relying on drug addict's inconsistent testimonySummary of this case from United States v. Delgado-Cortes
ARGUED OCTOBER 1, 1999
DECIDED JANUARY 11, 2000
Appeal from the United States District Court for the Southern District of Illinois. No. 97 CR 40069 — J. Phil Gilbert, Chief Judge.
Ralph M. Friederich (argued), Office of the U.S. Attorney, Criminal Division, Fairview Heights, IL, J. Christopher Moore, Office of the U.S. Attorney, Benton, IL, for Plaintiff-Appellee.
Jeffrey M. Brandt (argued), Cincinnati, OH, for Defendant-Appellant.
Before Cudahy, Easterbrook and Kanne, Circuit Judges.
In October 1997, the Drug Enforcement Agency (DEA) received a tip that Roger Galbraith was manufacturing methamphetamine at his home in Cisne, Illinois. Acting on that tip, agents went to Galbraith's home where they found significant evidence of drug production. The agents waited at the home, interviewed Galbraith when he arrived, and later arrested him. Galbraith moved to suppress evidence gathered and statements given at the officers' original search; his motion was denied. Galbraith pleaded "not guilty," but changed his plea after jury selection began. The trial judge sentenced him on the basis of statements given by his codefendants and others. The judge enhanced Galbraith's offense level for obstruction of justice, and denied Galbraith's requests for sentence reductions based on acceptance of responsibility and application of the United States Sentencing Guidelines' so-called safety valve provision. Galbraith now appeals the denial of the motion to suppress, the judge's calculation of his relevant conduct, the obstruction of justice enhancement and the denials of his two requests for sentence reduction. We do not reach the merits of the suppression issue, and we affirm on all other issues.
On October 31, 1997, officers from the county sheriff's department and the DEA went to Galbraith's home. Galbraith lived in a house trailer, and maintained on his land a small cinder block building near the road and a shed closer to the residence. The officers smelled ammonia and other odors associated with methamphetamine manufacture. They traced the odors to the cinder block building, put on protective suits and breathing devices, and entered the building. There, they found methamphetamine production apparatus. They then approached the residence, but before reaching it, smelled ammonia and ether fumes coming from the shed. An agent later testified that the shed was open. Looking inside, the agent saw a tank leaking ammonia. He also saw Coleman fuel and empty ether cans. All are used in methamphetamine production. Next, the agents noticed ether and ammonia smells near the residence. When they examined the house, they saw a hose stuck in the front door, and heard rustling. The agents testified that they were concerned that occupants of the home might be in trouble because ether and ammonia can be dangerous. They entered the home, and found it empty, but noticed two jars that appeared to contain methamphetamine.
About three hours later, the Galbraiths drove onto their land. What happened next is disputed. Galbraith contends that he was immediately handcuffed. He also notes that two agents present at the scene have different recollections about whether the officers drew their guns, and when the officers read Galbraith and his wife their Miranda rights. The government contends that both were read their rights, waived them and were interviewed by the agents and released.
The government arrested Galbraith in November 1997, and he was released on bond. The government later moved to revoke the bond because Galbraith had allegedly continued to manufacture methamphetamine after his arrest. Galbraith was taken into custody on April 16, 1998. He moved to suppress evidence seized from his residence during the search described above. Galbraith also moved to suppress the statement he gave on that day. The trial court denied both motions. On April 23, a grand jury returned a two-count indictment naming Galbraith, his wife and five others as defendants. Count one alleged a conspiracy with six codefendants and two others indicted in related proceedings to possess and distribute methamphetamine in violation of 21 U.S.C. § 841(a)(1) and 846. Count two alleged conspiracy to manufacture methamphetamine, in violation of the same provisions.
Galbraith initially opted for trial, but changed his mind after jury selection began. At that point, he entered an unconditional plea of guilty to both counts. He was sentenced on March 3, 1999. At the sentencing hearing, DEA agent Christopher Hoyt was the government's sole witness. Hoyt established the amount of drugs Galbraith was responsible for based on Galbraith's post-search interview and the statements of two codefendants. In addition, Hoyt related the statements of a third man, George Songer, who claimed knowledge of additional drug amounts. Galbraith disputes the reliability of Songer's statement.
