January 22, 2004.
Marilyn M. Trubey, Office of Federal Public Defender, Topeka, KS, for Defendant.
Randy M. Hendershot, Office of United States Attorney — Topeka, Topeka, KS, for Plaintiff.
This case came before the court for evidentiary hearing on the defendant's motions to suppress evidence and to dismiss the indictment. The court, having heard the testimony and reviewed the exhibits and pleadings, rules as follows.
On February 9, 2003, Russell County Sheriff's Deputy Kelly Schneider was on routine patrol, monitoring traffic. When westbound on 1-70, he noticed a gold car traveling eastbound, and determined by radar that it was going 77 m.p.h. in a 70 m.p.h. zone. Deputy Schneider routinely stops cars going that speed in a 70 m.p.h. zone, and decided to stop this one as well. Deputy Schneider testified that at the time he decided to stop the gold car, he did not know who was in the car, and did not know the number or race of the occupants.
Deputy Schneider then turned his patrol car around and pursued the vehicle for approximately two miles before catching up to it. Upon approaching the vehicle, he noticed that it was registered in California. He then pulled alongside the vehicle to determine how many people were in the vehicle, where they were located, and whether they were wearing seatbelts, as he routinely does. He gathers this information for the purpose of protecting his own safety, and testified that his intent is not to determine the race of the occupants, as race is immaterial to him both personally and professionally. Deputy Schneider testified that when he pulled alongside the vehicle, he saw two men in it. Although he was unsure of their race, they appeared to be Hispanic.
Deputy Schneider stopped the vehicle, and motioned for defendant to exit the vehicle. Defendant produced his license and registration, as requested. Deputy Schneider saw that defendant's license was from New York and that defendant had purchased the vehicle in California three days earlier. When asked whether he owned the vehicle, defendant told the deputy that he had taken a short trip to California, had been there one and onehalf days, had purchased the car, and was returning to New York where he lived. Deputy Schneider noticed that the registration showed defendant's address as Huntington Beach, California and not New York. He believed that defendant's statement that he had been in California only one and one half days contradicted the earlier date of purchase shown on the registration. He additionally found defendant to be "extremely nervous," as exhibited by lots of pacing, his hands shaking when he handed the deputy his license, and his shifting and moving around. Deputy Schneider testified that these acts of nervousness did not diminish throughout their contact, even after the deputy told the defendant he would receive a warning instead of a ticket.
After Deputy Schneider returned defendant's documents to him and was writing him a warning, he asked what defendant did for a living. Defendant told him that he did landscaping, but had not worked the last two months. Deputy Schneider found it unusual that defendant had purchased a new car when he was unemployed, and that defendant had traveled from New York to California to purchase a car which was not distinctive or unique.
As defendant opened the door and began to leave the patrol car, Deputy Schneider asked if he could ask him some more questions. Defendant agreed, and the deputy then asked whether defendant had any contraband or illegal drugs in the car. After defendant replied "no," the deputy asked if defendant would consent for him to look in the car. Defendant replied, "yes," and the deputy asked him to open the trunk, then asked the passenger to exit the car. Both complied with the deputy's requests and were cooperative throughout the encounter.
In the trunk, Deputy Schneider saw some duffle bags, then inspected the wheel well and the spare tire for drugs. He noticed that the floor in the trunk area was green, instead of gold as the original undercoating on the car would ordinarily have been. He saw that the caulking around the floor was new, thick and white, and that the carpet padding in or near the spare tire compartment was glued to the floor, both of which he found unusual. Suspecting a concealed compartment, he did a "finger test" inside the trunk and determined that the three inch gap he detected was too large, confirming his suspicions of a concealed compartment.
Because of his training and experience, Deputy Schneider knew that the reason for concealed compartments in vehicles such as this is to hide illegal drugs or contraband. He then approached the defendant and advised that there was a false compartment in the trunk, to which defendant said, "yes." Deputy Schneider then asked, "You know there's a false compartment in the trunk?" and defendant again replied, "yes." When the deputy asked whether there was anything in the compartment, defendant said "nothing," and told the deputy he could look. The vehicle was then transported to the Sheriff's Department, where it was searched. At no time did defendant protest, object, attempt to limit the scope of the search, or revoke his consent to search.
The search of the vehicle uncovered a hidden compartment in the trunk which contained approximately three kilograms of cocaine and one kilogram of heroin. Defendant and his passenger were then arrested and subsequently charged with possession with the intent to distribute these substances.
II. MOTION TO SUPPRESS
Defendant's motion to suppress challenges the validity of the initial stop, the scope and duration of the detention, and the validity of defendant's consent to search the car.
A. Initial Stop
As the Tenth Circuit established in United States v. Botero-Ospina, 71 F.3d 783, 787 (10th Cir. 1995), cert. denied, 518 U.S. 1007, 116 S.Ct. 2529, 135 L.Ed.2d 1052 (1996):
[A] traffic stop is valid under the Fourth Amendment if the stop is based on an observed traffic violation or if the police officer has reasonable articulable suspicion that a traffic or equipment violation has occurred or is occurring.
Thus this court's sole inquiry is whether this particular officer had a reasonable suspicion that this particular motorist violated "any one of the multitude of applicable traffic and equipment regulations" of the jurisdiction. See United States v. D'Armond, 65 F. Supp.2d 1189 (D.Kan. 1999).
In the present case, Deputy Schneider's testimony that defendant's vehicle was speeding, and that he routinely stops cars traveling 77 m.p.h. in a 70 m.p.h. zone for a violation of the traffic laws is credible and uncontradicted. Although Deputy Schneider admitted that he does not stop all speeders, and that the defendant denied speeding, the court does not find this sufficient to defeat the government's showing that the deputy had reasonable suspicion to stop the defendant's vehicle. Accordingly, defendant's challenge to the initial stop fails.
