Free to leave? An empirical look at the fourth Amendment's seizure Standard

Journal of Criminal Law and Criminology

Volumn 99, Issue 1, 2009, Pages 51-88

  Kessler, David K ^a  

^a HARVARD LAW SCHOOL   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 77951071072     PISSN: 00914169     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (192)

1. Transcript of Oral Argument at 43, Brendlin v. California, 127 S. Ct. 2400 (2007) (No. 06-8120) (Breyer, J.).
2. Id. at 44 (Scalia, J.).
3. 127 S. Ct. 2400 (2007).
4. United States v. Mendenhall, 446 U.S. 544, 554 (1980) (plurality opinion). The actual language in Mendenhall is whether a reasonable person "would have believed he was not free to leave," id., but later references to the standard have used the wording "felt free to leave." See, e.g., Florida v. Bostick, 501 U.S. 429, 435-36 (1991); California v. Hodari D., 499 U.S. 621, 639-40 (1991) (Stevens, J., dissenting); Transcript of Oral Argument at 43, Brendlin, 127 S. Ct. 2400 (No. 06-8120) (Breyer, J.). In addition, the Court seems never to have drawn a distinction between "believed he was not free" and "did not feel free."
5. United States v. Drayton, 536 U.S. 194, 202 (2002).
6. Id. This Article collectively refers to these tests as asking whether a person felt "free to leave."
7. See, e.g., Muehler v. Mena, 544 U.S. 93, 101 (2005) ("We have held repeatedly that mere police questioning does not constitute a seizure." (internal quotation marks omitted)).
8. As discussed below, the reasonable person described by the Court should act and feel roughly the way the average person described by the results of the survey should. See infra text accompanying notes 161-68.
9. Janice Nadler has noted that "[t]he simplest way to determine whether a reasonable person voluntarily consented to a police search is simply to ask them, 'To what extent did you feel free to decline the officer's request?'" Janice Nadler, 8o 8eed to Shout: Bus Sweeps and the Psychology of Coercion, 2002 SUP. CT. REV. 153, 201. But she concluded "that no one has asked this question." Id.; see Daniel J. Steinbock, The Wrong Line Between Freedom and Restraint: The Unreality, Obscurity, and Incivility of the Fourth Amendment Consensual Encounter Doctrine, 38 SAN DIEGO L. REV. 507, 552 (2001) ("The author knows of no scientific study on this issue directly, but there is no reason one could not be undertaken."). Burkoff does cite an unpublished undergraduate thesis that surveyed Florida State University students and law enforcement officers about whether they believed a person would feel free to leave once he had refused an officer's request to search his car. John M. Burkoff, Search Me?, 39 TEX. TECH L. REV. 1109, 1119-20 (2007). But that survey focused on the voluntariness of consent searches, a question related to, but not identical to, the subject of this Article.
10. See Muehler v. Mena, 544 U.S. 93, 101 (2005) ("We have held repeatedly that mere police questioning does not constitute a seizure." (internal quotation marks omitted)).
11. See, e.g., Arnold H. Loewy, Cops, Cars, and Citizens: Fixing the Broken Balance, 76 ST. JOHN'S L. REV. 535, 555 (2002) ("[M]ost people who have been stopped understand that they are not free to leave until the police officer tells them so.");
12. Tracey Maclin, "Black and Blue Encounters" - Some Preliminary Thoughts About Fourth Amendment Seizures: Should Race Matter?, 26 VAL. U. L. REV. 243, 249 (1991) [hereinafter Maclin, "Black and Blue Encounters"] ("The average, reasonable individual-whether he or she be found on the street, in an airport lobby, inside a factory, or seated on a bus or train-will not feel free to walk away from a typical police confrontation.");
13. Tracey Maclin, The Good and Bad 8ews About Consent Searches in the Supreme Court, 39 MCGEORGE L. REV. 27, 28 (2008) [hereinafter Maclin, The Good and Bad 8ews] ("[E]veryone else knows [that] a police 'request' to search a bag or automobile is understood by most persons as a 'command.'");
14. Stephen A. Saltzburg, The Supreme Court, Criminal Procedure and Judicial Integrity, 40 AM. CRIM. L. REV. 133, 137 (2003) ("Would any reasonable worker feel free to leave under these circumstances? In the world in which most people live, the answer is no."); see also RONALD JAY ALLEN ET AL., CRIMINAL PROCEDURE: INVESTIGATION AND RIGHT TO COUNSEL 404 (2005) ("[D]oes the average person when approached by a police officer feel free to terminate the encounter . . . ? Isn't the seizure test in fact a legal fiction . . . ?").
15. See infra text accompanying notes 88-110.
16. Steinbock, supra note 9, at 535-36 (discussing other troubling consequences of the Court's jurisprudence).
17. A seizure is unconstitutional if it occurs without a warrant or probable cause, or, in some cases, without reasonable suspicion. 1 JOSHUA DRESSLER & ALAN C. MICHAELS, UNDERSTANDING CRIMINAL PROCEDURE §§ 7-9 (2006) (reviewing the kinds of justifications that make a seizure "reasonable," that is, constitutionally permissible).
18. Evidence gathered during such a seizure would be inadmissible in court. E.g., Mapp v. Ohio, 367 U.S. 643 (1961) (extending the exclusionary rule to state trials); Weeks v. United States, 232 U.S. 383 (1914) (establishing the federal exclusionary rule).
19. Terry v. Ohio, 392 U.S. 1, 19 n.16 (1968) ("Obviously, not all personal intercourse between policemen and citizens involves 'seizures' of persons.").
