UNITED STATES of America, Plaintiff–Appellee, v. Andrew MODJEWSKI, Defendant–Appellant.
Andrew Modjewski presented the testimony of a psychiatric expert during his sentencing hearing in hopes that it would reduce his sentence for possession and delivery of an extremely large amount of child pornography. At the sentencing hearing, after the defense conducted its direct examination and the government cross-examined the expert, the district court judge engaged in lengthy and direct questioning of the expert, much of it regarding whether Modjewski could be classified as having a pedophilic identification. Though the expert said Modjewski could not, the judge came to her own contrary conclusion based on her knowledge of the field. Modjewski now argues the judge should have sua sponte recused herself. We reject Modjew-ski's argument that the judge's questioning rose to the level of personal bias since she was testing the reliability of the expert's opinion. We also find neither the judge's general knowledge of the relevant field nor her determination that Modjewski was more accurately classified having a pe-dophilic identification constituted personal knowledge of a disputed evidentiary fact, since neither was a fact. Further, we find that Modjewski waived his right to challenge the district court's purported failure to address the arguments in mitigation raised in the appeal. We therefore affirm Modjewski's sentence.
Modjewski pled guilty to three counts of possession and transportation of electronic child pornography. His collection was massive, consisting of over 12,500 images and 700 videos. One analyst from the National Center for Missing & Exploited Children called it the “most complete collection” of child pornography she had investigated.
At sentencing, Modjewski presented the expert testimony of Dr. Lisa Rone, an Assistant Professor of Clinical Psychiatry at Northwestern University, Feinberg School of Medicine. Dr. Rone reiterated the findings in her report that Modjewski suffered from post-traumatic stress and bipolar disorders. She testified that he had minimal risk factors for re-offending since he did not have a personality or impulse control disorder and was being adequately treated. She also opined that he was not a pedophile, based on her “experience with people who have had post-traumatic stress disorder from childhood sexual abuse and treating them” and her discussion with Modjewski. The government elicited testimony that Dr. Rone's opinions were based on only one meeting with Modjewski and she did not conduct any tests on him, relying on tests conducted by others, and she did not review any of the images he downloaded or traded. After the government rested, the district court judge asked questions of Dr. Rone. Those questions spanned eleven minutes and eleven pages of transcript, and primarily related to Dr. Rone's diagnosis that Modjewski was not a pedophile. For example:
Court: So in reaching your conclusion that he's not a pedophile, the type of images that he has possessed or collected would be relevant, wouldn't it?
Dr. Rone: They absolutely would be relevant.
Court: And there would be a difference, in your opinion, if, for example, he had pornography in the area of-child pornography of 16–to 18–year–old only. That would make a difference in your assessment of whether he was a pedophile, right?
Dr. Rone: Well, again, your Honor, I was making the assessment about his proclivity to being a pedophile not on the basis of just the images, but also on the basis of his psychiatric history and his own experiences.
Court: Right, but that's not my question. My question is: It would make a difference in making an assessment of pedophilia if these images were solely those between 16– and 18–year–olds, right?
Dr. Rone: Certainly it would make a difference. I don't think we would be talking about pedo-philia—
Court: You wouldn't be able to diagnose him with pedophilia.
Dr. Rone: That's correct.
Court: Okay. So when you say aroused by prepubertal images, if an individual has those in a collection and you didn't view them, how do you know that he's not a pedophile? You never looked at them, right?
Dr. Rone: Your honor, I don't think I needed to look at them based on the descriptions I read.
Defense counsel did not object to the court's questions (aside from an occasional relevance objection) or the fact that the court was questioning the expert. When the judge finished, defense counsel asked Dr. Rone further questions on redirect.