Defendant next challenges the scope of the detention. An officer making an investigative detention or a Terry stop must possess articulate reasonable suspicion. United States v. Carhee, 27 F.3d 1493, 1496 and n. 2 (10th Cir. 1994). Reasonable suspicion is one that would "warrant a man of reasonable caution in the belief that [a stop] was appropriate." Terry v. Ohio, 392 U.S. 1, 22, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968) (quotation and citation omitted). Police officers cannot rely upon an "unparticularized suspicion or hunch." Brown v. Texas, 443 U.S. 47, 52 n. 2, 99 S.Ct. 2637, 61 L.Ed.2d 357 (1979); United States v. Fernandez, 18 F.3d 874, 878 (10th Cir. 1994). "While the necessary level of suspicion is considerably less than proof of wrongdoing by a preponderance of the evidence, the Fourth Amendment requires some minimal level of objective justification." United States v. Gntierrez-Daniez, 131 F.3d 939, 942 (10th Cir. 1997) (quotation and citation omitted), cert. denied, 523 U.S. 1035, 118 S.Ct. 1334, 140 L.Ed.2d 494 (1998). The court considers both the quantity and quality of the evidence when evaluating whether there is reasonable suspicion.
The court looks at the factors enumerated at the suppression hearing, "both individually and in the aggregate, and determine[s] whether, under the totality of the circumstances, those factors give rise to a reasonable suspicion of criminal activity." United States v. Salzano, 158 F.3d 1107, 1111 (10th Cir. 1998) (citation omitted). The court "judge[s] the officer's conduct in light of common sense and ordinary human experience." United States v. Mendez, 118 F.3d 1426, 1431 (10th Cir. 1997) (citation omitted). "This approach is intended to avoid unrealistic second-guessing of police officers' decisions and to accord appropriate deference to the ability of a trained law enforcement officer to distinguish between innocent and suspicious actions." Gutierrez-Daniez, 131 F.3d at 941 (quotation omitted).
Generally, an investigative detention must "last no longer than is necessary to effectuate the purpose of the stop." United States v. Patten, 183 F.3d 1190, 1193 (10th Cir. 1999) (quoting Florida v. Royer, 460 U.S. 491, 500, 103 S.Ct. 1319, 75 L.Ed.2d 229 (1983)). It must be temporary, and its scope must be carefully tailored to its underlying justification. Gutierrez-Daniez, 131 F.3d at 942. A longer detention for additional questioning is permissible under two circumstances: (1) the officer has an objectively reasonable and articulable suspicion that illegal activity has occurred or is occurring; or (2) the initial detention changes to a consensual encounter. United States v. Hunnicutt, 135 F.3d 1345, 1349 (10th Cir. 1998).
Government's exhibit 1, a videotape of the stop, shows that the detention lasted approximately 11 minutes, six of which were spent waiting on the dispatcher to return information relating to defendant's driver's license. This length of time is not unreasonable under the circumstances of this case.
The scope of the detention is challenged because Deputy Schneider allegedly asked defendant questions unrelated to the traffic stop, while in the patrol car. Deputy Schneider contends that he was just being conversational with defendant while awaiting a response from the dispatcher, while defendant contends that the questions were illegally broad.
The Tenth Circuit has held in several cases that an officer conducting a routine traffic stop may inquire about "identity and travel plans," United States v. Rivera, 867 F.2d 1261, 1263 (10th Cir. 1989), and may request a driver's license and vehicle registration, run a computer check, and issue a citation. See United States v. Soto, 988 F.2d 1548, 1554 (10th Cir. 1993). An officer conducting a routine traffic stop may not ask questions unrelated to the purpose of the stop, even if the questioning does not extend the normal length of the stop, unless the officer has reasonable suspicion of illegal activity. United States v. Holt, 264 F.3d 1215, 1230 (10th Cir. 2001). However, officers can ask about travel plans, because such questions typically fall within the scope of a traffic stop. United States v. Williams, 271 F.3d 1262, 1267 (10th Cir. 2001).
The court agrees that certain questions asked by Deputy Schneider, such as how long defendant had been out of work, the amount he paid for the car, and whether he had paid in cash, may be unrelated to the purpose of the traffic stop. Nonetheless, by the time Deputy Schneider asked those questions, the following factors, viewed under the totality of the circumstances, had given rise to a reasonable suspicion of criminal activity: 1) defendant had traveled across the country to purchase a common vehicle in California instead of near his home in New York; 2) defendant was extremely nervous and remained so throughout their conversation; 3) defendant's statement to the officer regarding his address conflicted with information on the document the deputy saw; 4) defendant's statement about the date of purchase or dates he was in California conflicted with the date of sale on the document the deputy saw; and 5) defendant purchased a vehicle and traveled a great distance to do so at a time when he was out of work. Based upon these facts, the court finds that the detention did not exceed the scope of a permissible Terry stop.
C. Consent to Search
Defendant next challenges the search, contending that he did not understand what was being asked of him, and that the search exceeded the scope of his consent, if any.
A warrantless search is "per se unreasonable" unless one of the specifically established exceptions, such as consent, is present. Schneckloth v. Bustamonte, 412 U.S. 218, 219, 93 S.Ct. 2041, 36 L.Ed.2d 854 (1973) (quotations omitted). Valid consent is that which is freely and voluntarily given. Patten, 183 F.3d at 1194 (citation omitted). Voluntariness is a question of fact to be determined from the totality of all the circumstances. Schneckloth, 412 U.S. at 227, 93 S.Ct. 2041. A court makes this determination without presuming the consent was voluntary or involuntary. United States v. Hernandez, 93 F.3d 1493, 1500 (10th Cir. 1996).