20. Id. ("Only when the officer, by means of physical force or show of authority, has in some way restrained the liberty of a citizen may we conclude that a 'seizure' has occurred."); see Brendlin v. California, 127 S. Ct. 2400, 2405 (2007) ("A police officer may make a seizure by a show of authority and without the use of physical force, but there is no seizure without actual submission; otherwise, there is at most an attempted seizure, so far as the Fourth Amendment is concerned.").
21. These seizures involved handcuffs or other ways of physically restraining a suspect. See, e.g., California v. Hodari D., 499 U.S. 621, 629 (1991) (holding that defendant was seized when "he was tackled" by a police officer).
22. United States v. Mendenhall, 446 U.S. 544, 554 (1980) (plurality opinion).
23. Hodari D., 499 U.S. at 628. Because the test is objective, "neither the subjective intentions of the particular officer nor the subjective expectations of the particular suspect are determinative. Wayne R. LaFave, Pinguitudinous Police, Pachydermatous Prey: Whence Fourth Amendment "Seizures"?, 1991 U. ILL. L. REV. 729, 738.
24. Mendenhall, 446 U.S. at 554.
25. Mendenhall, 446 U.S.. at 555.
26. Mendenhall, 446 U.S.. at 555.
27. Mendenhall, 446 U.S.. at 555.
28. Mendenhall, 446 U.S.. at 555.
29. Brendlin v. California, 127 S. Ct. 2400, 2405 (2007); Ohio v. Robinette, 519 U.S. 33, 46 (1996); California v. Hodari D., 499 U.S. 621, 627 (1991); Brower v. County of Inyo, 489 U.S. 593, 600-01 (1989) (Stevens, J., concurring in the judgment); Michigan v. Chesternut, 486 U.S. 567, 573 (1988) ("The Court has . . . embraced this test"); INS v. Delgado, 466 U.S. 210, 214 (1984); Florida v. Royer, 460 U.S. 491, 516 n.1 (1983). The test has been widely used in the federal circuit courts as well. See, e.g., Belcher v. Norton, 497 F.3d 742, 747-48 (7th Cir. 2007) ("In order to establish that Deputy Marshal Norton's actions
30. Royer, 460 U.S. at 502.
31. Royer, 460 U.S. at 501.
32. Royer, 460 U.S at 502 ("These circumstances surely amount to a show of official authority such that 'a reasonable person would have believed he was not free to leave.'" (quoting Mendenhall, 446 U.S. at 554)).
33. Delgado, 466 U.S. at 218.
34. Delgado, 466 U.S. at 218.
35. Delgado, 466 U.S. at 218.
36. Michigan v. Chesternut, 486 U.S. 567, 575 (1988).
37. Michigan v. Chesternut, 486 U.S. 567, 575 (1988). at 576.
38. Michigan v. Chesternut, 486 U.S. 567, 575 (1988) at 574.
39. Michigan v. Chesternut, 486 U.S. 567, 575 (1988)
40. Florida v. Bostick, 501 U.S. 429, 436 (1991).
41. Florida v. Bostick, 501 U.S. 429, 436 (1991). at 431-32.
42. Florida v. Bostick, 501 U.S. 429, 436 (1991) at 436. The Court also made clear that the reasonable person it had in mind was an "innocent person," not a reasonable criminal. Id. at 438.
43. Florida v. Bostick, 501 U.S. 429, 436 (1991). at 434, 437.
44. United States v. Drayton, 536 U.S. 194, 202 (2002) ("The proper inquiry is whether a reasonable person would feel free to decline the officers' requests or otherwise terminate the encounter." (internal quotation marks omitted)).
45. Brendlin v. California, 127 S. Ct. 2400, 2405-06 (2007) ("[W]hen a person has no desire to leave for reasons unrelated to the police presence, the coercive effect of the encounter can be measured better by asking whether a reasonable person would feel free to decline the officers' requests or otherwise terminate the encounter . . . ." (internal quotation marks omitted)).
46. Drayton, 536 U.S. at 203-04.
47. Drayton, 536 U.S. at 204.
48. Brendlin, 127 S. Ct. at 2410 ("Brendlin[, the passenger,] was seized from the moment Simeroth's[, the driver,] car came to a halt on the side of the road . . . .").
49. Brendlin, 127 S. Ct at 2407.
50. Drayton, 536 U.S. at 201.
51. Florida v. Royer, 460 U.S. 491, 497 (1983); see, e.g., Florida v. Rodriguez, 469 U.S. 1, 5-6 (1984) (per curiam) (holding that such interactions in airports are "the sort of consensual encounter[s] that implicat[e] no Fourth Amendment interest").
52. See, e.g., Muehler v. Mena, 544 U.S. 93, 101 (2005) ("We have held repeatedly that mere police questioning does not constitute a seizure." (internal quotation marks omitted)); Terry v. Ohio, 392 U.S. 1, 34 (1968) (White, J., concurring) ("[T]here is nothing in the Constitution which prevents a policeman from addressing questions to anyone on the streets.").
53. Royer, 460 U.S. at 498.
54. Drayton, 536 U.S. at 210.
55. INS v. Delgado, 466 U.S. 210, 218 (1984).
56. Michigan v. Chesternut, 486 U.S. 567, 575 (1988).
57. Cf. Marcy Strauss, Reconstructing Consent, 92 J. CRIM L. & CRIMINOLOGY 211, 212 (2002) ("Only if the police behave with some extreme degree of coercion beyond that inherent in the police-citizen confrontation will a court vitiate the consent [to search].").
58. United States v. Mendenhall, 446 U.S. 544, 544 (1980).
59. See Brendlin v. California, 127 S. Ct. 2400, 2407 (2007).
60. Nadler, supra note 9, at 166-67 (2002) ("[T]he Court assumed these questions can be answered from intuition alone."); see also Ric Simmons, 8ot "Voluntary" but Still Reasonable: A 8ew Paradigm for Understanding the Consent Searches Doctrine, 80 IND. L.J. 773, 800-01 (2005) ("The most frequent criticism of the consent search cases is that the Supreme Court is unaware of the realities on the street . . . .").