At the hearing's conclusion, the judge noted: “I am not going to make any finding today, nor do I need to make any finding today, that this defendant will act out on any behaviors and will be a contact offender. I don't think that is necessary under the guideline calculations or the Sentencing Commission's report. I don't think that this psychiatrist who testified had enough experience in the field to opine on that conclusion, nor will I.” The court continued:
The Court does not credit [Dr. Rone's] finding that [Modjewski is] not a pedophile. And I don't think it matters if I make a determination of the term, because the term is really not necessary for my findings. But I don't think that one can come to a conclusion under the DSM–5 [Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition] that an individual is not a pedophile if she hasn't taken into account the actual images, viewing the actual images, looking at the types of images, the time period that they were stored, the amount of time that they were accessed, and the period of time over which he accessed the materials. That's critical to understanding whether someone has a preference and a motivation and a determination to obtain these images under her own definition․
The Court doesn't come to this study blindly. As the parties know, I've—I have continually read on the subject and stay up on the literature in this area and have written a treatise on the issue of child exploitation. And so for the appellate record, I have referenced some of the materials from “Child Exploitation and Trafficking: Examining the Global Challenges and U.S. Responses,” many of which do support some of the things that the doctor said, but not all of the things.
I'm not, again, making a conclusion that he's acted out, nor that he will act out. That's not a conclusion I think that I need to make. So with that in mind, the Court does find that the number of images, the type of images, the long period of time that the images were distributed, and the exacerbating fantasy and/or direct languages contained in the chats show a very aggravating circumstance and an individual who is obsessed with sexual activity with prepubescent individuals, and that that prepubescent activity also includes harm. And, therefore, with that in mind, a pedophilic identification, in the Court's opinion, is much more accurate.
The court found three mitigating factors and sentenced Modjewski to a 15–year sentence, which was 30 months below the 210–262 month advisory guideline range. After the sentence was imposed, the government asked, “if the defendant has other issues he felt should have been addressed by the [sentencing judge] in [her] statement [explaining the sentence], that he raise them now at this time.” Thereafter, the sentencing judge, addressing Modjewski's counsel, stated, “I have read all of the filings, but if you feel that there was something that I needed to address specifically, ․ I would be happy to do so.” In response, counsel only raised two issues related to treatment programs for Modjewski.
Modjewski appeals his sentence.
A. Recusal Was Not Required
Modjewski argues the district court judge should have sua sponte recused herself because she was biased or because a reasonable person would question her impartiality. He argues both 28 U.S.C. § 455(a) and § 455(b)(1) required recusal. Under § 455(a), a judge shall recuse herself “in any proceeding in which [her] impartiality might reasonably be questioned.” Under § 455(b)(1), a judge should recuse herself where she “has a personal bias or prejudice concerning a party, or personal knowledge of disputed evidentiary facts concerning the proceeding.” Modjewski must present “compelling evidence” of bias so that a reasonable person would be convinced the judge is biased. Grove Fresh Distribs., Inc. v. John Labatt, Ltd., 299 F.3d 635, 640 (7th Cir.2002).
First, we reject the government's argument that Modjew-ski's motion was untimely. He appealed to this court when the relevant acts-the judge's questions and statementscame to light. See United States v. Diekemper, 604 F.3d 345, 351 (7th Cir.2010) (“[W]e have permitted a judicial bias concern to be raised after trial when the bias did not become known until the trial's cessation .”). However, he did so without filing a motion to disqualify with the district court or a writ of mandamus with this court. We cannot review a motion under 28 U.S.C. § 455(a) without such filings. Id. at 352. Recusal under § 455(a) is of little use to Modjewski anyway since “it's too late for us to order the judge removed from the case, because she's through with it.” In re Bergeron, 636 F.3d 882, 884 (7th Cir.2011); see also Diekemper, 604 F.3d at 352. Our ability to review a recusal argument under § 455(b) without the writ and what standard of review we use are “less clear.” Diekemper, 604 F.3d at 351. As the government points out, we have reviewed recusal arguments under § 455(b) for the first time on appeal without a writ or motion filed with the district court, and have done so under both the clear and plain error standards. Id. (clear error); United States v. Smith, 210 F.3d 760, 764 (7th Cir.2000) (same). But see United States v. Lara–Unzueta, 735 F.3d 954, 958 (7th Cir.2013) (plain error); United States v. Ruzzano, 247 F.3d 688, 695 (7th Cir.2001) (same). Ultimately the argument fails under either, so we assume without deciding that we can review the matter.
