, 696 (2001)));
229. Detroit Free Press v. Ashcroft, 303 F.3d 681, 707 (6th Cir. 2002) ("Inasmuch as these agents' declarations establish that certain information revealed during removal proceedings could impede the ongoing anti-terrorism investigation, we defer to their judgment. These agents are certainly in a better position to understand the contours of the investigation and the intelligence capabilities of terrorist organizations.");
230. cf. United States v. Curtiss-Wright Exp. Corp., 299 U.S. 304, 319 (1936) ("In this vast external realm [of foreign affairs and international relations], with its important, complicated, delicate and manifold problems, the President alone has the power to speak or listen as a representative of the nation.").
231. Courts do not drop this deferential posture lightly. Cf. Pell v. Procunier, 417 U.S. 817, 827 (1974) ("[I]n the absence of substantial evidence in the record to indicate that the officials have exaggerated their response to these considerations, courts should ordinarily defer to their expert judgment in such matters. Courts cannot, of course, abdicate their constitutional responsibility to delineate and protect fundamental liberties.").
232. And sufficiently low probability as to render its constitutional significance debatable.
233. The only possible exception is the possibility of a riot or other such domestic disturbance, about which civilian social scientists might have as much to say as any executive-branch policymaker. Still, the danger inherent to such a disturbance, which includes the issue of how many police or national guard units stand ready to quell it, is a topic about which the executive is likely to have far better information than even any outside expert.
234. See Near v. Minnesota ex rel. Olson, 283 U.S. 697, 716 (1931) (describing the permissibility of government prevention of the publication of the sailing dates of troop transports).
235. See United States v. Progressive, Inc., 467 F. Supp. 990, 993 (W.D. Wis. 1979) (considering a request for an injunction to prohibit the publication of plans for a thermonuclear weapon).
236. N. Jersey Media Group, Inc. v. Ashcroft, 308 F.3d 198, 217-20 (3d Cir. 2002) (regarding the closing of a special-interest INS detainee hearing);
237. Detroit Free Press v. Ashcroft, 303 F.3d 681, 705-06 (6th Cir. 2002) (same).
238. Brandenburg v. Ohio, 395 U.S. 444, 447 (1969) (per curiam).
239. POSNER, supra note 56, at 316. Judge Posner writes: But if lawyers are not equipped to formulate sound legal policy regarding international terrorism, who is? The President is, virtually by default. The relevant expertise, on which he can draw, is widely distributed both within and outside government, but often in a form that cannot easily be presented in a legal forum, and not only because of the need for secrecy.
240. Id.
241. See supra notes 144-51 and accompanying text.
242. It is reasonable to assume that decision-makers-and the general public-will be risk averse to some degree and will rationally favor restrictions on low-probability harms that do not quite pass the cost-benefit test. But this fact does not change this Section's analysis. Risk averseness will only shift the cost-benefit curve slighdy downwards; it will not change the shape of that curve or eliminate the long tail. A subset of low-probability harms that fail even riskaverse cost-benefit balancing and are susceptible to the cognitive errors described here will remain.
243. For background on the application of cognitive science to legal decision-making, see generally Christine Jolis et al,, A Behavioral Approach to Law and Economics, 50 STAN. L. REV. 1471 (1998) (describing "how law and economic analysis may be improved by increased attention to insights about actual human behavior").
244. See Chris Guthrie et al., Inside the Judicial Mind, 86 CORNELL L. REV. 777, 816-18 (2001) (discussing studies that show that judges perform only slighdy better than laypeople at avoiding typical cognitive errors);
245. Frederick Schauer, On the Supposed Jury-Dependence of Evidence Law, 155 U. PA. L. REV. 165, 185-93 (2006);
246. Paul Slovic et al., Facts and Fears: Understanding Perceived Risk, in THE PERCEPTION OF RISK 137, 152 (Paul Slovic ed., 2000);
247. Wistrich et al., supra note 145.
248. See Mark Seidenfeld, Cognitive Loafing, Social Conformity, and Judicial Review of Agency Rulemaking, 87 CORNELL L. REV. 486, 494-508 (2002).
249. But see Jeffrey J. Rachlinski, The Uncertain Psychological Case for Paternalism, 97 NW. U. L. REV. 1165, 1216 n.270 (2005) (listing studies that demonstrate that experts make many of the same cognitive errors as laypeople) .