61. See, e.g., Missouri v. Seibert, 542 U.S. 600, 615-16 (2004) ("[S]ince a reasonable person in the suspect's shoes could have seen the station house questioning as a new and distinct experience, the Miranda warnings could have made sense as presenting a genuine choice whether to follow up on the earlier admission.").
62. See, e.g., Nadler, supra note 9, at 167 ("[T]hese are questions that depend crucially on
63. empirical inquiries . . . . [R]elying on casual intuition to infer why someone acted the way
64. they did in a situation where all of the details and circumstances are important and must be taken into account (as the Court has emphasized repeatedly) almost always leads to mistaken and erroneous judgments."); see also Michael C. Dorf, Foreword: The Limits of Socratic Deliberation, 112 HARV. L. REV. 4, 56 (1998) ("To begin, the Court could rely to a greater extent on empirical and policy analysis in its written opinions.").
65. See, e.g., Transcript of Oral Argument at 43-44, Brendlin, 127 S. Ct. 2400 (No. 06-8120) (noting the lack of "studies" concerning whether people feel free to leave during encounters with police).
66. See Burkoff, supra note 9, at 1119-20 (discussing an undergraduate thesis that asked whether people would feel free to leave once they had declined a police request to search a car); Nadler, supra note 9, at 201 (noting the absence of studies on this question); cf. Daniel L. Rotenberg, An Essay on Consent(less) Police Searches, 69 WASH. U. L.Q. 175, 188 (1991) ("Empirical studies [on consent searches] are either nonexistent or inconclusive."). 62 Burkoff also refers to "anecdotal" evidence from an unpublished undergraduate thesis. Burkoff, supra note 9, at 1119-20.
67. Illya Lichtenberg, Miranda in Ohio: The Effects of Robinette on the "Voluntary" Waiver of Fourth Amendment Rights, 44 HOW. L.J. 349, 366 (2001).
68. Id. at 367. The study found that the rate of consent actually went up slightly once the police began using the warning. Id. But there is no discussion in Lichtenberg's article about whether this difference is statistically significant, and the difference may have been caused by an overly sensitive dependent variable.
69. Steven L. Chanenson, Get the Facts, Jack! Empirical Research and the Changing Constitutional Landscape of Consent Searches, 71 TENN. L. REV. 399, 452 (2004) (discussing data from Lichtenberg's study available in Lichtenberg's doctoral dissertation, but not published in the final article).
70. Simmons, supra note 57, at 774 ("[N]o outsider viewing the interaction would conclude that the defendants voluntarily consented to a search when surrounded by police at close quarters, especially if the defendants knew (as they must have) that giving the consent would ultimately result in serious criminal charges being filed against them.").
71. For a thorough review of this literature, see Nadler, supra note 9. Cf. Simmons, supra note 57, at 801 ("[A] few legal scholars have applied various psychological experiments to the question of consent searches, and without exception they have concluded that the studies provide evidence that most of the 'consents' approved of by the Supreme Court are in fact involuntary.").
72. Nadler, supra note 9, at 173.
73. The Court itself has recognized the power of psychological pressure to compel people to act in ways they would not otherwise act. For example, the Court recognized that students who listened to the reading of a prayer at a graduation ceremony might feel compelled to accept the religious messages even if they did not want to do so. See Lee v. Weisman, 505 U.S. 577, 588 (1992). Justice Kennedy wrote that the role of authority-the school district-in establishing the prayer "places public pressure, as well as peer pressure, on attending students to stand as a group or, at least, maintain respectful silence . . . . This pressure, though subtle and indirect, can be as real as any overt compulsion." Id. at 594.
74. See STANLEY MILGRAM, OBEDIENCE TO AUTHORITY: AN EXPERIMENTAL VIEW (1974); see also Chanenson, supra note 65, at 448-50 (2004); Nadler, supra note 9, at 175-77 (discussing MILGRAM, supra); Simmons, supra note 57, at 802-08 (discussing MILGRAM, supra).
75. Simmons, supra note 57, at 802-08.
76. Nadler, supra note 9, at 176.
77. Id. at 175-76.
78. Martha Minow, Living Up To Rule Holding Soldiers Responsible for Abusive Conduct and the Dilemma of the Superior Orders Defence, 52 MCGILL L.J. 1, 32 (2007).
79. Leonard Bickman, The Social Power of a Uniform, 4 J. APPLIED SOC. PSYCHOL. 47 (1974).
80. Simmons, supra note 57, at 808.
81. Simmons, supra note 57, at 808.
82. Simmons, supra note 57, at 808.
83. Brad J. Bushman, Perceived Symbols of Authority and Their Influence on Compliance, 14 J. APPLIED SOC. PSYCHOL. 501, 502-06 (1984); Simmons, supra note 57, at 809 n.191 (discussing Bushman's experiment).
84. Chanenson, supra note 65, at 449 ("Yet, it is not reasonable to rely on the studies as full and direct support for attacking the voluntariness of all consent searches."); cf. Simmons, supra note 57, at 807 ("In summary, Milgram's experiments are not very useful in helping us assign the appropriate level of police pressure in the context of consent searches. Many of those that obeyed protested vigorously; all did so knowing full well the results of their actions.").
85. Minow, supra note 74, at 32; Moti Nissani, Comment, A Cognitive Reinterpretation of Stanley Milgram's Observations on Obedience to Authority, 45 AM. PSYCHOL. 1384, 1384 (1990).
