Full opinion text
OPINION KATHARINE S. HAYDEN, District Judge. INTRODUCTION Pronouncing sentence on David Grober has required this Court to look long and hard at the sentencing guidelines and the truly remarkable punishment the government seeks under them. The ultimate tension for a sentencing judge is between a mechanical application of the sentencing guidelines on the one hand, and a fair, reasonable sentence that does justice on the other. The crime of conviction in this case is David Grober’s downloading of child pornography from the internet, having accessed images of child pornography through file-sharing and attachments to emails. The ultimate question in terms of punishing him is; When is enough enough? In Koon v. United States, 518 U.S. 81, 113, 116 S.Ct. 2035, 135 L.Ed.2d 392 (1996), the Supreme Court wrote: “It has been uniform and constant in the federal judicial tradition for the sentencing judge to consider every convicted person as an individual and every case as a unique study in the human failings that sometimes mitigate, sometimes magnify, the crime and the punishment to ensue.” In United States v. Faulks, 201 F.3d 208, 209 (3d Cir.2000), then Chief Judge Becker wrote about the act of sentencing as the district judge’s “most important judicial responsibility, whose daunting character has not been eliminated by the Sentencing Reform Act and the Sentencing Guidelines.... ” In this post-Booker world, the Supreme Court wrote in Rita v. United States, 551 U.S. 338, 127 S.Ct. 2456, 2469, 168 L.Ed.2d 203 (2007), that the district court’s “reasoned sentencing judgment rest[s] upon an effort to filter the Guidelines’ general advice through § 3553(a)’s list of factors.... ” This is how the guidelines’ “general advice” works in David Grober’s case. Grober entered a plea of guilty without stipulations to all counts in a six-count superseding indictment. The sixth count is for possession of child pornography, and the other five counts charge individual incidents of “receipt” and “transportation” of child pornography over the internet. The possession count carries a sentence of up to 10 years, and each of the receipt and transportation counts carries a mandatory minimum of five years and a statutory maximum of 20 years. According to the sentencing computation in the Presentence Report (“PSR”), the base offense level for Grober’s crimes, after grouping the counts for sentencing purposes, is 22, calling for a sentencing range of 41-51 months. But specific offense characteristics laid out in § 2G2.2, the sentencing guideline applied by United States Probation in the Presentence Report (“PSR”), increased the offense level by 18 levels to level 40. Because Grober pleaded guilty after the government had taken substantial steps to prepare for trial, he was afforded a two-level adjustment for acceptance of responsibility. At level 38, then, Grober — a first-time offender who entered a plea and accepted responsibility for his crimes — faced a guidelines sentencing range of 235 to 293 months, most of which exceeds the statutory maximum of 240 months/20 years. It is this staggering sentence that the government is seeking. The sentencing guidelines “should be the starting point and the initial benchmark” in all sentencing proceedings. Gall v. United States, — U.S.-, 128 S.Ct. 586, 596, 169 L.Ed.2d 445 (2007). But when the bottom of the guidelines for a defendant entering a plea is just months shy of the 20 year statutory maximum, one gets busy asking questions about how that happened. This Court has asked itself: Am I working with a rational sentencing structure, or administering the Code of Hammurabi? A Texas defense lawyer in a similar case posed the question starkly in a brief published on the internet: Have we gone mad? After taking testimony over days of hearings and reviewing numerous written submissions, the Court has concluded that U.S.S.G. § 2G2.2, fails to provide a just and reasoned sentencing range given the facts of this case and the background of the defendant. As a consequence the Court has significantly varied downward in sentencing David Grober. This opinion records and supplements the reasons placed on the record on his sentencing day. For ease of reference, the opinion is divided into sections and headings as follows: 385 SECTION ONE: DOES § 2G2.2 PROVIDE A REASONABLE SENTENCE IN THIS DOWNLOADING CASE?........................................... i. Unpacking the Guidelines................................................385 ii. The Proofs.............................................................386 iii. Events Leading to Plea..................................................387 iv. Evidentiary Hearings ...................................................389 v. Testimony of Prof. Douglas Berman and Research of Troy Stabenow..........390 vi. How Courts Have Sentenced in the Absence of § 2G2.2 So That the Punishment Fits the Crime ............................................394 SECTION TWO: WHY THE APPLICATION OF § 2G2.2 FAILS TO PROVIDE A REASONABLE SENTENCE FOR THIS DEFENDANT .......................397 i. The § 2G2.2 Enhancements Apply Just About All the Time and Operate Exponentially........................................................397 ii. The § 2G2.2 Enhancements Are Promoting Sentencing Disparity.............397 iii. There is a Paucity of Direct Judicial Experience to Use in Fashioning Fair Sentences............................................................400 SECTION THREE: BECAUSE OF THE FLAWS IN § 2G2.2, THE COURT CANNOT APPLY THE GUIDELINE RANGE THE GOVERNMENT REQUESTS; RATHER THE COURT MUST BE GUIDED BY THE 3553(a) FACTORS AND THE MANDATORY MINIMUM SET BY CONGRESS............402 i. Unpacking the Sentencing Factors in 18 U.S.C. § 3553(a)....................402 ii. Applying the § 3553(a) Sentencing Factors.................................404 iii. The § 3553(a) factors “over-arching” instruction: the parsimony clause........411 CONCLUSION.................................................................412 SECTION ONE: DOES § 2G2.2 PROVIDE A REASONABLE SENTENCE IN THIS DOWNLOADING CASE? Unpacking the Guidelines U.S.S.G. § 2G2.2 applies to offenses that involve possession and transportation of child pornography. It has a long title and covers a variety of offenses, all relating to pornographic material. § 2G2.2. Trafficking in Material Involving the Sexual Exploitation of a Minor; Receiving, Transporting, Shipping, or Advertising Material Involving the Sexual Exploitation of a Minor; Possessing Material Involving the Sexual Exploitation of a Minor with Intent to Traffic; Possessing Material Involving the Sexual Exploitation of a Minor Under this guideline, the base offense level is 22 (§ 2G2.2 (a)(2)), calling for a sentencing range for a criminal category I offender of 41 to 51 months. Under § 2G2.2 (b), which lists specific offense characteristics, if the material involves prepubescent minors or minors under age 12, add 2; if the material portrays sadistic and masochistic conduct or other depictions of violence, add 4; if the defendant possesses a quantity of child pornography in excess of 600 images, add 5; if the defendant used a computer and an interactive computer service in acquiring the child pornography, add 2; and if the defendant distributed images of child pornography “for the receipt of a thing of value, namely additional images or videos of child pornography,” add 5. The adjusted offense level becomes 40, calling for a sentencing range of 292 to 365 months. This is almost 25 to over 30 years. The Proofs At his plea hearing, David Grober admitted the following facts through negotiated questioning: he sent someone an e-mail message on his computer that contained a video of child pornography on July 9, 2005, using his AOL account; on his computer, using the same AOL account, on July 27, 2005, he received an e-mail message from someone else that contained an image of child pornography and minutes later he sent back an e-mail message that contained approximately 17 images of child pornography; on his computer, using the same AOL account, on August 16, 2005, he received two separate e-mails from yet another sender that contained an image of child pornography; during the month of December, 2005, he collected images and videos containing child pornography from the internet that he stored and possessed on computer hard drives and portable compact discs; in the child pornography above there were numerous individuals who were clearly minors, some of whom were under the age of 12, and that on the images and videos he received, exchanged, possessed and stored, these individuals were engaging in sexual conduct with other minors or adults and/or posing in a sexually explicit manner. The government presented testimony of a forensic specialist, Special Agent Michell Chase (“SA Chase”) of Immigration and Customs Enforcement (“ICE”), who analyzed the content of David Grober’s seized computer hard drives and compact discs. She stated that Grober possessed a large collection of pornography, most of it adult porn and therefore legal. In the mix were more than 1500 images and over 200 videos of child pornography. SA Chase testified that she examined all the images and videos of child pornography, and classified them as to content. On two hard drives, she found 12 e-mails sent to and received by Grober that had images of pornography attached, among them child pornography. In support of the sentencing enhancements, the government prepared a CD for the court’s review of 14 images and 10 videos chosen by SA Chase that according to her testimony, “applied to the charges at hand as far as sadistic or bondage or prepubeseent.” (SA Chase Test., July 31, 2008, 98.) In open court, using a bench laptop computer, the Court personally reviewed the disc. The selected images the Court reviewed are of prepubeseent minors engaged in sexual activity, including images of posed bondage and images of what appeared to be actual penetration, which is considered a depiction of violence. As support for the distribution activity, the documented e-mails that David Grober sent and received establish the use of a computer or an interactive computer service for the possession, transmission, receipt or distribution of child pornography. Quickly, all of the special offense characteristics in § 2G2.2(b)(l) through (7) were established with one exception. Section 2G2.2(b)(5) increases the base offense level by 5 levels if the defendant “engaged in a pattern of activity involving the sexual abuse or exploitation of a minor.” There was no evidence or charge that David Gro-ber ever committed a contact offense against a child. The government has never wavered from this position. Anticipating it had evidence to support the specific offense characteristics, at the plea hearing government counsel stated she might be seeking a sentence as high as a level 40, and David Grober acknowledged the government might be making an argument for a sentence around the 30-year mark. He also acknowledged that he was pleading guilty to offenses that carry a five-year mandatory minimum sentence. Events Leading to Plea At 7:00 a.m. on December 7, 2005, officers from the Passaic County Sheriffs Office executed a search warrant at David Grober’s home. They had been alerted by a tip from AOL that child pornography was attached to e-mail exchanges involving David Grober’s account. An electronic warrant was obtained and ultimately authorization was granted for the search. Passaic County law enforcement showed up unannounced and seized computer equipment. Grober, his wife and 11 year old daughter were separately questioned. Their son, then 14, had already left for school. According to letters from members of the Grober family