250. In re Aguinda, 241 F.3d 194, 205 (2d Cir. 2001) ("As generalists, [judges] need to continue to acquire general information .... We cannot make intelligent fact decisions or evaluate the effect of our legal decisions on society unless we have some understanding of that society." (internal citations omitted)) ;
251. Eric A. Posner, Transfer Regulations and Cost-Effectiveness Analysis, 53 DUKE L.J. 1067, 1092 (2003) ("Judicial deference to agency action is justified by division of labor: agencies are specialists, judges are generalists.").
252. See N. Jersey Media Group, Inc. v. Ashcroft, 308 F.3d 198, 219 (3d Cir. 2002) ("[N]ational security is an area where courts have traditionally extended great deference to Executive expertise.").
253. These conclusions are naturally tentative and subject to revision as the social-science literature advances.
254. Viscusi & Aldy, supra note 46, at 18-23 (reviewing the literature and performing a meta-analysis of studies of wage-risk tradeoffs in the United States labor market).
255. See id. at 24-26 (reviewing past studies of price-risk tradeoffs in the United States housing and product markets).
256. See id. at 18-26, 63. Viscusi and Aldy's meta-study obtained a median value of life of $6.7 million with a standard deviation of $5.6 million, and half of the studies from the United States labor market arrived at values between $5 million and $12 million.
257. Id. at 18, 63.
258. See id. at 19-21, 25. The studies surveyed involved such events as typical workplace injuries, fatal automobile accidents, air pollution, cigarette smoking, and bicycle accidents.
259. Id. Most risks fell in the vicinity of 1 in 10,000.
260. Id. Viscusi and Aldy's meta-analysis also revealed an inverse correlation between the value of a statistical life and the risk of death.
261. Id. at 23.
262. Sunstein, supra note 133, at 131. This is despite the optimism that individuals typically display toward their own chances of danger.
263. See Christine Jolls, Behavioral Economics Analysis of Redistributive Rules, 51 VAND. L. REV. 1653, 1658-63 (1998).
264. Id. By contrast, individuals typically underestimate larger-probability risks on the order of 1 in 1,000. W. KIP VISCUSI, FATAL TRADEOFFS: PUBLIC AND PRIVATE RESPONSIBILITIES FOR RISK 150 (1992).
265. Viscusi & Zeckhauser, supra note 24, at 116.
266. Id.
267. V. Kerry Smith & William H. Desvousges, An Empirical Analysis of the Economic Value of Risk Changes, 95 J. POL. ECON. 89, 100 tbl.2 (1987). Respondents to this study exhibited a willingness to pay more to reduce a small risk by 90% than to reduce a larger risk by the same proportion-meaning that they were paying extra for a smaller reduction in absolute risk-even when the two choices were presented sequentially to the same study participants.
268. Id.
269. See supra notes 175-83 and accompanying text.
270. See Viscusi & Zeckhauser, supra note 24, at 116.
271. Howard Kunreuther et al., Making Low Probabilities Useful, 23 J. RISK & UNCERTAINTY 103, 106 (2001).
272. Id. Some surveys phrased the information in terms of insurance premiums, offering the surveytaker an estimate of the premium that the insurance company would charge to insure against the harm (between fifteen cents and fifteen dollars). Since the survey obtained essentially identical results regardless of whether it was posed in terms of premiums or probabilities, I have described only the probability context here.
273. Id. at 107.
274. Id. The survey asked respondents to answer each of the four questions on a scale of 1 to 5, with 1 the "lowest risk" answer and 5 the "highest risk" answer. Respondents who were told that the chances of an accident were 1 in 100,000 rated the risk at 3.03; those who were told that the probability was 1 in 1 million rated the risk at 2.93; and those who were given a risk assessment of 1 in 10 million rated the risk at 3.01. The differences were not statistically significant.
275. Sunstein, supra note 135, at 71. Indeed, studies have shown a correlation between the probability of a harm and the likelihood that an individual will underestimate the threat posed by that harm.
276. See BARUCH FISCHHOFF ET AL., ACCEPTABLE RISK 29 (1981).
277. This is known as a "within subjects" design. The fact that survey respondents were unable to draw meaningful distinctions between different risk cases, even when they were presented simultaneously, renders these results all the more robust-not to mention irrational and counterintuitive.