At sentencing, the judge accepted Songer's statement and used it to raise Galbraith's relevant conduct calculation. He also enhanced Galbraith's offense level by two levels for obstruction of justice, denied a two-level downward adjustment for acceptance of responsibility and found Galbraith ineligible for the safety valve provisions of United States Sentencing Guidelines section 5C1.2. Ultimately, the judge sentenced Galbraith to 151 months in prison. Galbraith now appeals the denial of his motion to suppress, the calculation of his relevant conduct, the obstruction of justice enhancement, the denial of his downward adjustment for acceptance of responsibility and the denial of safety valve treatment. We do not reach the merits of the motion to suppress, and we affirm the trial court on the remainder of the issues.
II. Analysis A. Motion to Suppress
Galbraith contends that the court below erred by denying his motion to suppress evidence seized from his property and to suppress the statement he gave on the day of the search. He argues that because the officers had no warrant, they were not entitled to go onto his land. He further argues that the exigent circumstances exception to the warrant requirement is not available to the officers because the situation was not sufficiently urgent to justify immediate action. The evidence was inadmissible because it was obtained in violation of the Fourth Amendment, he argues. And his post-search statement was inadmissible fruit of the poisonous tree. Whether or not these arguments have merit, Galbraith foreclosed his right to raise them when he entered an unconditional plea of guilty to the charges. He did not, as permitted by Federal Rule of Criminal Procedure 11, reserve the right to seek appellate review of the denied motion to suppress. See Fed.R.Crim.P. 11(a)(2). "[A] guilty plea constitutes a waiver of non-jurisdictional defects occurring prior to the plea. . . . This waiver includes Fourth Amendment claims." United States v. Cain, 155 F.3d 840, 842 (7th Cir. 1998) (citations omitted). Galbraith cites United States v. Yasak for the proposition that a guilty plea may not always amount to a waiver. 884 F.2d 996, 1000 (7th Cir 1989). But all we stated in Yasak was that the court could consider statements of conditionality at a defendant's plea hearing even if there was no written conditional plea, or was a written unconditional plea. 884 F.2d at 999-1000. A review of the transcript from Galbraith's plea hearing merely reaffirms that Galbraith's guilty plea was unconditional. Because Galbraith waived his right to appeal the suppression issue by entering this unconditional plea, we will not review his Fourth Amendment claims. See Cain, 155 F.3d at 842; see also United States v. Newman, 148 F.3d 871, 879 (7th Cir. 1998) (relinquishment of known rights constitutes waiver, which extinguishes appellate review, while rights forfeited by failure to raise them timely may be reviewed for plain error).
B. Relevant Conduct Calculation
Under the Sentencing Guidelines, the relevant conduct of one charged with manufacture, possession and distribution of drugs depends on the quantity of drugs manufactured, possessed and distributed. See U.S.S.G. §§ 2D1.1(a)(3), (c). At sentencing, DEA Agent Hoyt testified to statements given by Galbraith and Rodney Calhoun, a defendant in a related conspiracy case, which established that Galbraith had produced 361.46 grams of methamphetamine. Then Hoyt testified to statements given by codefendants Harlis Moulton, John Bierman and a third man, David Wood, which corroborated that amount. Next, Hoyt testified to two statements given by George Songer, who was introduced to Galbraith by codefendant Wood and bought methamphetamine from Wood starting in late 1996. According to Hoyt, some time after Galbraith's arrest, Galbraith asked Songer if he could cook methamphetamine at Songer's home. Hoyt related Songer's description of these sessions, and reported that Galbraith eventually taught Songer how to cook methamphetamine. In exchange for this tutorial, Songer alleged that he gave Galbraith a Ford Falcon, a motor and a chain saw. Hoyt testified that in total Songer saw Galbraith produce 148.83 grams of methamphetamine in addition to the amount established by Calhoun and those who corroborated his testimony. Hoyt also testified that in the second of the two interviews, Songer reported receiving a total of between 30 and 60 ounces (850.5 to 1701 grams) of Galbraith's methamphetamine. It is not clear whether he received this methamphetamine directly from Galbraith or through intermediate dealers.