Modjewski argues the judge, “in essence, stepped into the role of the prosecution” when she asked numerous direct questions of the expert “for the purpose of countering [her] conclusions.” He argues this demonstrates the judge's “personal bias or prejudice.” 28 U.S.C. § 455(b)(1). A judge may examine a witness. Fed.R.Evid. 614(b). However, she abuses that authority when she “abandons [her] proper role and assumes that of an advocate.” Fed.R.Evid. 614 advisory com-mittee's notes; cf. United States v. Alfaro, 336 F.3d 876, 883 (9th Cir.2003) (applying Rule 614 at sentencing). One of the main reasons that such questioning can be prejudicial is that it can influence the jury. See United States v. Barnhart, 599 F.3d 737, 744–46 (7th Cir.2010) (finding judge in jury trial erred by asking questions that “read like a cross-examination”). Here, there was no jury, and “in non-jury proceedings, questioning by the judge will rarely be prejudicial to the defendant.” United States v. Webb, 83 F.3d 913, 917 (7th Cir.1996). Moreover, many of the formalities that come with a criminal trial are relaxed in sentencing. At sentencing, a court “may appropriately conduct an inquiry broad in scope, largely unlimited either as to the kind of information [she] may consider, or the source from which it may come.” United States v. Johnson, 489 F.3d 794, 796–97 (7th Cir.2007) (internal quote omitted).
While the tone and content of the judge's questions read somewhat like a cross-examination, it is the judge's role to fashion a sentence based on information with a sufficient in-dicia of reliability. See id. The judge also plays the role of the “gatekeep[er]” when it comes to expert evidence, ensuring that it “both rests on a reliable foundation and is relevant to the task at hand.” Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 597 (1993). At times a determination of reliability will call for lengthy questioning and, at others, for direct questions (if for no other reason than efficiency). See Webb, 83 F.3d at 917 (finding no abuse of discretion in judge's questioning, which was “intended to speed the suppression hearing along”). Here, there were eleven minutes of mostly to-the-point questions. Yet the sheer volume of questions cannot per se show the questioning is improper since there is no principled way to determine how many questions are sufficient or excessive, whether the judge was required to ask more as a result of an uncooperative witness, and so forth. See, e.g., United States v. Kidding, 560 F.2d 1303, 1314 (7th Cir.1977) (holding no error in judge asking fifty-six questions over twelve pages of transcript during bench trial). The judge did not cut off the witness, demean her or impeach her. Rather, the judge posed various pointed inquiries to “question and even challenge” the reliability of Dr. Rone's testimony. United States v. Vallone, 698 F.3d 416, 468 (7th Cir.2012), reinstated by United States v. Vallone, 752 F.3d 690 (7th Cir.2014). Moreover, to meet his burden that a judge should be disqualified based on “judicial remarks during the course of a trial that are critical or disapproving of, or even hostile to, counsel, the parties, or their cases,” Modjewski must show “such a high degree of favoritism or antagonism as to make fair judgment impossible.” Liteky v. United States, 510 U.S. 540, 555 (1994). Modjewski has not presented compelling evidence that the judge reached that level by asking a large number of direct questions while determining which evidence was reliable.