86. See, e.g., David Luban, Alan Strudler & David Wasserman, Moral Responsibility in the Age of Bureaucracy, 90 MICH. L. REV. 2348, 2362 (1992) ("Milgram discovered that compliance was extraordinarily sensitive to peer pressure. When the other team members refused to proceed with the experiment, only ten percent of the subjects remained obedient to the experimenter and 'went all the way.' Conversely, when a teammate rather than the
87. Nadler, supra note 9, at 193 ("In sum, people approached at a close distance by an authority in a tightly enclosed space with no opportunity to move further away or leave feel discomfort and tension; at the same time, people who find their space invaded in this manner are more willing to comply with the request of the person making them feel uncomfortable.").
88. Id. at 194.
89. See Lichtenberg, supra note 63.
90. Some of the people who consented to the search gave answers such as: "I knew legally I didn't have to, but I kind of felt that I had to" or "I felt a little pressured that I didn't have much choice." Nadler, supra note 9, at 202-03 (citing Lichtenberg, supra note 63).
91. See infra notes 91-97.
92. See infra notes 98-105.
93. See infra notes 106-10.
94. See, e.g., Simmons, supra note 57, at 782-83 (arguing that the two tests are really the same).
95. See, e.g., Wesley MacNeil Oliver, With an Evil Eye and an Unequal Hand: Pretextual Stops and Doctrinal Remedies to Racial Profiling, 74 TUL. L. REV. 1409, 1466 (2000) ("Consent to lawful authority therefore does not necessarily evince a willingness to have the personal property in one's car thoroughly inspected. Absent knowledge of the right to withhold consent, a suspect is left with the belief that he has no choice.").
96. See Alan C. Michaels, Rights Knowledge: Values and Tradeoffs, 39 TEX. TECH L. REV. 1355, 1379 (2007) ("To the extent one believes the right is designed to protect individual dignitary interests, greater knowledge of the right would provide some further protection to individual autonomy . . . .").
97. See, e.g., id.; Loewy, supra note 11, at 554 ("The Court's suggestion that such [a warning] would destroy the informality of the interchange and be thoroughly impractical is so palpably false as to be laughable. A simple 'I'd like you to let me search your car Joe, but you don't have to' would add to the informality, while minimizing the show of force."); Simmons, supra note 57, at 821 ("The Supreme Court has shown concern that a notification would be an 'unrealistic' burden to put on law enforcement and would be 'thoroughly impractical;' but if the notification was brief enough, these concerns seem unjustified . . . . Thus, there would be no harm and perhaps a bit of good in requiring a brief notification of a suspect's rights before a law enforcement officer can request consent."); Adrian J. Barrio, Note, Rethinking Schneckloth v. Bustamonte: Incorporating Obedience Theory into the Supreme Court's Conception of Voluntary Consent, 1997 U. ILL. L. REV. 215, 244-47 (arguing that the costs to law enforcement of a warning requirement would be small);
98. Rebecca A. Stack, Note, Airport Drug Searches: Giving Content to the Concept of Free and Voluntary Consent, 77 VA. L. REV. 183, 206 (1991) ("[A]ll that is required for an effective warning are the six small words, 'You have the right to refuse,' appended to each
99. Gerard E. Lynch, Why 8ot a Miranda for Searches?, 5 OHIO ST. J. CRIM. L. 233, 245 (2007) ("I believe it would be an equally good thing if similar words accompanied police requests for consent to search. . . . Such a rule, like its Miranda parallel in custodial interrogation, would probably reduce the number of consents obtained, but most likely not by much.").
100. See, e.g., Devon W. Carbado, (E)racing the Fourth Amendment, 100 MICH. L. REV. 946, 1030 (2002) (arguing that warnings about the right to refuse consent may help to counteract the greater coercion some races feel when interacting with the police);
101. Carol S. Steiker, How Much Justice Can You Afford? A Response to Stuntz, 67 GEO. WASH. L. REV. 1290, 1294 (1999) ("Requiring police officers, when seeking consent to search, to advise all suspects of their right to refuse to consent would help to close the information and power gap currently existing between the rich and poor in their encounters with law enforcement agents.").
102. See, e.g., Chanenson, supra note 65, at 465 (arguing that warnings alone are not enough).
103. For a discussion of potential political reasons why the Court has not adopted this position, see Lynch, supra note 94.
104. See Schneckloth v. Bustamonte, 412 U.S. 218, 227, 231 (1973) (arguing that "valid consent may be the only means of obtaining important and reliable evidence" and that "it would be thoroughly impractical to impose on the normal consent search the detailed requirements of an effective warning"); see also Loewy, supra note 11, at 554 (arguing that a warning requirement would "[o]f course . . . decrease the number of consent searches").
105. Schneckloth, 412 U.S. at 227. But see id. at 277 (Marshall, J., dissenting) ("I would have thought that the capacity to [consent to the police] necessarily depends upon knowledge that there is a choice to be made.").
106. Id. at 227 ("While knowledge of the right to refuse consent is one factor to be taken into account, the government need not establish such knowledge as the sine qua non of an effective consent.").
107. Ohio v. Robinette, 519 U.S. 33 (1996).
108. See id. at 39-40 ("[It would] be unrealistic to require police officers to always inform detainees that they are free to go before a consent to search may be deemed voluntary.").
109. Chief Justice Rehnquist devoted only two paragraphs to consideration of the potential need for informed consent. See id.
110. Chief Justice Rehnquist devoted only two paragraphs to consideration of the potential need for informed consent. See id.