All citations are to the 1997 Sentencing Guidelines, since those are the Guidelines the trial court found applicable.
The prosecutor did not rely on Songer's statements regarding either the 148 grams or the 850 to 1700 grams when calculating the drug amount constituting Galbraith's relevant conduct. The prosecutor stated that "[o]ur position will be that [the drug amount] falls between the 350 to 500 in accordance with the PSI. . . . Roger Galbraith establishes more than . . . 350 grams. That's corroborated by Calhoun. . . . If you put Songer in, it puts him nine grams over the 500, but I'm willing to concede that." Sent. Tr. at 47. The trial court, however, did rely on Songer's statements, noting that "[h]ere we have a Probation report that establishes relevant conduct between 350 and 500 grams, and yet I'm provided evidence that it's clearly over 500 grams. . . . [T]here's nothing to counter Songer here. The evidence that's produced is such that I don't see how this Court can make a finding below 500 grams. The evidence is clearly that it's over 500 grams." Sent. Tr. at 54. Later, speaking to Galbraith's attorney, the trial court stated: "[Y]our client takes the position that he doesn't even know Songer in his objections. And yet at this hearing I have been presented no evidence that he didn't know him. Your client's elected not to testify. If he was so convinced with his position and so sure of it, he probably would have testified. I'm not holding that against him. I'm just saying that I have evidence that this thing is over 500 grams." Sent. Tr. at 56. As a result of holding Galbraith responsible for more than 500 grams, the judge was required under the Sentencing Guidelines to place Galbraith's base offense level at 32 rather than 30.
Galbraith challenges the trial court's decision to credit information provided by George Songer. The district court's determination of the quantity of drugs involved in a defendant's conduct is a finding of fact reviewed for clear error. United States v. Lanterman, 76 F.3d 158, 160 (7th Cir. 1996). We will reverse a district court's conclusion regarding drug amount only if "after reviewing the record, we are left with the firm and definite conviction that a mistake has been made." Id. (quoting United States v. Corral-Ibarra, 25 F.3d 430, 437 (7th Cir. 1994)). The government has a considerable advantage in proving a defendant's relevant conduct. At sentencing it must prove the quantity of drugs only by a preponderance of the evidence. Lanterman, 76 F.3d at 160. Further, the Federal Rules of Evidence do not apply at sentencing, meaning the court may consider hearsay evidence and other information not admissible at trial. United States v. McEntire, 153 F.3d 424, 435 (7th Cir. 1998). Not only procedural but also substantive advantages go to the government in the contest over calculating relevant drug conduct. For instance, the testimony of just one witness, even a potentially biased witness, is sufficient to support a finding of fact. See United States v. Cedano-Rojas, 999 F.2d 1175, 1180 (7th Cir. 1993). Further, the trial court is entitled to credit testimony that is "totally uncorroborated and comes from an admitted liar, convicted felon, large scale drug-dealing, paid government informant." McEntire, 153 F.3d at 436 (quoting United States v. Garcia, 66 F.3d 851, 857 (7th Cir. 1995)).