278. M.W. Jones-Lee et al., The Value of Safety: Results of a National Sample Survey, 95 ECON. J. 49, 65-66 (1985).
279. M.W. Jones-Lee et al., Valuing the Prevention of Non-Fatal Road Injuries: Contingent Valuation vs. Standard Gambles, 47 OXFORD ECON. PAPERS 676, 688 (1995).
280. Smith & Desvousges, supra note 183, at 99-100.
281. Young Sook Eom, Pesticide Residue Risk and Food Safety Valuation: A Random Utility Approach, 76 AM. J. AGRIC. ECON. 760, 769 (1994).
282. C.-T. Jordan Lin & J. Walter Milon, Contingent Valuation of Health Risk Reductions for Shellfish Products, in VALUING FOOD SAFETY AND NUTRITION 83-114 (J.A. Caswell ed., 1995).
283. M. W. Jones-Lee et al., supra note 192, at 65-66.
284. See Sunstein, supra note 135, at 73-74 ("For most of us, most of the time, the relevant differences-between, say, 1/100,000 and 1/1,000,000-are not pertinent to our decisions, and by experience we are not well equipped to take those differences into account.").
285. See generally Viscusi & Aldy, supra note 46. These estimations have become particularly common within the United States government, where executive orders and regulations from the Office of Management and Budget mandate cost-benefit pricing of most proposed regulations and analysts employ risk-risk analysis (a weighing of the risks involved in promulgating the regulation against the risks involved in not promulgating the regulation) in the remainder.
286. See id. at 5-6, 53-54.
287. This calculation was based on approximately 2,700 deaths at a cost of $6.7 million per death and an order of magnitude estimate of property damage at $10 billion.
288. See Moisan v. Loftus, 178 F.2d 148, 149 (2d Cir. 1949). Judge Hand wrote:
289. The injuries are always a variable within limits, which do not admit of even approximate ascertainment; and, although probability might theoretically be estimated, if any statistics were available, they never are; and, besides, probability varies with the severity of the injuries. It follows that all such attempts are illusory; and, if serviceable at all, are so only to center attention upon which one of the factors may be determinative in any given situation.
290. Id.
291. See, e.g., Ctr. for Nat'l Sec. Studies v. U.S. Dep't of Justice, 331 F.3d 918, 923 (D.C Cir. 2003) ("[T]he declarations state that release of the requested information could hamper the ongoing investigation by leading to the identification of detainees by terrorist groups, resulting in terrorists either intimidating or cutting off communication with the detainees . . . ." (emphasis added));
292. United States v. Progressive, Inc., 467 F. Supp. 990, 993-94 (W.D. Wis. 1979) ("[T]he article could possibly provide sufficient information to allow a medium size nation to move faster in developing a hydrogen weapon. It could provide a ticket to by-pass blind alleys. The Morland piece could accelerate the membership of a candidate nation in the thermonuclear club." (emphasis added)).
293. Here, I bracket all epistemological questions related to the "actual" probability of some event that depends on any number of path-independent factors. For purposes of this analysis, I simply assume that any given future event has some fixed probability of occurring given the presence or absence of antecedent events, including the speech act in question.
294. See, e.g., Eric A. Posner, Law and the Emotions, 89 GEO. L.J. 1977, 1979-84 (2001) (collecting and describing psychological studies demonstrating the influence of emotions on decision-making and preference formation).
295. See Sunstein, supra note 133, at 125. Sunstein's subjects were willing to pay, on average, $188.33 to avoid the emotionally charged risk and only $129.61 to dodge the unemotional version.
296. Id.
297. Yuval Rottenstreich & Christopher K. Hsee, Money, Kisses, and Electric Shocks: On the Affective Psychology of Risk, 12 PSYCHOL. SCI. 185, 185-90 (2001).
298. See, e.g., Slovic et al., supra note 170, at 143, 148 (describing the influence that the dread, severity, and "catastrophic potential" of a harm have upon conceptions of that harm's likelihood).