111. Strauss, supra note 54, at 254.
112. Strauss, supra note 54, at 254.
113. Susan R. Klein, Lies, Omissions, and Concealment: The Golden Rule in Law Enforcement and the Federal Criminal Code, 39 TEX. TECH L. REV. 1321, 1322 (2007) ("Thus, the Court may be incorrect in its prediction that knowledge about rights will lead to the exercise of these rights, along with the negative effects on crime clearance."); Michaels, supra note 92, at 1380 ("The best guess here, though, is that the effect [of a warning] would, in any event, be minimal."); Nadler, supra note 9, at 205 ("Given the magnitude of situational pressures brought to bear on citizens in bus sweeps and similar situations, there is no reason to think that police advising citizens that they have a right not to cooperate with their request for consent to search will significantly reduce coercion experienced by citizens in this situation."). But see Loewy, supra note 11, at 554 (arguing that a warning requirement would "[o]f course . . . decrease the number of consent searches").
114. Robinette, 519 U.S. 33.
115. See Lichtenberg, supra note 63, at 367 (explaining that the percentage of people consenting to a search did not decrease when drivers were given a warning by the police).
116. Lichtenberg, supra note 63, at 367 (explaining that the percentage of people consenting to a search did not decrease when drivers were given a warning by the police).
117. Eight students conducted surveys. All but one of the students were white and four were female. All the students were paid at an hourly rate through a grant from Harvard Law School.
118. "Would you like to complete a one-minute, anonymous survey for Harvard Law School?"
119. "How people interact with the police."
120. To test whether or not survey-takers might provide biased results by guessing the real purpose of the survey, I tested a draft questionnaire on a group of Harvard Law School firstyear students who had not yet taken criminal procedure or studied the Fourth Amendment. Even among that relatively knowledgeable group, only nine out of sixty-five respondents, or 13%, believed the survey was meant to explore a search-and-seizure topic.
121. Surveyors did not use a truly random method, such as selecting every fifth person who walked past a reference point. It is unlikely, therefore, that the survey group was truly randomly selected. As the demographic data suggest, however, the sample did not seem skewed in any particular direction. See infra Part IV.A.
122. Before it was used in the field, the questionnaire was tested on sixty-five Harvard Law School first-year students to ensure that it could be completed quickly and without confusion. See NORMAN M. BRADBURN, SEYMOUR SUDMAN & BRIAN WANSINK, ASKING QUESTIONS: THE DEFINITIVE GUIDE TO QUESTIONNAIRE DESIGN 319 (Rev. ed. 2004) (explaining that pilot tests need not be conducted with a sample identical to the one targeted in the actual surveying). The pre-test indicated that the survey needed no significant changes. See note 113, supra (describing pilot test).
123. The average person shopping in Quincy Market, for instance, is likely to be of a different background than the average person shopping at the supermarket in Porter Square. Three of the locations, Quincy Market and the two train stations, were likely to over-sample wealthier people, either shoppers or commuters returning to Boston's suburbs.
124. Having respondents complete the survey themselves reduces, if not eliminates, bias that might have been attributable to differences in surveyors. See ROGER TOURANGEAU, LANCE J. RIPS & KENNETH RASINSKI, THE PSYCHOLOGY OF SURVEY RESPONSE 298 (2000). In addition, respondents' self-administration of the survey reduced concern that respondents would mischaracterize how free they felt to leave a situation as a result of embarrassment. See id. at 306.
125. See infra app. A.
126. Respondents were told, "Assume you do not want to talk to the officer," because the purpose of the survey was not to see whether people are generally helpful and cooperative with the police, but rather to explore the more legally significant question of whether people feel they must cooperate with the police even when they do not want to do so.
127. See infra app. A. Continuous scales with between four and six options are used widely in surveys. KEITH F. PUNCH, SURVEY RESEARCH: THE BASICS 59 (2003). While the survey might have been even more effective had the two ends of the scale been evenly balanced, the survey results all fall below the "somewhat free to leave" middle position, reducing any concern about imbalance.
128. See infra app. A.
129. Placing the "correct" answer, number four, at one extreme might have prompted people who were unsure about whether number three or number four was correct to select number three, inflating the number of people who seemed not to know their rights in the seizure situation. See HOWARD SCHUMAN & STANLEY PRESSER, QUESTIONS AND ANSWERS IN ATTITUDE SURVEYS: EXPERIMENTS ON QUESTION FORM, WORDING, AND CONTEXT 162-69 (1981) (discussing the tendency of people to discriminate against options at the extreme ends of a list). But someone who is unsure, for purposes of this study, is just like someone who is sure but wrong-neither person knows his rights.
130. The correct answer read: "You have the legal right to refuse to talk with the officer with no consequence to yourself." See infra app. A. No one is forced to comply with a police officer's request for questions. As the Court has explained, "[t]he personapproached . . . need not answer any question put to him; indeed, he may decline to listen to the questions at all and may go on his way." Florida v. Royer, 460 U.S. 491, 497-98 (1983). In addition, "refusal to cooperate, without more, does not furnish the minimal level of objective justification needed for a detention or seizure." Illinois v. Wardlow, 528 U.S. 119, 125 (2000) (citing Florida v. Bostick, 501 U.S. 429, 437 (1991)); see Steinbock, supra note 9, at 530 ("Unless a police officer has 'probable cause' to make an arrest or a 'reasonable suspicion' to conduct a 'stop and frisk', a person has the legal right to walk away from a police officer." (quoting PAUL BERGMAN & SARA J. BERMAN-BARRETT, THE CRIMINAL LAW HANDBOOK (3d ed. 2000)). Of course, doing more than merely refusing to cooperate might create such a justification. Wardlow, 528 U.S. at 124 (explaining that "unprovoked flight . . . [or] nervous, evasive behavior is a pertinent factor in determining reasonable suspicion").