For this exception to apply, the government must prove by a preponderance of the evidence that consent was freely and voluntarily given. Soto, 988 F.2d at 1557. The government does not discharge its burden "by showing no more than acquiescence to a claim of lawful authority." Bumper v. North Carolina, 391 U.S. 543, 549, 88 S.Ct. 1788, 20 L.Ed.2d 797 (1968). The government first "must present `clear and positive testimony that consent was unequivocal and specific and freely and intelligently given.'" United States v. Pena, 143 F.3d 1363, 1367 (10th Cir.) (quoting United States v. Angulo-Fernandez, 53 F.3d 1177, 1180 (10th Cir. 1995)), cert. denied, 525 U.S. 903, 119 S.Ct. 236, 142 L.Ed.2d 194 (1998). The government also must prove that the officers used no implied or express duress or coercion in obtaining the consent. Id. This determination is made upon considering the totality of the circumstances. Schneckloth, 412 U.S. at 225-27, 93 S.Ct. 2041.
1. Language barrier
The court first examines defendant's allegation that he did not understand English sufficiently to have consented to a search of his vehicle. Consent is valid if the facts show that defendant understood English well enough to respond to the officer's requests. See United States v. Sanchez-Valderuten, 11 F.3d 985, 990-91 (10th Cir. 1993) (finding adequate "receptive English language skills" to consent to the search, despite defendant's difficulty in speaking English); United States v. Corral, 899 F.2d 991, 994 (10th Cir. 1990) (defendant with limited knowledge of English voluntarily consented because he understood the officer's questions, answered questions in English and demonstrated an overall working knowledge of English).
Deputy Schneider testified that he had no difficulty communicating in English with defendant initially. At times, defendant asked him "what do you mean?" and the deputy would repeat himself, at which point the defendant responded appropriately to the questions. Deputy Schneider testified that the language barrier never made it necessary to stop their interaction. From Deputy Schneider's testimony and the videotape, the court finds that defendant understood the officer's questions sufficiently, and demonstrated an overall working knowledge of English. The challenge to his consent on the basis of a language barrier thus fails.
2. Scope of Consent
Defendant next alleges that the search exceeded the scope of his consent. Defendant admits that he consented to the deputy's "taking a look" in the vehicle, but denies that such consent extends to the deputy's physical intrusion into the vehicle itself.
The court examines the totality of the circumstances to determine whether a search remains within the boundaries of the consent given. United States v. Hooper, 47 Fed.Appx. 531, 534 (10th Cir. 2002). The general standard for measuring the scope of consent is based on "objective reasonableness." Pena, 143 F.3d at 1367-68. Scope "is generally defined by its expressed object," and "is limited by the breadth of the consent given." United States v. Elliott, 107 F.3d 810, 814-15 (10th Cir. 1997) (quotations and citations omitted). Failure to object to a search is an indication the search is within the scope of consent. See Pena, 143 F.3d at 1368. Where the expressed purpose of the search is to look for drugs or contraband, "that certainly implies that the officer could look wherever drugs might be hidden." United States v. Ramstad, 308 F.3d 1139, 1146-1147 (10th Cir. 2002) citing Pena, 143 F.3d at 1367-68.
Defendant's contention that he gave limited, rather than general consent to search, finds some support in United States v. Wald, 216 F.3d 1222, 1228 (10th Cir. 2000). There, the court found that consent was not general when limited to a "quick look inside the vehicle." The court recognized the general proposition that a defendant's "failure to object when the search exceeds what he later claims was a more limited consent is an indication the search was within the scope of consent," (citation omitted), but found that rule to apply only when the defendant initially gave a general authorization to search. Id. The court found no consent despite defendant's failure to object to the officer's search of the trunk, because the limited consent to take a "quick look inside" the passenger compartment did not encompass a prolonged look inside the trunk.
In United States v. Pena, 920 F.2d 1509 (10th Cir. 1990), cert. denied, 501 U.S. 1207, 111 S.Ct. 2802, 115 L.Ed.2d 975 (1991), the Tenth Circuit rejected an argument strikingly similar to the one made by this defendant. There, the scope of consent for an officer to "look" inside a car was allegedly exceeded by the officer's use of a screwdriver to remove the rear panel of the vehicle. The Tenth Circuit stated that it would not "attach an unduly restrictive meaning to the officer's request to `look' inside the vehicle," and concluded that the "search was conducted within the general scope of the permission granted." Pena, 920 F.2d at 1515.
Similarly, in Hooper, 47 Fed.Appx. at 534, the defendant told the officer that he could "go ahead and look" after the officer asked if he could "look inside the vehicle." While conducting his search of the van, the officer noticed that the plastic molding from the floor and the rear quarter panel did not seem to fit together properly, so he removed the ashtray from the molding and saw a package which contained illegal drugs. The Tenth Circuit rejected defendant's assertion that the officer's removal of the ashtray was beyond the scope of his consent to "look inside the vehicle," applying the court's reasoning in Pena.
Based upon the authorities noted above, the court finds no support for defendant's assertion that his consent for Deputy Schneider to "take a look" was limited, rather than general. The evidence shows that the search was within the scope of the consent granted. Accordingly, defendant's motion to suppress shall be denied.
III. MOTION TO DISMISS
By this motion, defendant moves the court to order the government to provide certain discovery relating to defendant's defense of racial profiling, and to dismiss the indictment because of racial profiling. Defendant contends that Deputy Schneider was motivated to stop defendant because of his race, thus the stop violates the equal protection clause of the Fifth Amendment. See Wayte v. United States, 470 U.S. 598, 608-09, 105 S.Ct. 1524, 84 L.Ed.2d 547 (1985). Defendant concedes that he has the burden to show not only adverse impact in the government's prosecutions or investigations of Hispanics, but also discriminatory intent.