A second matter of calculation is unchallenged. The Presentence Investigation Report states that Calhoun tied Galbraith to between 212.62 to 240.97 grams of methamphetamine. At the sentencing hearing, however, Hoyt presented Calhoun's testimony regarding a much larger amount — 361.46 grams. Apparently, this larger amount included some methamphetamine that Calhoun produced on his own. Hoyt testified that Calhoun was "directly involved with" Galbraith for the "joint purpose of manufacturing methamphetamine." Sent. Tr. at 17-18. Galbraith himself stated in his one interview with police that he provided Calhoun with cold pills needed to make the drug and knew Calhoun made methamphetamine on his land and elsewhere. Galbraith's attorney did not object at sentencing to the government's larger offer of proof, and admitted in his brief to this court that "the government met its burden on . . . 360 grams." Appellant's Br. at 22. Therefore, Galbraith has waived this issue, and we will not disturb it. See United States v. Newman, 148 F.3d 871, 879 (7th Cir. 1998). For the record, the Guidelines determine relevant conduct of a conspirator with reference to "all reasonably foreseeable acts and omissions of others in furtherance of the jointly undertaken criminal activity." U.S.S.G. § 1B1.3(a)(1)(B). The testimony that Galbraith acted jointly with Calhoun, and Galbraith's own statements suggest that he could reasonably foresee Calhoun's independent drug production; therefore, the entire 361.46 grams was properly included in Galbraith's relevant conduct.
There is one significant counterweight to these government advantages: the defendant has a due process right to be sentenced on the basis of reliable information. See Lanterman, 76 F.3d at 160. We have suggested that inconsistent evidence may be unreliable. See McEntire, 153 F.3d at 436. When evidence is inconsistent, the district court must undertake a "sufficiently searching inquiry into the government's evidence to ensure its probable accuracy." Id. This inquiry is particularly warranted where a witness has a history of drug use and admits his memory is not sharp. See id. (collecting cases). For instance, in McEntire, a witness offering information on the defendant's relevant conduct first stated in a proffer that he gave the defendant 50 pounds of methamphetamine; he then testified at trial that he gave the defendant 80 to 100 pounds; the witness later signed an affidavit stating that he could not estimate the amount, and then he stated at a sentencing hearing that he gave him more than 100 pounds. Id. The witness admitted that he used "a lot" of methamphetamine and that it sometimes affected his memory. Id. The trial court accepted the 80 to 100 pound estimate without specifying why that was the most reliable estimate, and we reversed and remanded for a more searching inquiry. Similarly, in United States v. Beler, 20 F.3d 1428 (7th Cir. 1994), a witness to relevant conduct first stated that he purchased 150 to 200 ounces of cocaine from the defendant, then disclaimed any ability to pinpoint an amount, and finally set the amount at 15 to 20 ounces. See id. at 1430-33. We found the judge's decision to credit the smaller amount unacceptable because he failed to explore the factual basis for these bare estimates. See id. at 1433-34. Similarly, in McEntire, we rejected as unreliable a witness's "conclusory estimates . . . not supported . . . with any further explanations or details as to how he arrived at the amounts." 153 F.3d at 437. Thus, consistent facts, details and explanations suggest the reliability we require before crediting one of several inconsistent statements.
Galbraith complains that Songer had been convicted of prior drug offenses, and was facing possible drug charges himself. He also decries the fact that Songer's testimony was uncorroborated. These facts do not necessarily render him unreliable. See, e.g., Cedano-Rojas, 999 F.2d at 1180 (testimony of one biased witness may be sufficient to support a finding of fact); McEntire, 153 F.3d at 436 (trial court may credit uncorroborated testimony of a convicted felon and government informant).
We are more disturbed by the fact that Songer's two interviews yielded vastly different information. In Songer's first interview, he stated that Galbraith cooked about six ounces of methamphetamine in his home. No mention was made of additional drugs procured through Galbraith. In the second interview, Songer again stated that Galbraith cooked between six and eight ounces of methamphetamine. He added, however, that he had purchased between 30 and 60 ounces of methamphetamine produced by Galbraith. Songer's initial silence and later loquaciousness on this score are troubling though not, strictly speaking, contradictory. The wide gulf between these stories, coupled with Songer's admitted heavy drug use — he became addicted to hallucinogenic drugs at age 15; to methamphetamine at age 18, to cocaine powder at age 20, to crack cocaine at age 35 and apparently used between 850 and 1700 grams of methamphetamine during the period in question — suggest that the trial court might have performed a more searching inquiry than it did. The judge noted the discrepancy, and decided to credit Songer's statement about the 148 grams, but not the larger amount. He did not explain why he found the latter statement reliable but the former suspicious. Given the government's disavowal of Songer's entire statement, an explanation demonstrating the judge's scrutiny of the evidence might have been advisable.