Modjewski next contends that the judge's general knowledge about child exploitation was a “disputed evidentiary fact[ ] concerning the proceeding” that was used to contradict the expert and therefore required her recusal. 28 U.S.C. § 455(b)(1). But, the judge's broad extrajudicial knowledge was not a “fact[ ] concerning the proceeding.” She has written a treatise on the subject, as she stated on the record, but there is no evidence she knew any extrajudicial facts about Modjewski. Moreover, all of the information she knew was available to the general public. Therefore, she did not need to recuse herself. See, e.g., In re Hatcher, 150 F.3d 631, 635 (7th Cir.1998) (finding no recusal necessary under § 455(b)(1) since the judge “learned nothing ․ that any member of the public could not also have learned”); United States v. Bonds, 18 F.3d 1327, 1331 (6th Cir.1994) (Boggs, J., in chambers) (holding recusal not necessary in criminal case involving DNA matches despite judge's attendance at DNA conference); United States v. Payne, 944 F.2d 1458, 1476 (9th Cir.1991) (finding “expertise on and exposure to a subject ․ does not necessitate recusal” in sexual abuse case where judge had been on Attorney General's Commission on Pornography).
Nor does the fact that the judge discounted the expert's opinion that Modjewski was a pedophile mandate recusal since Dr. Rone's own testimony-not any “personal knowledge of disputed evidentiary facts” allegedly held by the judge-provided the reasons to reject the diagnosis. The first reason the judge gave for dismissing Dr. Rone's opinion was because “I don't think that one can come to a conclusion under the DSM–5 that an individual is not a pedophile if she hasn't taken into account the actual images.” Dr. Rone testified to that exact point, stating it would “absolutely be relevant” under the DSM–5 to consider the type of images that Modjewski has possessed or collected. Dr. Rone also admitted she did not view those images. The judge then discredited the conclusion because Dr. Rone did not determine the age group of individuals in the images, which Dr. Rone admitted would “certainly ․ make a difference” in assessing pedophilia. Dr. Rone acknowledged she did not make such a determination. So, based on Dr. Rone's own testimony, the judge rejected the conclusion as unreliable. See United States v. Garthus, 652 F.3d 715, 720 (7th Cir.2011) (affirming sentence in child pornography case when the “judge was un-persuaded [by an expert's testimony] that the defendant would forgo criminal activity after release from prison”); United States v. Coopman, 602 F.3d 814, 818 (7th Cir.2010) (affirming sentencing judge who rejected expert's testimony because judge “harbored serious concerns about the doctor's specific experience, methods, and analysis” in the child pornography case).
The judge's more problematic action, from Modjewski's perspective, is not that she knew facts from outside of the proceeding, but that she knew enough to ask informed questions. Yet, to hold that a judge may not ask such questions would be to ignore that “she sits to see that justice is done in the cases.” Collins v. Kibort, 143 F.3d 331, 336 (7th Cir.1998) (internal quotation omitted). The judge credited that testimony which she found to be reliable and discredited that which she did not, while giving reasons for both. That is the role of the judge at sentencing, and Modjewski does not challenge those determinations. To find recusal necessary here, we would punish the judge for being well-informed. Yet, we have previously noted that judges are not subject to “eternal[ ] disqualif[ication]” in cases where they know the subject matter well. See Schurz Commc'ns., Inc. v. FCC, 982 F.2d 1057, 1061 (7th Cir.1992) (Posner, J., in chambers); see also United States v. Alabama, 828 F.2d 1532, 1543 (11th Cir.1987) (per curiam) (“All judges come to the bench with a background of experiences, associations, and viewpoints․ A judge is not required to recuse himself merely because he holds and has expressed certain views on a general subject.” (internal citations omitted)). Any other ruling would result in, for example, judges being forced to sit idly by if one party presented uncontradicted, but known to the judge as wrong, evidence.
Finally, Modjewski points to the court's affirmative finding of his pedophilic identification as cause for recusal. But, again, the judge's opinion was not a “fact” and therefore no recusal was necessary. Cf. Fed.R.Evid. 702 (noting expert “opinion[s]” based on “facts”). The court acknowledged as much by defining Dr. Rone's pedophilic identification as a “conclusion” and the court's own classification as her “opinion” on the matter. By comparison, the facts in the case were Modjewski's actions and his collection, and there is nothing in the record showing the court had any extrajudicial knowledge of those facts.