In his discovery motion, defendant asks the court to order the government to produce the following:
1. All standards, policies, practices or criteria employed by the Russell County Sheriff to guard against the influence of racial, political or other arbitrary or invidious factors in the selection of cases and defendants for drug trafficking prosecution;
2. Regulations and policies governing the Russell County Sheriff units, including but not limited to any mission statements, and the amount of discretion exercised by individual officers;
3. Personnel files, records of complaints against Deputy Schneider, records of investigating bodies, and records of lawsuits and motions to suppress involving similar police conduct rising from Deputy Schneider's actions;
4. Document retention policies in the Russell County Sheriff (sic);
5. Records of stops made by Deputy Schneider from the beginning of calendar year 2000 to February 9, 2003, including but not necessarily limited to warning citations issued, notices to appear issued, arrests made, and evidence of traffic stops which did not result in any of the above. These records should include the race and ethnicity of the person stopped, vehicle and license information, and dispositions of the tickets and arrests.
6. Written training materials relating to Deputy Schneider's interdiction training by the Kansas Highway Patrol (sic).
A. Selective prosecution and selective enforcement
Although defendant's motion is styled as a motion regarding selective prosecution, the court finds that it is in fact a motion regarding selective enforcement. Defendant challenges the decision and actions of the law enforcement officer, not the exercise of discretion by the prosecutor. A defendant seeking discovery on a selective enforcement claim must meet the same ordinary equal protection standards outlined for selective prosecution claims, however. As stated in United States v. Barlow, 310 F.3d 1007, 1010 (7th Cir. 2002), cert. denied, ___ U.S. ___, 123 S.Ct. 2236, 155 L.Ed.2d 1123 (May 27, 2003):
Barlow complains not of selective prosecution, but of racial profiling, a selective law enforcement tactic. But the same analysis governs both types of claims: a defendant seeking discovery on a selective enforcement claim must meet the same "ordinary equal protection standards" that Armstrong outlines for selective prosecution claims. (Citations omitted). To prevail on his motion, therefore, Barlow needed to demonstrate that the agents' actions had a discriminatory effect and that the agents had a discriminatory purpose when they approached him in Union Station. (Citations omitted).United States v. Barlow, 310 F.3d 1007, 1010 (7th Cir. 2002).
B. Disparate effect and discriminatory intent
The threshold standard for obtaining such discovery is set forth in United States v. Armstrong, 517 U.S. 456, 116 S.Ct. 1480, 134 L.Ed.2d 687 (1996):
A defendant who claims he was targeted for prosecution because of race is entitled to discovery on that claim only if he presents "some evidence tending to show the existence of the essential elements" of the selective prosecution defense, i.e., discriminatory effect and discriminatory intent.United States v. James, 257 F.3d 1173, 1178 (10th Cir. 2001) (quoting Armstrong, 517 U.S. at 468, 116 S.Ct. 1480).
This standard, although it requires only "some evidence," is not insignificant.
In light of Armstrong's seemingly less stringent "some evidence tending to show" standard, the defendants need not establish a prima facie case of selective prosecution to obtain discovery on these issues. (citations omitted). Nevertheless, given the heavy burden that discovery can impose on the government, see Armstrong, 517 U.S. at 468, 116 S.Ct. 1480, 134 L.Ed.2d 687, the showing necessary to obtain discovery for a selective prosecution defense must "itself be a significant barrier to the litigation of insubstantial claims," id. at 464, 517 U.S. 456, 116 S.Ct. 1480, 134 L.Ed.2d 687.James, 257 F.3d at 1178. Defendant thus has the burden to show some evidence tending to show both disparate effect and discriminatory intent.
C. Disparate Effect
Defendant can meet its burden to show some evidence of a disparate effect by naming a similarly-situated individual of another race who was not prosecuted for the crime defendant allegedly committed, or through the use of statistical evidence.
To prove discriminatory effect in a racebased selective prosecution claim, a defendant must make a credible showing that a similarly-situated individual of another race could have been prosecuted for the offense for which the defendant was charged, but was not. Armstrong, 517 U.S. at 465, 116 S.Ct. 1480, 134 L.Ed.2d 687. If such a claim is based on the investigative phase of the prosecution, however, the defendant must instead make a credible showing that a similarly-situated individual of another race could have been, but was not, arrested or referred for federal prosecution for the offense for which the defendant was arrested and referred. Jones, 159 F.3d at 977. The defendant may satisfy the "credible showing" requirement by identifying a similarly-situated individual or through the use of statistical evidence. Chavez v. Illinois State Police, 251 F.3d 612, 636 (7th Cir. 2001).James, 257 F.3d at 1179.
Here, defendant intends to rely solely upon statistics to make his case. In support of his allegation of racial profiling, defendant submits a "Racial Profiling Study" published in February of 2003 by the Police Foundation in Washington, D.C. This study purports to show that certain law enforcement agencies in Kansas stop Hispanic drivers for traffic offenses in numbers disproportionately large in comparison to the number of Hispanic drivers on those same roads. Defendant additionally submits certain statistics reflecting the number of stops, by racial group, made by Deputy Schneider and others in the Russell County Sheriff's Department.
This study was requested by the Kansas legislature in 2000. See Kan. L.2000 Ch. 180.
1. Use of Statistics-generally
In United States v. Borrego, 66 Fed. Appx. 797 (10th Cir. 2003), the Tenth Circuit rejected a defendant's claim of selective enforcement of traffic laws, noting: "while helpful, purely statistical evidence is rarely sufficient to support an equal protection claim." The court found, "as a matter of law, long-standing equal protection jurisprudence has recognized that some measure of selectivity in the law enforcement arena is constitutionally permissible." Id. Because Borrego presented no evidence that the officer in that case used race as a factor in deciding to stop defendant's vehicle, and testified that he did not know the race of the occupants before stopping it, no prima facie case of selective enforcement of traffic laws was made.