299. CHARLES PERROW, NORMAL ACCIDENTS 324-28 (1999);
300. see also Clark R. Chapman & David Morrison, Impacts on the Earth by Asteroids and Comets: Assessing the Hazard, 367 NATURE 38 (1994);
301. Slovic et al., supra note 170, at 143;
302. Sunstein, supra note 133, at 121.
303. Sunstein, supra note 135, at 100.
304. Slovic et al., supra note 170, at 148.
305. CASS R. SUNSTEIN, RISK AND REASON 45 (2002);
306. Sunstein, supra note 133, at 128;
307. Viscusi Sc Zeckhauser, supra note 24, at 99.
308. See Paul Slovic, Perception of Risk, 236 SCI. 280, 281-83 (1987);
309. cf. John Rather, Dreading a Replay of the 1938 Hurricane, N.Y. TIMES, Aug. 28, 2005, at L11 (citing a study that found that before Hurricane Katrina, sixty-four percent of homes in the United States were underinsured against floods and hurricanes).
310. See, e.g., Posner & Vermeule, supra note 140, at 626-34.
311. Data strongly suggest that individuals tend to underestimate the risk they face from everyday dangers such as strokes (approximately 1 in 2,000) and automobile accidents (1 in 5,000). VISCUSI, supra note 180, at 150.
312. These effects are undoubtedly exacerbated by the more general phenomenon that, as one commentator put it, "the costs of freedom of expression are often more salient than the benefits." Posner, supra note 22, at 744. Posner attributes this imbalance to the fact that speech that challenges common assumptions can be painful in the immediate term but is "an unavoidable concomitant of social progress," and so "the cost of heterodox speech is immediate and its benefit deferred."
313. Id.
314. Sunstein, supra note 135, at 82.
315. W. Kip Viscusi, Alarmist Decisions with Divergent Risk Information, 107 ECON. J. 1657, 1657-59 (1997).
316. Id. at 1666;
317. see also Sunstein, supra note 135, at 82-83.
318. Sunstein, supra note 135, at 120-21. Sunstein views this phenomenon as yet another manifestation of low-probability insensitivity.
319. Id.
320. See supra Part IV.A.
321. See supra Part IV.A.
322. Sunstein, supra note 133, at 121;
323. see also STONE, supra note 22, at 537-38.
324. SUNSTEIN, supra note 211, at 34;
325. Amos Tversky & Daniel Kahneman, Availability: A Heuristic for Judging Frequency and Probability, in JUDGMENT UNDER UNCERTAINTY: HEURISTICS AND BIASES 163, 163-78 (Daniel Kahneman et al. eds., 1982);
326. Roger G. Noll & James E. Krier, Some Implications of Cognitive Psychology for Risk Regulation, 19 J. LEGAL STUD. 747, 755 (1990).
327. Cf. Posner & Vermeule, supra note 140, at 622-25 (arguing that publicly salient examples of governmental over-reaction can catalyze expansions in civil liberties).
328. See PERROW, supra note 208, at 325-27.
329. See id.;
330. Slovic et al., supra note 170, at 142. These risks are in contradistinction to such daily terrors as car accidents or workplace injuries.
331. See generally, e.g., Brandenburg v. Ohio, 395 U.S. 444 (1969) (per curiam) (concerning the unrealized threat of race riot);
332. Watts v. United States, 394 U.S. 705 (1969) (per curiam) (concerning the unrealized threat of urban riot).
333. See STONE, supra note 22, at 311-426.
334. SUNSTEIN, supra note 211, at 50-52.
335. By way of illustration, a Lexis search of the Wall Street Journal for the words "terrorism" or "terrorist" produced 135 hits within the month of March 2006 alone. A similar search of the Atlanta Journal-Constitution produced 102 hits. The terrorist threat obviously retains preeminent salience within the public consciousness.
336. See generally, e.g., Ctr. for Nat'l Sec. Studies v. U.S. Dep't of Justice, 331 F.3d 918 (D.C. Cir. 2003);
337. N. Jersey Media Group, Inc. v. Ashcroft, 308 F.3d 198 (3d Cir. 2002);
338. Detroit Free Press v. Ashcroft, 303 F.3d 681 (6th Cir. 2002).
339. Even the D.C. Circuit's recent decision denying the existence of a constitutional reporter-source privilege rooted in the First Amendment sounded in terrorism. In re Grand Jury Subpoena, Judith Miller, 397 F.3d 964, 965-67 (D.C. Cir. 2005).
340. Whether a reporter-source privilege should exist is very much a question of the costs and benefits involved in providing reporters with a shield against forced disclosure of information. McKevitt v. Pallasch, 339 F.3d 530, 531 (7th Cir. 2003).