However, the judge's own diagnosis was not reliable evviously held that sentencing based on “speculation or unfounded allegations” constitutes plain error. See United States v. Halliday, 672 F.3d 462, 475 (7th Cir.2012) (reversing because district court focused on defendant's purported belief in the lawfulness of his offenses without any evidence defendant held that belief); cf. United States v. Bradley, 628 F.3d 394, 400 (7th Cir.2010) (per curiam) (“Sentencing judges necessarily have ‘discretion to draw conclusions about the testimony given and evidence introduced at sentencing,’ but ‘due process requires that sentencing determinations be based on reliable evidence, not speculation or unfounded allegations.’ “ (internal citation omitted)); United States v. England, 555 F.3d 616, 622 (7th Cir.2009) ( “[D]ue process requires that sentencing determinations be based on reliable evidence, not speculation or unfounded allegations.”). This plain error was not raised by either party, but that does not prohibit our review. “In exceptional circumstances, especially in criminal cases, appellate courts, in the public interest, may, of their own motion, notice errors to which no exception has been taken, if the errors are obvious, or if they otherwise seriously affect the fairness, integrity, or public reputation of judicial proceedings.” United States v. Atkinson, 297 U.S. 157, 160 (1936); see also Fed.R.Crim.P. 52(b) (noting plain error “may be considered even though it was not brought to the court's attention”).
The judge's opinion that Modjewski exhibited a pe-dophilic identification cannot be based on Dr. Rone's expert opinion, since Dr. Rone testified to the contrary. Therefore, the only basis for this opinion came from the judge herself, but such a conclusion cannot be a reliable expert opinion here. Expert opinions are based on specialized knowledge and result from reliable principles and methods being applied to the facts of this case. See Fed.R.Evid. 702. Whether someone should be diagnosed as a pedophile or having a pedophilic identification is a conclusion for experts to make. See, e.g., United States v. Boroczk, 705 F.3d 616, 620 (7th Cir.2013) (noting conflicting expert testimony on pedophilic identification); Garthus, 652 F.3d at 719 (noting expert testified as to pedophilic fantasies); United States v. Mantanes, 632 F.3d 372, 376 (7th Cir.2011) (same). But the judge was not admitted as an expert. She was not subject to a Daubert hearing, and it is unclear who would even conduct such a hearing. She did not disclose her methods for classifying Modjewski, nor was she subject to cross-examination on her opinions. Before a purported expert opinion is found reliable, it is “crucial that a Daubert analysis of some form in fact be performed.” Naeem v.. McKesson Drug Co., 444 F.3d 593, 608 (7th Cir.2006) (internal quotation omitted). That was not done here. Since there was no other evidence in the record upon which the judge could support her conclusion, her finding of a pedophilic identification was speculative and an error. Cf. United States v. Miller, 601 F.3d 734, 739–40 (7th Cir.2010) (reversing where judge based sentence on belief that sex offenders have higher than normal recidivism rates, but there was an “absence of support in the record for the court's views”); United States v. Dorvee, 604 F.3d 84, 94 (2d Cir.2010) (finding sentence substantively unreasonable when district court made the “assumption” defendant “was a ‘pedophile’ likely to engage in sexual conduct with a minor” despite “expert record evidence to the contrary”).