The Tenth Circuit, in James, reaffirmed that a defendant cannot meet his burden by generic statistical evidence. There, the Federal Public Defender's office asserted that the government (the Topeka Police Department and the U.S. Attorney's office) improperly targeted black cocaine dealers. As evidence of discriminatory effect, defendant showed that the percentages of blacks arrested during three separate investigations in Topeka in 1998 and 1999 were 100%, 67%, and 93%, and that 92% of all crack cocaine defendants represented by their office since 1998 were black. This was contrasted to the fact that only 50% of persons who voluntarily committed themselves for treatment of crack cocaine addiction in 1999 were black. The Tenth Circuit affirmed the district court's denial of defendant's discovery requests, stating:
Of course, a defendant cannot satisfy the discriminatory effect prong by providing statistical evidence which simply shows that the challenged government action tends to affect one particular group. Rather, the proffered statistics must address the critical issue of whether that particular group was treated differently than a similarly-situated group. Chavez, 251 F.3d at 638.James, 257 F.3d at 1179. The James defendant erred in comparing the number of blacks arrested for cocaine offenses to the number of blacks voluntarily admitted to cocaine addiction treatment, rather than to the number of non-blacks who could have been arrested for cocaine offenses but were not arrested.
Statistical data has proven a useful tool in some high-profile state racial profiling cases. See, e.g., State v. Soto, 324 N.J. Super. 66, 734 A.2d 350 (1996) (statistical evidence that blacks were 4.85 times more likely than whites to be stopped for traffic violations established prima facie case of discriminatory effect). "Although statistics alone rarely establish an equal protection violation, they may be sufficient to establish the discriminatory effect prong of the Armstrong test. Chavez, 251 F.3d at 640. But such statistics must be relevant and reliable . . ." Barlow, 310 F.3d at 1011.
Statistics were based upon another survey by Dr. Lamberth, the same individual who conducted the study upon which our defendant relies.
In Barlow, the Seventh Circuit examined another study by Dr. Lamberth, the author of the Kansas Racial Profiling Study at issue here. It found that Dr. Lamberth used "flawed statistical methodology," and even if his methodology had been statistically valid, it was insufficient to meet defendant's burden under Armstrong because it stated nothing about the behavior of white travelers in that area and thus failed to provide any basis for concluding that any of the white travelers were similarly situated to defendant.
2. The Kansas Racial Profiling Study
Upon request of the parties, and subject to their express agreement, the court admitted certain exhibits in this case which had been previously introduced in another racial profiling case in this district, United States v. Mesa-Roche, 288 F. Supp.2d 1172 (D.Kan. 2003). Those exhibits are admitted in this case to the same extent that they were admitted in that case. See Gvmt. Exh. 2, p. 56. Accordingly, the court admits the data contained in the survey under the "public records" exception to the hearsay rule, Fed.R.Evid. 803(8), because it was compiled at the direction of the Kansas legislature. Its conlusions remain hearsay, however, given the lack of testimony by the author of the survey. See Beech Aircraft Corp. v. Rainey, 488 U.S. 153, 109 S.Ct. 439, 102 L.Ed.2d 445 (1988).
This and all subsequent references to Gvmt. Exh. 2 refer to the transcript of the hearing, and not to other documents contained therein.
The study (Def.Exh.1) used data collected over a four-month period beginning between October 18 and November 15, 2001, and ending between February 18 and March 15, 2002. It examined stops along certain sections of highway rather than the entire stretch of the highway system in the state or the entire length of I-70 within the state. Def. Exh. 1, pp. 119-120. The study attempts to measure the proportion of persons of certain racial or ethnic groups stopped on the relevant part of the highway (stop data), in comparison to the proportion of those racial or ethnic groups traveling on that same stretch of road (benchmark data). The focal point of the study, for purposes of this case, is its finding that the highway patrol recorded 6.8% stops of Hispanics between Colorado and Colby, in comparison to 1.8% of Hispanics reported to have been traveling that road, producing a 5% disparity.
To measure the racial composition of the traveling public, teams of stationary research assistants in some locations counted the cars driving by and decided whether the driver or another occupant appeared to be "Asian, Black, Hispanic, White, Other or Unknown," and determined their age and sex as well. Def. Exh. 1, p. 27-28. In other locations, a "rolling survey" was conducted by teams of observers in cars moving in traffic on the highway, which observed cars in traffic and noted what they believed to be the race, sex, and age of the driver of each car. These observations produced the benchmark data.
The study does not state whether the survey ors recorded their perceptions for occupants of each car they passed or which passed them, or whether they "selected drivers" for identification. Compare Gvmt. Exh. 2, p. 21 (expert's testimony that observers selected drivers) to Def. Exh. p. 28-29 (stating observers recorded all cars during stationary surveys, but not stating the same for rolling or highway surveys.) In the event surveyors selected drivers during rolling surveys, this adds increased unreliability or bias to the collection methodology.
3. Reliability of the Racial Profiling Study
Having carefully reviewed the study, the court finds it to be irrelevant and unreliable for the following reasons, some of which find support in the testimony of expert witness Dr. Brian Withrow, a statistician and researcher at Wichita State University.
a. The study states nothing about Russell County, or this Deputy in particular
Russell County, in which this defendant's stop occurred, was not included in any of the four sections of highway selected for the Racial Profiling Study. Further, Deputy Schneider has not been shown to have worked in any of the other three areas surveyed. Although the results of a random survey from other law enforcement agencies in other counties might be transferrable to this county and/or this deputy if one assumes that the percentages of racial groups in both the transient and local populations and drivers are the same in both locations, no reason has been shown to credit that assumption. In fact, the author of the study notes that an increased number of Hispanic motorists might be on the roadways in certain areas because of several plants that employ a large number of Hispanic motorists, and concludes: "this probably contributes to the increased disparities between the proportion of Hispanic motorists stopped and those on the roadways. . . ." Def. Exh. 1, p. 125. See also Gov. Exh. 2, pp. 50-52 (assumption underlying transferability would be valid if no other highways or routes between Colby and Russell would divert drivers from I-70); Gov. Exh. 2, pp. 92-97 (showing other highways between Colby and Russell.)