to the Court, written as part of the defense presentation at sentencing, the local law enforcement officers who executed the search told David Gro-ber and his wife that someone in the home was involved with child pornography and they did not know if it was David Grober or his teenage son. The family was told that the police might have to go to the boy’s school and arrest him in class. The Grobers’ son wrote in a letter to the Court submitted in connection with the sentencing that David Grober “immediately chose to sacrifice himself. Given the initial confusion as to who was actually responsible, the police threatened to pull me from my classroom in handcuffs, and arrest me at [the high school] in front of my peers and mentors, who held me in high esteem. Dad wouldn’t hear of it: he threw himself on the chopping block to spare me from harm.” Eleanor Grober wrote that after the officers separated the family and questioned each one separately, including their 11-year old daughter, she got permission to drive the girl to school as long as she did not stop at the high school to see their son. When she returned she found David Gro-ber in handcuffs. A female officer told Grober in front of his wife to confess that he had been cheating on her, because looking at the photos was a form of cheating. Eleanor Grober’s letter goes on: They then searched the rest of our home and took shopping bags full of videos of our children performing in shows, recorded episodes of favorite tv shows as well as David’s computers. As they left with their vehicle full of our personal mementos and my husband[,] I was in shock, I didn’t even know who to call. I remembered that an employee of ours was using an attorney and as they pulled away to book David [I] asked if I should call him. The only time we’ve needed an attorney was for the closing on our home. David Grober was taken to the Passaic County Jail. At the end of the day he was released. Because his young daughter lived in the home, Grober was not permitted to return pending an investigation by the New Jersey Division of Youth and Family Services (“DYFS”). He lived in a motel for two weeks until DYFS closed its file, finding no evidence of abuse, and his bail conditions were modified to permit him to return home. Passaic County turned over the seized computer hard drives and discs to federal authorities, who sent them for analysis to ICE. SA Chase was the agent who received and analyzed the content. At some point she went to the Grober home to question Mrs. Grober because SA Chase thought that the Grobers’ daughter was pictured among the child pornography images. Eleanor Grober examined what SA Chase showed her and denied the young girl depicted was her daughter. In her testimony, SA Chase did not disavow her impression that the photography had images of the Grobers’ daughter. “Looking at the image there were just several similarities and I still see the similarities.” (SA Chase Test., July 31, 2008, 87:15-16.) The government, to its credit, was clear that it did not consider that the child pornography images featured the Grobers’ daughter or that David Grober knew any of the children pictured. A two-count indictment was handed down on October 26, 2006. The Court learned that during the pre-indictment period, the government had offered a plea to the possession count and dismissal of the more serious receipt and transportation count with its five year, mandatory minimum sentence. Post-indictment, the government continued the same offer. In August, 2007, the Court heard oral argument on defense motions, which were denied, and the Court set down a trial date for October 9, 2007 on the two counts in the indictment, one for possession and the other for receipt and transportation. Two weeks before the trial date, the government obtained a superseding indictment that charged five specific counts of receipt and distribution of child pornography, each carrying a five-year mandatory minimum, and a sixth count of possession of child pornography. On October 4, 2007, the date set for his arraignment on the superseding indictment, David Grober entered a guilty plea to all counts without entering into any stipulations regarding sentencing. Evidentiary Hearings The sentencing date was postponed from January to the end of March, 2008. The government filed a brief dated March 11, 2008, setting forth its sentencing position that David Grober be sentenced to (1) A term of imprisonment within the range of 235 to 293 months on Counts One through Five, (2) a term of 120 months’ imprisonment on Count Six to run concurrently, and (3) a term of at least 2 years of supervised release. Further, the Court should order David Grober to pay a fine within the advisory Guideline range of $25,000 to $250,000. The defense moved for an evidentiary hearing to present testimony of mental health professionals on behalf of David Grober. The defense also filed a motion for a downward departure under the guidelines, and for a downward variance from the guidelines based upon the Court’s sentencing authority under 18 U.S.C. § 3553(a). In May, 2008, the Court held a status conference. Because of the extraordinarily harsh sentence called for under the guidelines, the Court requested the government to produce a witness from, or involved with, the United States Sentencing Commission. The Court sought the data behind the guidelines calculation to assist it in understanding how a first offender who pleaded guilty was facing a sentence at, and above, the 20 year statutory maximum. The government responded in a letter of May 22 [D.E. 50] indicating it would not produce a witness on the subject of the guidelines computation. The government reasoned that: The Supreme Court continues to recognize that the Guidelines-which reflect the Commission’s work “over a long period of time” as well as the Commission’s ongoing