341. The Judith Miller case appears against a national-security backdrop; reporters were attempting to use the privilege in that case to block the investigation of the leaking of a CIA operative's name in conjunction with a public debate over the threat of terrorism that existed before the war in Iraq. In re Grand Jury Subpoena, 397 F.3d at 965-68.
342. The probability threshold is of course not meant as a full substitute for cost-benefit balancing, as it is certainly not the case that any likely harm-no matter how small in magnitude - provides an adequate justification for banning speech. Rather, an asserted threat suffices as a legitimate ground for suppressing speech only if it passes both the cost-benefit test and the probability threshold.
343. See, e.g., Chris Guthrie et al., supra note 170, at 822-26;
344. cf. Aaron J. Wright, Rendered Impractical: Behavioral Economics and the Impracticability Doctrine, 26 CARDOZO L. REV. 2183, 2212-13 (2005).
345. See Gregory N. Mandel, Patently Non-Obvious: Empirical Demonstration that the Hindsight Bias Renders Patent Decisions Irrational, 67 OHIO ST. L. J. 1391, 1316-18 (2006).
346. See, e.g., Korematsu v. United States, 323 U.S. 214, 218 (1944) ("[W]e cannot reject as unfounded the judgment of the military authorities and of Congress.");
347. Hamdi v. Rumsfeld, 316 F.3d 450, 473 (4th Cir. 2003) ("To delve further into Hamdi's status and capture would require us to step so far out of our role as judges that we would abandon the distinctive deference that animates this area of law." (emphasis added));
348. N. Jersey Media, 308 F.3d at 219 ("To the extent that the Attorney General's national security concerns seem credible, we will not lightly second-guess them.");
349. Detroit Free Press, 303 F.3d at 707 ("[W]e defer to their judgment. These agents are certainly in a better position to understand the contours of the investigation ....");
350. Padilla ex rel. Newman v. Bush, 233 F. Supp. 2d 564, 608 (S.D.N.Y. 2002)
351. ("[DJeference is due because of a principle captured in another 'statement of Justice Jackson - that we decide difficult cases presented to us by virtue of our commissions, not our competence.'" (quoting Dames & Moore v. Regan, 453 U.S. 654, 661 (1981))).
352. United States v. Progressive, 467 F. Supp. 990 (W.D. Wis. 1979).
353. Ian M. Dumain, No Secret, No Defense: United States v. Progressive, 26 CARDOZO L. REV. 1323, 1325 (2005) (explaining that the goal of the Article was to "puncture the secrecy mystique that surrounded nuclear weapons" (internal quotation marks and citation omitted)).
354. Progressive, 467 F. Supp. at 993;
355. see also id. at 995 ("[N]ot all the data is available in the public realm in the same fashion, if it is available at all." (emphasis added)).
356. Id. at 993-94 (emphasis added).
357. Id. at 995.
358. Id. at 996.
359. HOWARD MORLAND, THE SECRET THAT EXPLODED 35-37 (1981).
360. Id. at 23.
361. Id. at 24.
362. Nor could it have been of much value that Morland had collected all of the relevant sources. A thorough literature search requires negligible effort in comparison to the greater project of producing a nuclear weapon. See Progressive, 467 F. Supp. at 993 ("A number of affidavits make quite clear that a sine qua non to thermonuclear capability is a large, sophisticated industrial capability coupled with a coterie of imaginative, resourceful scientists and technicians. One does not build a hydrogen bomb in the basement.").
363. A. DEVOLPI ET AL., BORN SECRET: THE H-BOMB, THE PROGRESSIVE CASE, AND NATIONAL SECURITY 84 (1981).
364. Id.
365. Howard Morland, The H-bomb Secret, PROGRESSIVE, Nov. 1979, at 9-10.
366. Id.
367. Howard Morland, Errata, PROGRESSIVE, Nov. 1979, at 35.
368. See Dumain, supra note 236, at 1335 ("Perhaps surprisingly, all parties to The Progressive case agreed that there were significant inaccuracies in Morland's article."). In hindsight, Morland's errors are almost laughably dramatic. On the second page of his article, he explained breathlessly that "[r]adiation pressure - a term never mentioned in the open literature - is the essence of what remains of the H-bomb secret."