The fact that the survey was not random also prevents the court from finding that the results are transferable, as it is not possible to take what happened in one location and infer that it is representative of what is happening in another community. Id. at 167-68.
b. The benchmark data is not sufficiently reliable
A valid benchmark is one of the basic tenets of an accurate and reliable study. Gvmt. Exh. 2, p. 68. For the reasons which follow, the court finds that the benchmark data is not sufficiently reliable to warrant the court's reliance upon the study.
The court has serious concerns about the methodology in which the benchmark data was collected during rolling surveys. Some of these concerns are based in what the survey fails to do. For example, the study fails to show how the data was collected in the field, as would be routinely included in an appendix to such a study. Id. at 162, 165. Nor does the study record the "non-hit" rate, which would indicate the number of cars for which the surveyors were unable to determine the race of the occupants, or other details which would be important in verifying that the data was collected appropriately. Id. at 157. Further, the study collected data on fewer than 3,000 stops for all seven areas surveyed over a four-month period, rather than data on three to four thousand stops at one location over a period of two years, as defendant's expert conceded would be necessary to obtain a good sample. Thus the sample size is "too small to provide reliable statistical results." Fallis v. Kerr-McGee Corp., 944 F.2d 743, 746 (10th Cir. 1991).
Other concerns flow from information which is contained in the study. The study relies upon subjective and swift, if not split-second observations of Dr. Lamberth's surveyors for its collection of benchmark data regarding the race and ethnicity of drivers. In most cases, one person looked at a car and wrote down the race he or she believed the driver to be, and that conclusion served as the ultimate benchmark data. Yet one person's perception of what another's race is may be vastly different than someone else's perception of the same individual. Gvmt. Exh. 2, p. 172. The court agrees with the expert's observation that it would be "difficult to stand on the side of a road, particularly at night, and record both the race, gender, approximate age of a person driving a car, and at the same time turn and look at the license plate to figure out what county they came from, which is what they were attempting to do in those places by themselves." Id. at 158.
The court agrees with expert testimony that a better method of gathering benchmark data is to have two or three surveyors look at the same car and compare results to measure the extent to which surveyors uniformly perceived race. See Gvmt. Exh. 2, p. 164. Telling the difference between Whites and Hispanics is "a difficult task" and subject to error. Id. at 110-111. As the author of the study admits, "[w]hile there is little doubt that there is high reliability in determining race with regard to Blacks and Caucasians, there has been little empirical evidence that it is possible to make the same determinations accurately for Hispanics." Def. Exh. 1, p. 37. The difficulty of this task would appear to be compounded under the circumstances present in the rolling surveys, when surveyors were asked to record multiple data regarding motorists who they passed or who passed them in their vehicles.
Although "inter-rater reliability tests" were conducted at two intersections in Kansas City for some undesignated length of time, Def. Exh. 1, p. 37, those tests were necessarily done during stationary surveys. The study does not reveal that any such tests were conducted during a highway or rolling survey, or were conducted with any of the researchers who participated in rolling surveys. Because the ability of the researchers to distinguish between Hispanics and Whites under those circumstances has not been shown, their collection methodology is suspect, see id. at 16465, and the court has no assurance that they could accurately do what they attempted to do.
In these tests, two surveyors separately determine the race of drivers of the exact same cars, and their results are compared.
The court additionally notes the potential discrepancy between the nature of the stop data and the benchmark data, which could give rise to incorrect classification of persons. The benchmark data relied upon the subjective impressions of the surveyors to determine how many of which race were driving on the sections of highway studied, but the stop data relied upon either statements of the drivers or impressions of the officers who stopped them. Thus some persons who, when stopped, were reported in stop data as Hispanic, could have been recorded in the benchmark data as White because of their appearance as such to the surveyors.
c. The stop data lacks trustworthiness
The agencies and/or areas included in the study were not selected by the author at random, but were selected by the agencies who were to be the subjects of the study. Although the record does not reflect whether the agencies selected areas which they suspected would be problematic, it is clear that some basis for selection was used, thus the samples of roads and agencies included in the study are not random. In so doing, the study "breached an inviolate rule of the scientific method." Mesa-Roche, 288 F. Supp.2d at 1189. No evidence has shown a likelihood that the disparity represented in the study would be produced by a random selection from the law enforcement agencies or areas of highway in the state.
Additionally, the law enforcement officers who recorded the stop data were informed that their activity was going to be observed, and the purpose for such observation. Deputy Schneider, however, was not so informed. An awareness that one is being watched may lead a person to modify his behavior, or change the activity he would otherwise engage in. Gvmt. Exh. 2, p. 75. Law enforcement agencies were also required to self-report their stop data, and no attempt was made to verify it before it was included in the study. The court mentions these factors not to imply any bad faith by the law enforcement officers, but merely to show that the surveyors failed to use standard procedures in gathering the data. See id. at 170.
d. The study fails to rule out alternate explanations
Even the defendant's expert stated that having read the study, he could not give any opinion on racial profiling. Id., p. 30. Although the study, if accepted as reliable and trustworthy, shows that troopers between Colorado and Colby stop a disproportionately large number of Hispanics in comparison to the number of Hispanics driving on that stretch of road, it fails to show the reason for that event. Although racial profiling could be one possible explanation, both experts admitted that alternative explanations exist and are not addressed in the study. Id. at 32, 178-80. One such explanation is that "the violator population might have a different structure than what the benchmark data set has." Id. at 32.