work-“insofar as practicable, reflect a rough approximation of sentences that might achieve § 3553(a)’s objectives.” Rita v. United States, [551 U.S. 338], 127 S.Ct. 2456, 2464-65 [168 L.Ed.2d 203] (2007). The Third Circuit has similarly recognized that in addition to being based on the in-depth research of the Commission, the Guidelines “reflect Congress’s determination of potential punishments, as set forth in statutes, and Congress’s ongoing approval of Guidelines sentencing, through oversight of the Guidelines revision process” so that the Guidelines “deserve careful consideration in each case.” United States v. Goff, 501 F.3d 250, 257 (3d Cir.2007). According to the government, nothing more could be added to “what has been set forth by the Commission in Appendix B of the Guidelines and what is contained in the public hearing testimony and transcripts of the Commission, which are available at http://www.ussc.gov/HEARINGS.HTM.” The Sentencing Guidelines Commission wrote the Court a week later [D.E. 51] and declined to participate, because “[although the Commission does not have an official policy regarding appearances at sentencing proceedings, the Commission agrees with the United States that a representative would not be able to provide the Court with information beyond what is contained in the legislative history of the Guidelines.” The defense moved to provide the information the Court sought by calling Professor Douglas Berman, a professor of law at the Moritz College of Law at The Ohio State University. Professor Berman is widely published and is co-managing editor of two journals, the Ohio State Journal of Criminal Law and the Federal Sentencing Reporter. He owns and edits a scholarly blog entitled Sentencing Low and Policy. See Douglas A. Berman, Sentencing Law & Policy, available at http://sentencing. typepad.com/ (last visited Dec. 10, 2008). Over the government’s objection, Professor Berman testified. Thus, the hearings that followed the May status conference involved the government’s case-in-chief providing proofs in support of the specific offense characteristics found in the PSR; Professor Berman’s testimony, which the government vigorously opposed; and testimony from mental health experts. As part of its case-in-chief the government provided victim impact evidence by calling the mother of young boys sexually abused by a predator in Wisconsin. The predator had photographed his abuse of these boys and others, and the images became part of the stream of available internet child pornography. David Grober’s computer storage devices contained images of these boys, who are called “identified victims” by the National Council for Missing and Exploited Children (“NCMEC”). Through contacts between SA Chase and NCMEC, “identified victim” children or their families were contacted to provide impact statements if they wished. In addition to the testimony of this particular family member, the PSR included three written victim statements from another family regarding a separate incident of abuse. The hearings took place over several days, spaced out due to scheduling issues. In October, both sides submitted briefs about the rationality of the sentencing guidelines as applied in this case, and the Court took argument on this issue. The final hearing took place on the day the Court imposed sentence, when the Court heard from David Grober and his family, his Rabbi, neighbors, and friends and took argument from the government and the defense on the sentence to be imposed. To the end, the government argued for a sentence of almost 20 to almost 25 years. Preserving their arguments, defense counsel did not take a guidelines position and argued strenuously for a sentence below level 38. Defense counsel argued further that it was unconstitutional to sentence David Grober to the mandatory minimum. The Court rejected the constitutional argument because it did not find that the legal grounds set forth in the supporting brief [D.E. 89] were persuasive, and this point was made only at the conclusion of all the testimony and submission of final briefs. On the sentencing day, while not abandoning their arguments, David Grober’s lawyers urged the Court not to sentence above the mandatory minimum of 60 months/five years. Testimony of Professor Douglas Ber-man and Research of Troy Stabenow Moving to the substance of the foregoing hearings, one of the themes was the government’s massive resistance to hearing from Professor Douglas Berman. As seen above, from the first moment the Court and counsel addressed the offense computation in the PSR, the Court was actively seeking some rationale for the harsh sentence — close to and exceeding the statutory maximum of 20 years — that the government had adopted from the PSR as the appropriate sentence. First the government and then the Sentencing Commission declined to engage in any kind of dialogue. The Court denied the government’s motion to exclude and later to strike Professor Berman’s testimony, because the Court believes it is vital to gain some perspective on the operation of the sentencing guidelines in these downloading cases. It was through Professor Berman’s testimony that the Court initially gained an understanding of the role Congress has played in the increased penalties for child exploitation offenses and how intertwined the guidelines for child exploitation offenses have become with various pieces of legislation. Professor Berman, in addition to his own immersion in federal sentencing guidelines, has published an increasingly influential article written by a federal public defender, Troy Stabenow, on his sentencing blog. See Troy Stabenow, Deconstructing the Myth of Careful Study: A Primer on the Flawed Progression of the Child Pornography Guidelines, available at http://mow.fd.Org/3 July 2008 Editpdf (last visited Dec. 10, 2008). At the time Professor