Nevertheless, after reviewing the record as we are directed to do by Corral-Ibarra, 25 F.3d at 437, we are not left with a definite and firm conviction that the judge was mistaken to credit the testimony on the 148 grams but not on the larger amount. The testimony regarding the 148 grams bore indicia of reliability — facts and details — that were missing from the statement regarding the larger amount. Songer twice told officers that he observed Galbraith cook approximately this amount of methamphetamine in his home. He specified exact quantities cooked at various times. He described the ingredients used each time, and explained who provided those ingredients. He told officers about the items he gave Galbraith in exchange for the cooking lessons, including a car. This detailed testimony was far more reliable than the mere conclusions rejected in McEntire and Beler. In contrast, Songer estimated just once, and in passing, that he purchased between 30 and 60 grams of Galbraith's methamphetamine. He did not specify the quantities he purchased on various dates, or the locations of the purchases. Further, his statement suggests that he did not purchase all of the drugs directly from Galbraith, but instead through intermediate dealers. Therefore, Songer could very likely have misunderstood who supplied the drugs to his dealer. This single, isolated statement, like those rejected in McEntire and Beler, did not bear sufficient indicia of reliability. The judge was correct to ignore it. We are, therefore, not left with the conviction that the trial judge made a mistake in crediting Songer's testimony on the 148 grams. There was no clear error in the relevant conduct calculation, and we affirm it.
Unfortunately, as Galbraith points out, the government did not try to locate the car to corroborate Songer's story. However, because uncorroborated testimony is not necessarily unreliable, this oversight is not important.
C. Obstruction of Justice
Galbraith also contests the trial judge's decision to enhance his sentence two levels for obstruction of justice as provided for in section 3C1.1 of the Guidelines. The Presentence Investigation Report (PSR) recommended the enhancement because of Galbraith's alleged perjury at the suppression hearing. Galbraith did not object below to this enhancement, and thus we review the judge's decision to enhance under the plain error standard. See United States v. Santoro, 159 F.3d 318, 320-21 (7th Cir. 1998). "A plain error is not only a clear error but an error likely to have made a difference in the judgment, so that failure to correct it could result in a miscarriage of justice, that is, in the . . . imposition of an erroneous sentence." Newman, 965 F.2d at 213.
If a defendant does not object to the enhancement at the time of sentencing, the judge is entitled to adopt the PSR's findings without making independent findings on the record. See Fed.R.Crim.P. 32(b)(6)(D) ("Except for any unresolved objection . . . the court may, at the [sentencing] hearing, accept the presentence report as its findings of fact."). See also United States v. Dunnigan, 507 U.S. 87, 95 (1993) ("[I]f a defendant objects to a sentence enhancement resulting from her trial testimony, a district court must review the evidence and make independent findings. . . ."). Galbraith did not object to the enhancement at sentencing, and the judge both adopted the PSR and took judicial notice of the suppression hearing at which the alleged perjury took place. We take Galbraith's specific challenges in logical order.
First, Galbraith argues in essence that two of his alleged perjurious statements (that officers did not read his Miranda rights, and threatened him to alter the content of his statement) were not false at all. Instead, he posits, the trial judge wrongly credited the government's version of events rather than his own. Naturally, we review the trial court's credibility determination with great deference. See, e.g., United States v. Agostino, 132 F.3d 1183, 1198 (7th Cir. 1997). The judge below heard the testimony and decided whom to believe, and in light of Galbraith's subsequent turnaround on several of these issues, we cannot say this decision was erroneous.