This ties in to the government's suggestion that the typical behavior of drug couriers may result in their violating traffic laws more frequently than the motoring public at large, because they often drive long distances without sleeping, under strict deadlines and much pressure, and thus violate traffic laws more frequently than does the motoring public. See id, at 89. But no evidence was introduced on that issue or on the race of "typical drug couriers," so the court finds no support in the evidence for that argument.
Nonetheless, the study is based upon an assumption that the incidence of traffic violations for all racial and ethnic groups is the same. The court is similarly unaware of any evidence in support of this assumption, which may or may not be valid. Compare James E. Lange, PhD. Et al, Speed Violation Survey of the New Jersey Turnpike: Final Report (finding that black motorists have a statistically significant higher incidence of speeding than other races) with Mesa-Roche, 288 F. Supp.2d at 1192 n. 50 (finding New Jersey conclusion "suspect" and "seriously flawed" because of its methodology); see generally Armstrong, 517 U.S. at 469, 116 S.Ct. 1480 (rejecting an analysis that was based on a presumption that all races commit all types of crimes); James, 257 F.3d at 1179-80 (rejecting a challenge that was based on the assumption that whites and blacks are addicted to and distribute crack cocaine at an equal incidence).
Defendant's expert testified that he was unaware of any data indicating that drivers commit violations in exactly the same proportion to their representation on the highway, and admitted his uncertainty that the assumption underlying the study was correct. Gvmt. Exh. 2, p. 84-85. Defendant's expert also testified that he was aware of at least one other study showing that the violator benchmark is different than the racial distribution benchmark, id. at 86, and that this study would have been more accurate had it used a violator benchmark instead of a transient population benchmark. Id. at 116.
e. The study does not identify similarly-situated individuals
The study purports to show the percentage of Hispanic persons traveling on certain roads in Kansas in comparison to the percentage of Hispanic persons stopped for traffic offenses on those same roads. Although defendant was stopped for a traffic offense, this is not the offense that brings him before this court.
Defendant's arrest did not stem from his being stopped for a traffic violation alone, but from the deputy's decision to search the vehicle after the traffic stop. Although defendant may use statistics, his burden remains to "make a credible showing that a similarly-situated individual of another race could have been, but was not, arrested or referred for federal prosecution for the offense for which the defendant was arrested and referred. (citation omitted)" James, 257 F.3d at 1179. The offense for which the defendant was arrested was not a traffic offense, but a drug offense. Defendant has not attempted to provide credible evidence that similarly situated persons of other races could have been prosecuted for possession of drugs with intent to distribute, but were not.
To be relevant in showing disparate effect in this case, statistics should demonstrate that non-Hispanics stopped for traffic violations were not detained and searched, even if they displayed indicators of drug trafficking, while similarly situated Hispanics drivers were detained and searched. Cf., United States v. Bass, 536 U.S. 862, 863, 122 S.Ct. 2389, 153 L.Ed.2d 769 (2002) (a "credible showing" that "similarly situated individuals of a different race were not prosecuted was not made based on nationwide statistics demonstrating that `[t]he United States charges blacks with a death-eligible offense more than twice as often as it charges whites'" and that the United States enters into plea bargains more frequently with whites than it does with blacks); United States v. Turner, 104 F.3d 1180, 1184-85 (9th Cir. 1997) (noting that a defendant cannot establish a discriminatory effect by presenting evidence that a law was enforced exclusively against a particular ethnic group, finding that the defendants had not shown "that similarly-situated defendants of other races had been left unprosecuted"); United States v. Gaspar-Lupercio, 61 Fed. Appx. 422, 423 (9th Cir. 2003) (because defendant failed to present sufficient evidence that similarly situated non-Hispanic aliens were not prosecuted for illegal reentry, he failed to make the threshold showing required to obtain discovery).
4. Statistical Evidence re: Deputy Schneider
The court next addresses defendant's statistical evidence other than the Lamberth study. Defendant attaches a declaration from John Floyd, Dk. 45, Attachment A, an investigator for the Federal Public Defender's (FPD) office, who reviewed arrests for drug charges stemming from vehicle stops by Deputy Schneider on I-70 from January through October of 2002. Defendant contends those records show that Hispanics constituted 57% of such persons. Defendant's counsel additionally conducted a review of FPD records, and asserts they show that 52% of defendants represented by their office on drug charges arising from traffic stops from January 2001 to April 22, 2003, were Hispanic.
Such evidence consists of reports regarding Deputy Schneider's arrests of Hispanics for a period of approximately ten months, and the FPD's defense of Hispanics for approximately two years and four months. The government properly states that the records relating to Deputy Schneider are records of arrests, that the records relating to the FPD's office are records of charges, and that therefore they do not relate to the current claims of selective stops.
The record additionally shows that Deputy Schneider stops Hispanics more often than do other officers in the Russell County Sheriff's department. Def. Exhs. 3-12. The court finds this unremarkable given the facts of record that Deputy Schneider patrols I-70 almost exclusively, while his fellow officers patrol the surrounding city and county of Russell almost exclusively, and there are more Hispanics traveling on I-70 than there are living in the surrounding city and county of Russell.
The record additionally includes data comparing Deputy Schneider's stop data to KHP stop data between Colorado and Colby, as recorded in the study. That violator comparison shows that 33.69% of Deputy Schneider's stops were of Hispanics, while only 6.85% of the KHP stops were of Hispanics. Def. Exhs. 17, 19. Nonetheless, because the Lamberth study is fatally flawed for the reasons noted above, it cannot serve as a valid comparison for purposes of evaluating Deputy Schneider's stops.