Berman testified, Stabenow’s research was beginning to be cited in federal sentencing decisions. In cross-examining Professor Berman and in its main sentencing brief [D.E. 76], the government downplayed Stabenow as a biased “deconstructionist,” pointing out that he is a federal public defender and has an agenda that is apparent from the title of his article. Professor Berman was able to support Stabenow’s contributions and research. In commenting on Stabe-now’s article, Professor Berman confirmed the reliability of Stabenow’s observation that a third of the sentences imposed in these cases vary below the guidelines range. He testified further that judicial attentiveness toward the disturbing effect of the guidelines on child pornography offenses outstrips Commission attentiveness to it. This Court’s own experience mirrors Professor Berman’s observation that both in written opinions and anecdotally, sentencing judges are concerned about these guidelines because they are uncomfortable following a guideline table that has moved up so high on the statutory range. The Court emphasizes here that insofar as the government might have established a dialogue between Professor Berman and the Commission on the Stabenow research or any other issue Professor Berman testified about, it took a pass. So did the Commission. In July, at about the time Professor Berman was testifying, general counsel to the Commission wrote the Court, again declining to participate in the sentencing hearing. See Appendix A (Letter of the U.S. Sentencing Comm’n to the Court, dated July 15, 2008). Hence, despite the Court’s invitation and many opportunities, the testimony of Professor Berman is unrebutted by live testimony. The Stabenow research was not the only feature of Professor Berman’s testimony. He also stressed the critical role that charging decisions by federal prosecutors play in the reported statistics: “in many of these cases there may be plea agreements or other arrangements that are made, so that neither the presentence report, nor the judge, finds out the full scope of potentially applicable enhancements.” This occurs through stipulations that the judge accepts, or through “a plea that ends up limiting the applicable guideline range, the applicable statutory maximum, and I’m amazed, I guess this is the point I’ve actually focused on a little bit more, since looking at this case, how many cases seem like trading, distribution, but they get pled out to just pure possession.” (Berman Test., July 16, 2008, 88-89.) Professor Berman’s testimony and Troy Stabenow’s article are not lone voices. Published decisions in 2008 reflect judicial concern about sentences in downloading cases. Chief Judge Bataillon of the District of Nebraska concluded in United States v. Baird, 580 F.Supp.2d 889, 893-94 (D.Neb.2008) that based on the history of the child pornography guidelines as the Commission has keyed them into policy and statutory mandates, the sentencing ranges of imprisonment are “a less reliable appraisal of a fair sentence.” This point was made again by Chief Judge Pratt in United States v. Shipley, 560 F.Supp.2d 739, 744 (S.D.Iowa 2008): [T]he guidelines for child exploitation offenses were not developed using an empirical approach by the Sentencing Commission, but rather were mainly promulgated in response to statutory directives. Specifically, the Protect Act directly amended guideline 2G2.2 by amending the guideline enhancements for specific offense characteristics. These modifications do not appear to be based on any sort of empirical data, and the Court has been unable to locate any particular rationale for them beyond the general revulsion that is associated with child exploitation-related offenses. The day after Shipley was filed, Judge Lynn Adelman filed a sentencing memorandum in United States v. Hanson, 561 F.Supp.2d 1004 (E.D.Wis.2008). He liberally cites to Stabenow’s article, finding that the article’s observations and conclusions “hit home” in the case before him, where an offender “who appeared fairly typical, one with no prior record, found himself in a guideline range that exceeded the statutory maximum at the high end.” It is appropriate here to cite the same long passage in the Stabenow article that convinced Judge Adelman that “given all of the flaws” in 2G2.2, the sentencing range does not deserve deference. The flaw with U.S.S.G. § 2G2.2 today is that the average defendant charts at the statutory maximum, regardless of acceptance of responsibility for Criminal History. As noted by the Guidelines Commission, there are “several specific offense characteristics which are expected to apply in almost every case (e.g. the use of a computer, material involving children under the 12 years of age, number of images).” See Amendment 664, U.S.S.GApp. C “Reason for Amendment”, (November 1, 2004). The internet provides the typical means of obtaining child pornography resulting in a two-level enhancement. See U.S.S.G. § 2G2.2(b)(6). Furthermore, as a result of internet swapping, defendants readily obtain 600 images with minimal effort, resulting in a five-level increase. See U.S.S.G. § 2G2.2(b)(7)(D). The 2004 Guidelines created an Application Note defining any video-clip as creating 75 images. See U.S.S.G. § 2G2.2(b)(2),(4). Thus one email containing eight, three-second video clips would also trigger a five-level increase. Undoubtedly, as the Commission recognized, some of these images will contain material involving a prepubescent minor and/or material involving depictions of violence (which may not include “violence” per se, but simply consist of the prepubescent minor engaged in a sex act), thereby requiring an additional six-level increase. See U.S.S.G. § 2G2.2(b)(2), (4). Finally, because defendants generally distribute pornography in order to receive pornography in return, most defendants receive a five-level