Second, Galbraith complains that, even if his statements were false, the judge did not follow the proper procedure and go on to find that the falsehoods were material and willful. As discussed above, the judge needed only to adopt the PSR's findings to satisfy the requirements of Dunnigan. The PSR did not include a specific finding of materiality. Because Galbraith did not dispute the enhancement, this failure is reviewed for plain error. That is, we ask whether the record's silence on materiality was likely to have made a difference in the judgment. See Newman, 965 F.2d at 213. Here, because the PSR specifically detailed the falsehoods and the context constituting perjury, the judge was not prevented from drawing a meaningful conclusion about materiality. Further, the judge himself presided at the suppression hearing and took judicial notice of that hearing at the sentencing proceeding. The judge was better qualified than the author of the PSR to assess the materiality of Galbraith's perjury, and the PSR's silence did not affect the judgment or amount to plain error.
Next, Galbraith argues that, on the merits, his falsehoods were not material. We have stated that false testimony is material if it is "`designed to substantially affect the outcome of the case.'" United States v. Parker, 25 F.3d 442, 448 (7th Cir. 1994) (quoting Dunnigan, 507 U.S. at 95). In both Parker and Dunnigan, the lies at issue were alleged to have had a bearing on the final outcome of the case, namely the defendant's guilt or innocence. The situation here is slightly different, because Galbraith testified falsely at a suppression hearing rather than a trial. Thus, in a strict sense even the most blatant falsehood he told would have had a direct effect only on the judge's evidentiary ruling and at best an indirect effect on the outcome of the case. Notably, however, the Sentencing Guidelines define materiality for non-perjurious false statements to law enforcement officers or probation officers as those that if believed would affect the issue under determination. See U.S.S.G. § 3C1.1, Application Note 5. This definition essentially applies the Dunnigan "outcome of the case" standard to lies in settings where something less ultimate than guilt or innocence is at stake. Further, logic dictates that a lie influencing a pretrial issue will, in an attenuated sense, influence the ultimate outcome of the case itself. Thus, both Dunnigan and the Guidelines suggest that a falsehood told at a pretrial hearing is material if it is calculated to substantially affect the issue under determination at that hearing.
This result was foreshadowed by United States v. Emenogha, 1 F.3d 473, 485 (7th Cir. 1993), in which we affirmed the district court's application of an obstruction of justice enhancement based on perjury at a pretrial hearing. In Emenogha, the defendant stated at a suppression hearing that prior to executing a consent to search form, he and his family had been threatened by law enforcement officers. This lie was perjury because if believed it would have negated the defendant's consent and perhaps rendered the search unconstitutional and its fruits inadmissible. Galbraith told two similar lies that if believed would have influenced the outcome of the suppression hearing. He told the court that on the day of the search, law enforcement officers did not read him his Miranda rights before he gave a statement. See Suppression Hrg.Tr. at 110. And he told the court that law enforcement officers threatened him with jail in order to make him alter the content of his statement. See id. The trial court, in its Memorandum and Order denying the motion to suppress, specifically stated that in deciding whether to admit Galbraith's statement, "[t]he only question for the Court . . . is whether the defendants were informed of their Miranda rights." Mem. at 8. Obviously, if this was the only question, and the trial judge had believed Galbraith's version of events, he might well have suppressed Galbraith's statement. Therefore, at least two of Galbraith's falsehoods at the suppression hearing were designed to influence the outcome of the issue under determination, namely the admission of the statements. Other falsehoods, regarding the manufacturing operation in his home in the month before the search, were not obviously relevant to the outcome of the suppression hearing, and were not material.
Galbraith now argues that if the search of his home prior to his arrest was conducted in violation of the Fourth Amendment, his statements were fruit of the poisonous tree whether or not he was read his Miranda rights, thus rendering his falsehood on this point at the suppression hearing immaterial. This logic is, of course, specious. Materiality turns on a statement's effect if believed. When Galbraith spoke, the legality of the search and the admissibility of his post-search statements were both under review. Therefore, when Galbraith spoke, it was possible the judge would find the search legal. If so, his belief in Galbraith's lies about the Miranda warnings would have influenced his decision whether to suppress the post-search statements. Thus, at the time Galbraith made them, and still today, the falsehoods were material.