Thus the comparison is to a violator benchmark.
For all the reasons stated above, defendant's statistical evidence fails to show disparate treatment in this case. But even in the event defendant had shown disparate treatment, he must also show discriminatory intent.
D. Discriminatory Intent
Although the Tenth Circuit has not analyzed the discriminatory intent element of this defense, other circuits require some evidence that the decisionmaker in the particular case acted with discriminatory purpose.
Plaintiffs must show that the "decisionmakers in [their] case acted with discriminatory purpose." McCleskey v. Kemp, 481 U.S. 279, 292, 107 S.Ct. 1756, 95 L.Ed.2d 262 (1987); Nabozny v. Podlesny, 92 F.3d 446, 453 (7th Cir. 1996). "`Discriminatory purpose' . . . implies more than . . . intent as awareness of consequences. It implies that the decisionmaker . . . selected or reaffirmed a particular course of action at least in part `because of' . . . its adverse effects upon an identifiable group." McCleskey, 481 U.S. at 298, 107 S.Ct. 1756, 95 L.Ed.2d 262 (quoting Pers. Adm'r of Mass. v. Feeney, 442 U.S. 256, 279, 99 S.Ct. 2282, 60 L.Ed.2d 870 (1979)); Hearne v. Bd. of Educ. of City of Chi., 185 F.3d 770, 776 (7th Cir. 1999) (same).Chavez, 251 F.3d at 645; See Wayte, 470 U.S. at 610, 105 S.Ct. 1524 (absent a showing that the government prosecuted defendant because of his protest activities, his claim of selective prosecution fails.)
To prove that the individual officer intentionally discriminated against the defendant, some courts find it insufficient for a defendant to rely on statistics which allegedly show discriminatory effect. See, e.g., Chavez, 251 F.3d at 647-48 (in this context, "statistics may not be the sole proof of a constitutional violation"); Armstrong, 116 S.Ct. at 1488-89, 116 S.Ct. 1480 (reversing the court of appeals' discovery order, holding that absent an appropriate basis for comparison, statistical evidence alone cannot establish any element of a discrimination claim); United States v. Olvis, 97 F.3d 739, 745-746 (4th Cir. 1996) (finding "even if the study had a basis for comparison that showed discriminatory effect, it would not necessarily prove discriminatory intent," (citing McCleskey, 481 U.S. at 297, 107 S.Ct. 1756) (holding that statistical evidence that race played a role in other death penalty cases was insufficient to create inference that prosecutors sought death penalty against particular defendant because of his race)); Barlow, 310 F.3d at 1012 (finding no evidence of discriminatory purpose where defendant failed to demonstrate that agents made any racial comments during their encounter with defendant, and there was no evidence that the agency's policy, either officially or de facto, encouraged racial profiling.)
Other courts permit an inference of discriminatory intent to arise from a statistically disparate impact of sufficient magnitude. See United States v. Heatley, 1999 WL 61816, *17 (S.D.N.Y. 1999); Soto, 324 N.J. Super. 66, 83-85, 734 A.2d 350 (1996) (finding that discriminatory intent may be inferred from statistical proof presenting a "stark pattern" of racial disparity, then burden shifts to government to introduce specific evidence showing either defects which bias the results, or missing factors, which when properly organized and accounted for, eliminate or explain the disparity).
The court finds that the better-reasoned rule places the burden on defendant to present some non-statistical evidence to demonstrate that Deputy Schneider acted with discriminatory intent when he stopped defendant. Because Deputy Schneider is not alleged to have engaged in any racial language or actions other than the stop itself, the court examines the study once again.
The study does not attempt to show whether the law enforcement officers who made the stops knew the race or ethnicity of the drivers before deciding to stop them. Gvmt. Exh. 2, p. 104-05. This is admittedly, a factor which would be "really good information" in attempting to explain why a disproportionate number of Hispanics was stopped. Id. This failure is, in the opinion of expert Withrow, "the most serious flaw in the research." Id, at p. 161, 734 A.2d 350. Although the study defined racial profiling as "the use of race or ethnicity as a basis for making decisions involving law enforcement activities," the methods used in the study failed to measure whether race was, or even could have been, one of the factors in the officer's decision to stop a car. Id.
Without proof of the officer's knowledge of a driver's race, an intent to discriminate against a driver because of his race can rarely, if ever, be shown. In order to have any relevance in showing racial profiling, the study must rely upon the assumption that all law enforcement officers knew the race or ethnicity of the drivers they stopped, before they stopped them. Yet no evidence has been shown to support this assumption. Instead, the only testimony was that during traffic enforcement, officers "almost never" know the race or ethnicity of an individual in a car before the stop. Gvmt. Exh. 2, p. 159-60. This is because officers are meeting traffic at relatively high speeds, focusing their attention on the particular car, and looking at the radar and a lot of other things in an attempt to figure out which car is going faster, all of which leaves them little, if any, opportunity to observe the race of the occupant of the vehicle before stopping it. Id.
For all the reasons set forth above, the court finds that defendant has failed to carry his burden of producing some evidence to make a credible showing of both disparate effect and discriminatory intent. Because the threshold showing required in Armstrong is not met, defendant's motion for discovery shall be denied.
The court finds it unnecessary to reach the government's additional contention that defendant's discovery request should be denied because the items he seeks are not in the government's possession.
Additionally, because defendant has failed to make a prima facie case of selective enforcement, his motion to dismiss shall be denied.
IT IS THEREFORE ORDERED that defendant's motion to suppress (Dk.32) and defendant's motion for discovery and to dismiss (Dk.45) are denied.