enhancement for distribution of a thing of value. See U.S.S.G. § 2G2.2(b)(3)(B). Thus, an individual who swapped a single picture, and who was only engaged in viewing and receiving child pornography for a few hours, can quickly obtain an offense level of 40. Even after Acceptance of Responsibility, an individual with no prior criminal history can quickly reach a Guideline Range of 210-262 months, where the statutory maximum caps the sentence at 240 months. See U.S.S.G. § 5Gl.l(a). The results are illogical; Congress set the statutory range for first time distributors as five to twenty years. Congress could not have intended for the average first time offender with no prior criminal history to receive a sentence of 210 to 240 months. An individual with a Criminal History Category of II faces a Guideline range of 235 to 240 months, and any higher Criminal History score mandates the statutory maximum. These results run contrary not only to Congressional will, but also to a principal Guideline policy — providing harsher penalties to individuals with more significant Criminal History scores while still retaining an incentive for pleas at all Criminal History levels. Hanson, 561 F.Supp.2d at 1010-11 (quoting Troy Stabenow, Deconstructing the Myth of Careful Study: A Primer on the Flawed Progression of the Child Pornography Guidelines (July 3, 2008) 23-24.) And Chief Judge Pratt has provided another thoughtful analysis of these sentencing guidelines in United States v. Johnson, 588 F.Supp.2d 997, 1003-04 & n. 4 (N.D.Iowa 2008): At the urging of Congress, the Sentencing Commission has amended the guidelines under § 2G2.2 on several occasions over the past two decades, recommending more severe penalties. As far as this Court can tell, these modifications do not appear to be based on any sort of empirical data, and the Court has been unable to locate any particular rationale for them beyond the general revulsion that is associated with child exploitation-related offenses. See Skye Phillips, Protect Downward Departures: Congress and Executive’s Intrusion into Judicial Independence, 12 J.L. & Pol’y 947, 967-84 (2004) (discussing the history of the guideline ❖ * # [Footnote 4] See, e.g., Amendment 372, U.S.S.G.App. C (1991) (responding to the Treasury, Postal Service and General Government Appropriations Act, Pub. L. No. 102-141, § 632, 105 Stat. 834 (1991)); Amendment 537, U.S.S.G.App. C (1996) (responding to the Sex Crimes Against Children Prevention Act of 1995, Pub. L. No. 104-71, §§ 2-4, 6, 109 Stat. 774 (1995)); Amendment 592, U.S.S.G.App.C. (2000) (responding to the Protection of Children From Sexual Predators Act of 1998, Pub. L. No. 105-314, §§ 506-507, 112 Stat. 2974 (1998)); and Amendments 649, U.S.S.G.App.C. (2003) and 664, U.S.S.G.App.C (2004) (responding to the Prosecutorial Remedies and Other Tools to end the Exploitation of Children Today (PROTECT) Act of 2003, Pub. L. No. 108 P. L. 21, § 401, 117 Stat. 650 (2003)). For a thoughtful commentary on the progression of the sentencing guidelines for child pornography offenses, see Troy Stabenow, Deconstructing the Myth of Careful Study: A Primer on the Flawed Progression of the Child Pornography Guidelines, July 3, 2008, available at http://mow.fd.Org/3 July 2008 Editpdf (unpublished Comment, last visited Dec. 3, 2008). modifications made in the PROTECT Act and the fact that the changes were made without notifying or consulting the Sentencing Commission); see also United States v. Detwiler, 338 F.Supp.2d 1166, 1170-71 (D.Or.2004) • (discussing the Feeney Amendment to the PROTECT Act). Thus, because the guidelines at issue in this case do not reflect the unique institutional strengths of the Sentencing Commission in that they are not based on study, empirical research, and data, this Court, as it did in United States v. Shipley, “affords them less deference than it would to empirically-grounded guidelines.” 560 F.Supp.2d 739, 745-46 (S.D.Iowa 2008); see also Kimbrough, 128 S.Ct. at 574 (noting that guideline ranges created when the Commission departed from past practices are a less reliable appraisal of a fair sentence). Congress has created a fifteen-year window, between the statutory minimum (5 years) and maximum (20 years) sentences, within which this Court can penalize a convicted child pornographer. See 18 U.S.C. § 2252(a)(2), (b)(1). However, on account of Congress’ tinkering with the guidelines, the Commission now recommends that nearly all defendants be incarcerated near the twenty-year statutory maximum. Thus, strict adherence to the sentencing guidelines effective at the time of Defendant’s arrest, and even more so to those effective today, would make it difficult for the Court to consider the individualized factors that § 3553(a) requires. Stated differently, the Court would struggle to differentiate between the punishment appropriate for the most and the least egregious acts of child pornographers. As this Court noted in Shipley, the Court must consider the need to avoid unwarranted similarities in the punishment handed down to differently situated defendants. 