369. Morland, supra note 247. Perhaps the reason that "radiation pressure" had never made it into the open literature was that it was completely irrelevant to the construction of a fusion weapon.
370. DEVOLPI ET AL., supra note 245, at 91.
371. Id. at 91-92.
372. Id. at 108-09.
373. Id. at 130.
374. Dumain, supra note 236, at 1331-32.
375. Id.
376. Morland, supra note 249, at 34.
377. N. Jersey Media Group, Inc. v. Ashcroft, 308 F.3d 198 (3d Cir. 2002).
378. Detroit Free Press v. Ashcroft, 303 F.3d 681 (6th Cir. 2002).
379. The plaintiffs in each case brought suit under Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555 (1980), which held that "the right to attend criminal trials is implicit in the guarantees of the First Amendment; without the freedom to attend such trials, which people have exercised for centuries, important aspects of freedom of speech and 'of the press could be eviscerated.'"
380. Id. at 580
381. (citing Branzburg v. Hayes, 408 U.S. 665, 681 (1972)). In both cases, the trial courts extended such rights to INS deportation hearings (which have many of the qualities of criminal trials) and held that the government must demonstrate a compelling interest if it wished to close the hearings. Detroit Free Press v. Ashcroft, 195 F. Supp. 2d 937, 943-46 (E.D. Mich. 2002) (indicating that while the plaintiffs had a right of access in this case, the "press' and public's right is not absolute"),
382. aff'd, 303 F.3d 651 (6th Cir. 2002);
383. N. Jersey Media Group, Inc. v. Ashcroft, 205 F. Supp. 2d 288, 301 (D.N.J. 2002), rev'd, 308 F.3d 198 (3d Cir. 2002)).
384. N. Jersey Media, 308 F.3d at 199.
385. Id. at 218.
386. Id. at 219 ("The Newspapers are undoubtedly correct that the representations of the Watson Declaration are to some degree speculative, at least insofar as there is no concrete evidence that closed deportation hearings have prevented, or will prevent, terrorist attacks.").
387. N. Jersey Media, 205 F. Supp. 2d at 301 (emphasis added) ;
388. see also Detroit Free Press, 195 F. Supp. 2d at 947. The Michigan district court found that: Each of the interests the government relies upon addresses the dangers associated with disclosing the name of the detainee, as well as the date and place of the detainee's arrest. With respect to the present matter, however, that information was made public from the outset. . . . Furthermore, neither in the Creppy directive nor elsewhere does the Government prohibit detainees in special interest cases (or their counsel or families) from revealing that information to the press and public.
389. Id.
390. Detroit Free Press, 303 F.3d at 707 ("These agents are certainly in a better position to understand the contours of the investigation and the intelligence capabilities of terrorist organizations.");
391. see also N.Jersey Media, 308 F.3d at 219.
392. 5 U.S.C.A. § 552 (West 2005).
393. Ctr. for Nat'l Sec. Studies v. U.S. Dep't of Justice, 331 F.3d 918, 920 (D.C. Cir. 2003).
394. Id.
395. 5 U.S.CA. § 522 (b) (7) (A), (C), & (F) (West 2005) (emphasis added).
396. Ctr. for Nat'l Sec. Studies, 331 F.3d at 937.
397. Id. at 923.
398. Id. at 922.
399. Id. at 939 (Tatel, J., dissenting).
400. See, e.g., Roselink Investors, L.L.C. v. Shenkman, 386 F. Supp. 2d 209, 224 (S.D.N.Y. 2004) ("[T]he business judgment rule is intended to protect directors against just such attacks because their decisions are not to be second-guessed by courts with the benefit of hindsight.");
401. William J. Stuntz, Warrants and Fourth Amendment Remedies, 77 VA. L. REV. 881, 911 (1991) ("In almost any after-the-fact system of adjudication, decisionmakers must determine whether behavior was appropriate after they know how the behavior turned out.").
402. United States v. Virginia, 518 U.S. 515, 542-43 (1996) (citing Miss. Univ. for Women v. Hogan, 458 U.S. 718, 730 (1982)). The implication is that, regardless of the potential level of harm (and in Virginia, the government's witnesses had predicted nothing less than the destruction of the very character and importance of the Virginia Military Institute), the harm must reach a certain (high) level of probability for it even to become judicially cognizable.