131. The third option read: "You have a legal right to ignore the officer, but he may assume you are guilty of wrongdoing if you do." See infra app. A. There is some concern that the Court's decision in Wardlow, 528 U.S. 119, will open the door to allowing police to use refusal or retraction of consent as the source of probable cause. See Chanenson, supra note 65, at 416-32; see also Steinbock, supra note 9, at 543 (suggesting that good practice for a person walking away from the police is to "walk-do not run or otherwise seem to be 'evading' the police"); cf. WAYNE R. LAFAVE, SEARCH AND SEIZURE: A TREATISE ON THE FOURTH AMENDMENT § 8.2(f) (4th ed. 1996) (suggesting that "the suspect's earlier refusal to give consent is a factor which is properly taken into account [in] the 'totality of the circumstances' in judging the later consent under the Schneckloth formula" for determining whether consent is voluntary).
132. See infra app. A.
133. This defect in the question is unlikely to have caused problems. If we think that people believe the law is the same regardless of location, there is no problem. If we believe that people feel their rights differ by location, there is still little cause for concern. Since the actual right does not vary by location, at least one of the two answers given in each case would have been incorrect, and therefore, the overall answer for that respondent would also have been incorrect. Thus, at minimum, the wording of this question has no effect on the number of people who got the answer "right"; at maximum, the wording causes the data to understate respondents' legal ignorance.
134. See BRADBURN, SUDMAN & WANSINK, supra note 115, at 262 ("Demographic questions are almost always asked at the end of an interview, after the substantive questions."). 128 Zip codes were later used as proxies for income, although it is unclear whether zip codes are an effective proxy for individual incomes. See, e.g., Kathryn Moss et al., Unfunded Mandate: An Empirical Study of the Implementation of the Americans with Disabilities Act by the Equal Employment Opportunity Commission, 50 U. KAN. L. REV. 1, 86 (2001) (discussing problems with using zip codes as proxies for income); Margaret F. Brinig, Does Mediation Systematically Disadvantage Women?, 2 WM. & MARY J. WOMEN & L. 1, 14 n.56 (1995) (discussing a similar problem).
135. This question was written without enough specificity to ensure consistent results. Some respondents believed the question included traffic stops, such as being pulled over for a red light, while others believed it only included situations like the ones mentioned in the first part of the survey. Thus the analysis of the survey does not include these results.
136. See infra app. A. The two boxes on the lower left of the survey were to indicate male and female respectively, and the three boxes in the center corresponded to White, Black, and Other.
137. Only twenty-seven questionnaires, or 6.6% of responses, fell into this category. Even if all those questionnaires were dropped from the sample, the results discussed below would remain statistically significant and of the same order of magnitude.
138. There were eighteen surveys in this category.
139. No specific demographic or racial group had an average freedom-to-leave score that was far different from the overall sample average. Even if the surveyors over-selected for young respondents, for instance, the results would not be very different. See infra text accompanying note 158.
140. Income was measured using the median household income for each zip code from the 2000 U.S. Census. These results only suggest that no differences existed in this sample, and not that there are no differences between the groups in the actual world. More accurate measures of race and income, coupled with more diversity in the sample, might reveal differences. The question about previous stops by the police did not produce useable data. See supra note 129.
141. The responses break down as follows: 93 responses from Faneuil Hall, 168 at South Station, 78 at Porter Square, and 67 at Government Center / North Station.
142. Surveyors offered the questionnaire to a total of 1108 people. Four hundred and twenty-four people completed questionnaires, but eighteen results were discarded and counted as non-responses. See supra note 132. The response rate varied by survey site: 30.4% at Faneuil Hall, 65.2% at South Station, 19.2% at Porter Square, and 54.9% at Government Center / North Station. The higher response rates at the two train stations were likely the result of people having very little else to do while waiting for their trains.
143. All 406 respondents provided answers for both scenarios. Ten respondents did not provide a zip code, five did not provide an age, and two did not respond to the knowledgeof-rights question.
144. The larger-than-average number of people younger than twenty-five did not skew the data. See infra note 157.
145. Assuming the coding for race was accurate, this result makes sense. Two of the four locations were commuter rail terminals; therefore, the sample likely consisted of many people who lived outside of Boston.
146. U.S. Census Bureau, Boston (City) QuickFacts from the U.S. Census Bureau, http://quickfacts.census.gov/qfd/states/25/2507000.html (last visited Jan. 7, 2009).
147. U.S. Census Bureau, Norfolk County QuickFacts from the U.S. Census Bureau, http://quickfacts.census.gov/qfd/states/25/25021.html (last visited Jan. 7, 2009).
148. U.S. Census Bureau, DP-1 Profile of General Demographic Characteristics for the United States: 2000 http://factfinder.census.gov/servlet/QTTable?_bm=y&-geo_id=01000US &-qr_name=DEC_2000_SF1_U_DP1&-ds_name=DEC_2000_SF1_U (last visited Jan. 7, 2009).
149. There is a difference between how respondents believe they would feel and how they would actually feel. However, as discussed above, there is little reason to believe that respondents are overstating how free they would feel. See infra text accompanying note 159.
150. All 406 respondents answered both of these questions. While the averages at the four survey locations varied, the differences between averages for the sidewalk scenario were not statistically significant, and the differences for the bus scenario were different only at the ten percent level. The bus averages ranged from 2.81 to 2.39, but if each location had been weighted equally, the average freedom to leave would have been 2.55, reflecting almost no different from the sample average. There was also no statistically significant difference between the overall average for the sidewalk and bus scenarios. The absence of a statistically significant difference may suggest that the primary cause for feeling a lack of freedom is not the particular circumstances of an event, but rather the presence of the police officer. This result supports what commentators who have discussed the general coercive power of authority figures have said. See supra Part II.C.
151. 49.7% and 54.7%, respectively.
152. 78.0% and 78.8%, respectively.
153. The Census reports data in groups of ten years for people aged "25" and older, making "under 25" a natural category with which to capture high school students and college students. See U.S. Census Bureau, supra note 142.