560 F.Supp.2d at 745-46. The statute provides a broad range of punishments for the crime Defendant committed. If Congress does not want the courts to sentence individual defendants throughout that range based on the facts and circumstances of each case, then Congress should amend the sentencing statute, rather than manipulate the advisory guidelines, thereby blunting the effectiveness and reliability of the work of the Sentencing Commission. While the Court has consulted the advice of the guidelines, the Court finds this advice is less reliable in the present case than in others where the guidelines are based on study and empirical data. The foregoing decisions are practical, courageous, and use the research in Troy Stabenow’s article. Recently Stabenow published a chart on the web that summarizes the march of the guidelines to the top and beyond the statutory maximum for internet trading offenses, which is attached as Appendix B. Troy Stabenow, “Table of Changes” Chart Accompanying Deconstructing the Myth of Careful Study: A Primer on the Flawed Progression of the Child Pornography Guidelines, available at http://mow.fd.org/Table of Changes.pdf (last visited Dec. 18, 2008). The Stabenow chart is a one-page devastating commentary on guidelines sentencing in downloading eases. How Courts Have Sentenced In the Absence of § 2G2.2 So that the Punishment Fits the Crime The Court agrees with the reasoning of Judge Pratt and Judge Adelman that § 2G2.2 leads to a sentence that is too severe in a downloading case. Surely congress did not intend to provide a sentencing range of 19% to 20 years for a typical downloader, especially one who pleads guilty. Aside from overly punitive sanctions for a first offender, the guidelines sentence for a typical downloader squelches the ability of a judge to depart upward. The bottom line is that the skewed focus in § 2G2.2 on content and the strained application to conduct, see infra, lead to a vision whereby the predator becomes indistinguishable from the voyeur. This wasn’t the way it used to be. At the beginning of the offensive against internet pornography, courts used the guidelines creatively to capture the true nature of the offense, and used cross-references where there were genuinely aggravating circumstances. United States v. Alter, 985 F.2d 105, 106 (2d Cir.1993), involved the criminal conduct of a co-owner of a private halfway house where approximately 100 federal inmates resided. Defendant Alter was indicted in 1990 for coercing five Manhattan House residents to engage in sexual acts with him. Id. Faced with a 20-count indictment, Alter pleaded to a single bribery count for demanding and receiving sexual and sadomasochistic favors in exchange for payment, drugs, and assurances of favorable treatment at the halfway house. Pursuant to the plea agreement, for sentencing purposes the Government was entitled to seek an upward departure by proving details of Alter’s sexual advances on others in his custody. Id. The district court found the base level of 10 for the crime of conviction insufficient because of “three aggravating circumstances” calling for “a substantial upward departure,” specifically that Alter abused the warder/inmate relationship for sexual gain, that the episode greatly disrupted the halfway house and the federal corrections system, and that Alter facilitated one of the inmate’s drug abuse. Alter, 985 F.2d at 107. Because the guidelines fell short in punishing these aspects of Alter’s offenses, the sentencing judge resourcefully matched Alter’s bad acts with existing punishments for stand-alone crimes. Id. For his sexual bartering with the inmates, the sentencing judge drew a parallel to the crime of sexual intercourse with a person under one’s custody, which carries a base offense level of 9. Id. Thus, Court added 9 levels for the inmate with whom Alter had sex, and another 2 levels for another inmate on whom advances were made but with whom no intercourse took place. Id. For disrupting the corrections system, the judge fashioned an enhancement of three levels, even in the absence of an analogous “guidepost” offense. Id. For having given drugs to the inmates, the court added three levels based upon similar federal crimes. Id. Overall, the Alter court added 17 levels, and subtracted only two levels for acceptance of responsibility. In a more recent case, the appellate court affirmed a district court’s application of a cross-reference to another guideline to enhance a sentence where the defendant’s conduct was particularly egregious. In United States v. Madison, 477 F.3d 1312, 1313-14 (11th Cir.2007), the defendant, Madison, pled guilty to charges of sex trafficking of children by fraud, force, or coercion (18 U.S.C. § 1591(a)(2)), and conspiracy to transport a minor for purposes of prostitution. (18 U.S.C. § 2423(e)). Madison’s criminal conduct, however, which involved closely managing the prostitution of a 16-year old and physically abusing her, caused the district court to utilize a sentencing cross-reference to U.S.S.G. § 2A3.1, relating to criminal sexual abuse. Id. at 1314. The district court’s 168-month sentence, based upon Madison having forced the 16-year old into “commercial sex” by way of violence, was upheld by the Eleventh Circuit, which found Madison’s conduct to fit within the realm of criminal sexual abuse under 18 U.S.C. § 2241. Id. See also United States v. Froman, 355 F.3d 882 (5th Cir.2004) (affirming a sentence where the sentence departed upward under § 5K2.8 for extreme conduct, based on the defendant’s having produced a video tape that showed him having intercourse with his 12 year old daughter. The defendant was charged with possession and receipt of child pornography and conspiracy to knowingly transport, receive, and distribute child pornography). This year, the Fifth Circuit affirmed a sentence that departed from the guidelines range applicable to a possession count, 63-78 months, and applied the statutory maximum of 10 years based on relevant conduct. In that case, United States v. McGehee, 261 Fed.Appx. 771, 772-73 (5th Cir.2008), the district court departed upward on facts that defendant McGehee did not dispute, and on sentencing guidelines commentary that “upward departure may be warranted if the ... enhancement does not adequately reflect the seriousness of the sexual offense or exploitation involved.” McGehee admittedly “made use of religious and social service institutions and programs to gain access to children for his sexual gratification,” and in so doing “molested seven children between the ages of six and nine on more than 100 occasions.” Id. at 773. Further, the district