Galbraith next argues that the PSR erred in finding the falsehoods willful. Galbraith argues persuasively that several of the "false" statements identified in the PSR merely reflect his inability to follow the prosecutor's compound or complex questions. But the "confused" statements were the ones concerning Galbraith's pre-arrest drug activity, which would have no influence on the outcome of the motion to suppress and were therefore immaterial. Galbraith's material falsehoods, regarding the Miranda warnings and the officers' threats, were offered in response to simple questions posed by his own attorney. Galbraith's confusion on immaterial issues does not offset his willful untruths on material issues, and the finding of perjury was not plainly erroneous. The perjury finding justified the obstruction of justice enhancement, which we affirm.
Mr. Isaacson: Were you ever read those rights that day?
Galbraith: No, sir. . . .
Mr. Isaacson: Did [the officers] tell you anything before you made any statements?
Galbraith: That if I didn't tell them what they wanted to hear that I would do life in jail.
D. Acceptance of Responsibility
Galbraith next appeals the district court's finding that he failed to accept responsibility and therefore did not qualify for the two-level reduction in base offense level offered by the Sentencing Guidelines. See U.S.S.G. § 3E1.1 (two-level decrease if "defendant clearly demonstrates acceptance of responsibility for his offense"). Whether a defendant has accepted responsibility is a factual determination that the trial court is to make as a result of its conclusions about the defendant's conduct and credibility. United States v. Scott, 145 F.3d 878, 885 (7th Cir. 1998). We review a trial court's findings on acceptance of responsibility for clear error. See id.
The Guidelines specifically state that "timely" notice of the intention to enter a guilty plea is a clear demonstration of acceptance of responsibility. See U.S.S.G. § 3E1.1(b)(2). Conversely, last-minute guilty pleas do not demonstrate the requisite acceptance. See, e.g., United States v. Ewing, 129 F.3d 430, 436 (7th Cir. 1997) (affirming denial where defendant pleaded guilty on last business day before start of trial). Galbraith waited until jury selection was underway before pleading guilty. This plea was nothing if not last-minute, and therefore the trial court's finding that it did not demonstrate acceptance of responsibility was not error. The judge also noted that Galbraith had obstructed justice, a finding that ordinarily is inconsistent with acceptance of responsibility. See, e.g., Ewing, 129 F.3d at 435. Occasionally, in an extraordinary situation, a defendant may initially obstruct justice and later accept responsibility. See id. Here, however, after his performance at the suppression hearing, Galbraith declined to submit to additional interviews with the government and maintained his "not guilty" plea until the eleventh hour. The situation was ordinary, and the obstruction of justice enhancement is further support for our conclusion that the judge did not err in denying the "acceptance of responsibility" reduction to Galbraith.
E. Safety Valve Reduction
Finally, Galbraith contends that the judge erred by refusing to apply the provisions of section 5C1.2 of the Guidelines, known as the "safety valve" provision. Under this provision and U.S.S.G. § 2D1.1(b)(6), if Galbraith met five specific criteria, the court could lower his offense level by two. The defendant bears the burden of proving he is eligible for the safety valve reduction. See United States v. Ramirez, 94 F.3d 1095, 1099-1102 (7th Cir. 1996). We review the trial court's finding on this issue for clear error. United States v. Ramunno, 133 F.3d 476, 482 (7th Cir. 1998). The trial court found that Galbraith foundered on the fifth criteria: "not later than the time of the sentencing hearing, the defendant [must] truthfully provide to the Government all information and evidence the defendant has concerning the offense or offenses that were part of the same course of conduct or of a common scheme or plan. . . ." See U.S.S.G. § 5C1.2(5). Galbraith gave one statement to law enforcement agents, on the day his home was searched. Following that, he did not make a proffer and did not expand on his initial statement regarding his codefendants. The proffers of his codefendants suggest that Galbraith was not forthcoming in his post-arrest interview. The judge's finding that Galbraith did not truthfully provide all information he had was not clearly erroneous, and we affirm it.
In sum, we dismiss Galbraith's challenge to the denial of the motion to suppress. We affirm the trial court's relevant conduct determination, the two-level enhancement for obstruction of justice, the denial of the acceptance of responsibility reduction and the denial of the safety valve provision.