154. Three hundred ninety-two of 406 respondents were coded for gender. The differences between men and women are statistically significant at the 1% level. While the overall difference in the sample between genders was strongly significant, the differences between genders were not significant at three of the four survey locations. While some randomness is to be expected in sub-samples-which is the very reason one draws from multiple sub-samples-there is no reason to believe that there is an interaction effect between location and gender.
155. The difference in these distributions is statistically significant at the 1% level.
156. See supra note 95.
157. There were no statistically significant differences between the racial categories or between levels of income.
158. There were 241 respondents who did not know their rights and 163 who did. People coded as "knowing their rights" selected the fourth option for the third question in the survey: "You have the legal right to refuse to talk with the officer with no consequence to yourself." Those coded as not knowing their rights selected one of the other three options. See supra Part III.B. That 60% of the sample did not know about their rights indicates "the truth of the matter, namely that most people do not expect that they have the right not to accede a police officer's request that a search be authorized." Burkoff, supra note 9, at 1138 (internal quotation marks omitted).
159. The differences between the two groups are statistically significant at the 0.1% level.
160. See supra Part III; infra app. A.
161. These percentages refer to the sidewalk scenario.
162. Thirty-seven percent in the sidewalk scenario and 42% in the bus scenario.
163. For instance, the average freedom-to-leave score in the sidewalk scenario is 2.61. The average freedom-to-leave score for people over the age of twenty-four is 2.65, suggesting that over-representing young respondents does not falsely suggest that people do not feel free to leave.
164. See, e.g., Nadler, supra note 9, at 171 ("[R]esearch confirms the difficulty of accurately imagining the extent to which situational constraints shape our behavior.").
165. In that sample of sixty-four Harvard Law School first-year students who had not yet taken criminal procedure, the average freedom to leave scores were 2.34 for the sidewalk and 2.13 for the bus. These results are slightly lower than the sample averages for people under twenty-five, 2.41 and 2.26 respectively.
166. See supra Part II.A. In none of the opinions using the Medenhall-Bostick standard does that Court introduce any qualifier to "free to leave" that might suggest that a person who feels "somewhat free to leave" is not seized.
167. See, e.g., Maclin, "Black and Blue Encounters," supra note 11, at 250-51 (treating the average person and reasonable person similarly); Steinbock, supra note 9, at 522-23 (explaining that the reasonable person standard "operates like a bell curve, with the reasonable person defined as a certain number of standard deviations from the mean . . . [I]n a statistical sense, the reasonable person standard accurately captures the beliefs and attitudes of the general population . . . ."). In fact, Justice Breyer's dissent in Yarborough v. Alvarado went so far as to suggest that the reasonable person should be defined with even more, not less, particularity than merely the "average person." 541 U.S. 652, 673 (2004).
168. California v. Hodari D., 499 U.S. 621, 628 (1991).
169. See, e.g., LaFave, supra note 20, at 738 ("the subjective expectations of the particular suspect [are not] determinative"); Robert V. Ward, Consenting to a Search and Seizure in Poor and Minority 8eighborhoods: 8o Place for a "Reasonable Person," 36 HOW. L.J. 239, 241 (1993) ("[T]he reasonable person test assumes that a person's interaction with the police is a generic experience . . . ."); see also Florida v. Bostick, 501 U.S. 429, 438 (1991) (making it clear that the reasonable person is presumed to be innocent).
170. See, e.g., Maclin, "Black and Blue Encounters," supra note 11, at 250 (arguing that the police and courts must consider the race of the person interacting with the police in determining if a seizure occurred); Martha Minow & Todd Rakoff, Is the "Reasonable Person" a Reasonable Standard in a Multicultural World, in EVERYDAY PRACTICES AND TROUBLE CASES 40, 59-60 (Austin Sarat et al. eds., 1998) (arguing that the reasonable person should be defined relative to "group membership, social relations among groups, issues of knowledge, and issues of dominance and subordination").
171. See Ronald J. Bacigal, Choosing Perspectives in Criminal Procedure, 6 WM. & MARY BILL RTS. J. 677, 720 (1998) (emphasizing the need for the Court to "direct[ly] access . . . the community's shared understandings"); Minow & Rakoff, supra note 164, at 42 ("The imagined 'reasonable person' . . . acquires the morals, intelligence, and knowledge reflecting community standards and perceptions, whether or not the particular individual actually held them.").
172. See the discussion, for example, in United States v. Mendenhall, 446 U.S. 544, 553-55 (1980). Of course, the reasonable person envisioned by the Court is an innocent person. Bostick, 501 U.S. at 438.
173. As discussed above, there is no reason to believe they do not. See supra Part IV.A.
174. See Steinbock, supra note 9, at 51 ("Since the Court has phrased its predominant test in terms of the reasonable person, the constitutional question could clearly be merged with the factual one: how would the average American perceive certain police practices?").
175. See, e.g., Burkoff, supra note 9, at 1129 ("[J]udges are arguably not doing a very good job-after the fact and far removed from the scene-of appreciating the actual coercive impact of a police officer's request for consent under these circumstances.").