We must now ask whether the error affected the defend-ant's substantial rights by evaluating whether the district court's error was not only plain “but also likely to ‘have resulted in a different sentence.’ “ United States v. Corona–Gonzalez, 628 F.3d 336, 341 (7th Cir.2010). Modjewski bears the burden of persuasion as to this question. See United States v. Olano, 507 U.S. 725, 735 (1993); Corona–Gonzalez, 628 F.3d at 341. Modjewski fails to establish that the sentencing judge's pedophilic identification finding was likely to have resulted in a different sentence. While it is problematic that she made such a statement, in context of the whole hearing, we find that she did not rely upon it. The record shows that the sentencing judge repeatedly stated that she did not need to make a finding of pedophilic tendencies to support the sentence. See, e.g., Sent'g Hr'g Tr. 118, Dist. Ct. Dkt ., ECF No. 156 (“I [do not] need to make any finding today, that this defendant will act out on any [sexual] behaviors [toward children] and will be a contact offender.”); id. at 119 (“I don't think that the psychiatrist who testified had enough experience in the field to opine on [whether Modjewski will engage in sexual activity with a child] nor will I.”); id. (“[I]t will not be a part of my conclusion that he will or will not act out on his urges.”); id. at 127 (“The [c]ourt does not credit [Dr. Rone's] finding that [Modjewski] is not a pedophile. And I don't think it matters if I make a determination of the term, because the term is really not necessary for my findings.”); id. at 129 (“I'm not, again, making a conclusion that he's acted out [sexually against children], nor that he will act out. That's not a conclusion I think that I need to make.”); see also id. at 132 (”I am going to go below the guideline range, but I can't go to a level that would probably make everyone feel comfortable in your family because of the seriousness of the offense.). The sentencing judge explicitly and repeatedly denounced reliance on Modjewski's pedophilic tendencies, and under the circumstances here, we conclude that the finding was not likely to have resulted in a different sentence. Cf. McGee v. United States, 462 F.2d 243, 246 (2d Cir .1972) (finding the error likely impacted the sentence “absent some explicit indication to the contrary by the trial judge at the time of sentencing”). Therefore, while the error is plain, we cannot vacate the sentence because the plain error did not affect Modjewski's substantial rights.
B. Mitigation Arguments Were Addressed or Waived
“A sentencing court must address a defendant's principal arguments in mitigation unless they are too weak to merit discussion .” United States v. Garcia–Segura, 717 F.3d 566, 569 (7th Cir.2013). Modjewski argues that the district court never addressed all the arguments in mitigation put forth in the eight motions and exhibits that he filed requesting variances or contesting the government's or the probation officer's positions. However, the district court directly addressed many of Modjewski's arguments in mitigation including the U.S. Sentencing Commission's report to Congress about child pornography offenses under U.S.S.G. § 2G2.2 (the guideline used in this case), Modjewski's risk of recidivism, Modjew-ski's cooperation with law enforcement, the circumstances and nature of the offense, and offender characteristics, including Modjewski's own victimization.
Modjewski also argues that the district court did not specifically address his arguments regarding the continued risk of being assaulted in prison and the sentencing disparity among circuits in child pornography cases. Modjewski waived these issues for appeal. After imposing the sentence, the district court explicitly asked Modjewski's counsel whether he wished to raise any arguments. See Sent'g Hr'g Tr. at 137 (“I have read all of the filings, but if you feel that there was something that I needed to address specifically ․ I would be happy to do so.”). Modjewski did not raise the arguments he makes now, and instead, raised only two issues related to treatment programs. As a result, Modjewski waived the arguments in his appellate brief that the sentencing judge should have explicitly addressed his risk for assault in prison and the sentencing disparities among circuits in child pornography cases. See Garcia–Segura, 717 F.3d at 569 (encouraging sentencing courts after imposing the sentence “to inquire of defense counsel whether they are satisfied that the court has addressed their main arguments in mitigation” and finding that if the argument was not raised, a later challenge for failure to address the argument would be considered waived); see also United States v. Donelli, 747 F.3d 936, 940–41 (7th Cir.2014) (finding waiver of defendant's argument that the district court failed to address a principal argument in mitigation where at the sentencing hearing the judge asked defendant's counsel if he “required ‘any further elaboration’ of the reasons for the sentence,” and defendant's counsel did not raise the argument raised on appeal).
We therefore AFFIRM Modjewski's sentence.