176. See, e.g., Maclin, "Black and Blue Encounters," supra note 11, at 249-50 ("Common sense teaches that most of us do not have the chutzpah or stupidity to tell a police officer to 'get lost' . . . ."); Strauss, supra note 54, at 236 ("Numerous scholars and even judges have made the very basic observation that most people would not feel free to deny a request by a police officer."); cf. Bacigal, supra note 165, at 720 ("[L]acking direct access to the community's shared understandings, the Justices will not often find that a hypothesized reasonable person's assessment of Fourth Amendment reasonableness differs from the Justices' own assessment."); Simmons, supra note 57, at 800 ("The most frequent criticism
177. See supra note 60. Even beyond the discussion in Brendlin, the Court has sometimes been persuaded by empirical evidence in other areas of criminal law. See, e.g., District of Columbia v. Heller, 128 S. Ct. 2783, 2860 (2008) (Breyer, J., dissenting) (arguing that "the empirical evidence presented here [on gun control] is sufficient to allow a judge to reach a firm legal conclusion"); Stogner v. California, 539 U.S. 607, 649 (2003) (Kennedy, J., dissenting) (arguing that a legislature's determination about child abuse should be upheld because it "is amply supported by empirical studies"); Atkins v. Virginia, 536 U.S. 304, 316 n.21 (2002) (accepting results from polling data as evidence of a consensus against executing the mentally retarded); Withrow v. Williams, 507 U.S. 680, 714 (1993) (O'Connor, J., concurring in part and dissenting in part) (suggesting that the Court should reconsider Miranda's rule "when presented with empirical data"). On the other hand, the Court has not always been convinced by, or even willing to accept, empirical evidence. See, e.g., David L. Faigman, "8ormative Constitutional Fact-Finding": Exploring the Empirical Component of Constitutional Interpretation, 139 U. PA. L. REV. 541, 588-94 (1991) (surveying a series of cases where the Court has rejected empirical evidence that social scientists find persuasive);
178. Brandon L. Garrett, Judging Innocence, 108 COLUM. L. REV. 55, 57 n.4 (2008) (reporting the Court's general resistance to empirical evidence in Kansas v. Marsh, 548 U.S. 163 (2006)); see also Dorf, supra note 59, at 38 ("The typical Supreme Court opinion cites dozens of other Supreme Court cases, but scarcely any empirical data . . . ."). It is possible that the Court remains willing to consider empirical evidence only when the studies are "particularly unequivocal.
179. Shawn Kolitch, Comment, Constitutional Fact Finding and the Appropriate Use of Empirical Data in Constitutional Law, 10 LEWIS & CLARK L. REV. 673, 696 n.100 (2006).
180. See., e.g., Patricia M. Wald, Guilty Beyond a Reasonable Doubt: A 8orm Gives Way to the 8umbers, 1993 U. CHI. LEGAL F. 101, 106 ("Although judges daily proclaim piously that a reasonable person in those circumstances should have known she had the right to keep going, I doubt that any judge is completely convinced of that. Several of our D.C. Circuit cases have referred to it as a convenient, albeit necessary, fiction.").
181. Cf. Burkoff, supra note 9, at 1129 ("A second explanation for why judges routinely find such searches to be consensual is, to my mind, even more persuasive-and much more disturbing. Unfortunately, I believe that judges are holding that these searches are consensual strictly as a matter of what might be called knee-jerk, 'result stare decisis.' . . . [J]udges are following the lead of the Supreme Court in the application of prevailing consent doctrine, rather than following the consent-search doctrine itself and determining whether such consents have truly been tendered 'freely and voluntarily,' as the law requires.").
182. See Steinbock, supra note 9, at 537 ("[T]he Court's conception of a reasonable person has an implicit normative effect.").
183. The Court has never explicitly indicated that it includes such a prescriptive component in its standard. Some commentators, however, have argued that the Court's holdings in seizure cases can only be explained by a reasonable person who is "highly artificial," Edwin J. Butterfoss, Bright Line Seizures: The 8eed for Clarity in Determining When Fourth Amendment Activity Begins, 79 J. CRIM. L. & CRIMINOLOGY 437, 439 (1988), or even a reasonable "pachydermatous person"-a reasonable person with skin so thick that he still would believe he had a choice to leave when faced with police questioning. LaFave, supra note 20, at 739-40.
184. See Michigan v. Chesternut, 486 U.S. 567, 574 (1988) ("The test's objective standard-looking to the reasonable man's interpretation of the conduct in question-allows the police to determine in advance whether the conduct contemplated will implicate the Fourth Amendment.").
185. See supra text accompanying notes 106-08.
186. The Court has rarely addressed the question of how to consider the reasonable person's gender or age (or race). DRESSLER & MICHAELS, supra note 14, § 7.03(C)(3). Several justices in Mendenhall did suggest that age and gender were "not irrelevant factors" in the seizure analysis. See United States v. Mendenhall, 446 U.S. 544, 558 (1980) (plurality opinion); cf. Schneckloth v. Bustamonte, 412 U.S. 218, 226 (1973) ("In determining whether a defendant's will was overborne in a particular case, the Court has assessed the totality of all the surrounding circumstances . . . . Some of the factors taken into account have included the youth of the accused . . . .").
187. For instance, in Mendenhall, the Court found that a reasonable person who was twenty-two years old and female would not feel coerced when interacting with older, male police officers. 446 U.S. at 558.
188. LaFave, supra note 20, at 741 ("[A] literal application of the 'not free to leave' test would make virtually all police-citizen encounters of that type seizures . . . .").
189. The description Professor LaFave has provided of how the Court actually determines if a seizure occurs could also provide the wording for an alternative test. See supra note 20. 182 There is precedent for a standard that focuses on a "reasonable officer." See, e.g.,
190. Graham v. Connor, 490 U.S. 386, 396 (1989) ("[T]he 'reasonableness' of a particular use of force must be judged from the perspective of a reasonable officer on the scene . . . .").
191. See, e.g., Butterfoss, supra note 175, at 442 (1988) (proposing a per se rule based on the subjective intent of the officer initiating the encounter).
192. See Lynch, supra note 94, at 234 ("I don't know that scholars can penetrate these mysteries, either, but the broader perspective that comes from systematically surveying large numbers of police officers and defendants may give a more realistic view of that process than judges have.").