court discounted McGehee’s degree of acceptance of responsibility because he lightly referred to his crimes as “unfortunate incidents” and “errors” in letters to the court. Id. The Fifth Circuit weighed the “ “totality” of the circumstances” and affirmed the upward departure to the statutory maximum. Id. In the 2006 case United States v. McCaffrey, 437 F.3d 684, 686 (7th Cir.2006), the Seventh Circuit affirmed a sentence at the top of the statutory range after a plea to one count of receipt and possession. Vincent McCaffrey, a former priest, was sentenced to 240 months after the sentencing judge heard evidence from former victims and enhanced 5 levels upward for a pattern of sexual abuse and ordered a 5-level increase in McCaffrey’s criminal history. Id. Since his 1978 ordination, McCaffrey capitalized on priestly assignments to manage youth groups, choirs, and altar boys in various parishes in order to sexually molest boys in his supervision “hundreds” of times over. Id. His misdeeds persisted to the computer age, and in 1999 McCaffrey purchased memberships to child pornography websites. Id. It was not until 2002 that federal customs agents arrested McCaffrey, finding images on his computer containing children under 12 and depicting sadomasochistic situations, which formed the basis for his pled-to charges. Id. The affir-mance noted that the guidelines intend “to enable district judges to give extended sentences to those with a long history of abusing children.” Id. at 688. The circuit court concluded that the sentencing judge wanted to “put McCaffrey away for as long as he could.” There was plenty of evidence to support that intent and the guidelines were applied flexibly so that this sentence fit the many crimes committed that had never been punished. The last case discussed above, McCaf-frey, saw the results of proper exercise of judicial discretion where the defendant who possessed and consumed child pornography was punished for his predatory conduct. The district court departed upward, using discretion and common sense, based on evidence about the conduct of the offender. McCaffrey did terrible things to actual children aside from involving himself in images of child pornography available on the internet to predators like himself, as well as to voyeurs like David Grober. The 240-month sentence that McCaffrey received satisfies the goals of just punishment, protection of the public (long left unprotected, it appears from the facts of his case), and deterrence. In David Grober’s case, the government wants a higher sentence for a man who is not accused of, and has never been found to be, an active abuser. In so doing, the government unimaginatively, robotically, and with terrible consequences to this defendant, seeks application of “aggravating” factors that in reality define the core of the offense — transmitting child pornography by file sharing and, on occasion, attaching it to e-mails. That is what a consumer does, not what a predator does. Under § 2G2.2 as applied in the PSR, the consumer is punished as fiercely and even more fiercely than the predators described in the foregoing cases, leading naturally to the next discussion. SECTION TWO: WHY § 2G2.2 DOES NOT ACCOMPLISH THE SENTENCING GOALS OF § 3553(a) FOR THIS DEFENDANT To review, then, the Court has learned that actual working judges have decided to give these child pornography guidelines less weight in downloading cases; an experienced law professor with a widely used sentencing blog has been unable to probe the Commission’s thinking why it sentences downloaders this way; and an influential article makes persuasive arguments against the rationality of § 2G2.2 in downloading cases. Also, reported cases demonstrate that bad people involved with child pornography can get long sentences without using § 2G2.2 as the basis. There are a few more practical concerns to add. The § 2G2.2 Enhancements Apply Just About All the Time and Operate Exponentially Stabenow and Professor Berman commented on the onerous and universal application of the offense characteristics called for under § 2G2.2. Essentially, they point out that using these characteristics to enhance the base offense level is irrational because logically and factually, the characteristics are simply not genuine aggravating factors. Rather, they are inherent in just about any downloading offense. As persuasive as Stabenow and Professor Berman were on this point, ultimately the most persuasive evidence came from the government’s own witness. SA Chase is the veteran of 100,000 images from 180 collections. In her testimony, SA Chase recognized that every one of her 180 investigations involved a possessor with 600 or more images. (SA Chase Test., Dec. 1, 2008, 81:21-25.) SA Chase testified that every one of the cases she had worked on — -“100 percent” — “involved the use of a computer and of interactive computer service.” (SA Chase Test., Dec. 1, 2008, 80:21-25.) Further, according to SA Chase, “all” of the cases she has worked on involved images of prepubescent minors under age 12, either posing or engaged in sexual activity. (SA Chase Test., Dec. 1, 2008, 82:16-17.) Even a vast majority— “80 percent” — had at least one image and video depicting sadomasochistic content. (SA Chase Test., Dec. 1, 2008, 82:23-24.) SA Chase’s experience on the ground punctuates Stabenow’s and Berman’s view that the purportedly aggravating factors are actually inherent in the possession of child pornography. Another problem arises from the wholesale application of the enhancements: since the increase in the base offense level from 17 to 22, the exponential rise in sentence exposure when the special offense characteristics are factored in produces an outrageously high sentence. This is forced by way the numbers work in the sentencing table. When the base offense level for receipt/distribution/trading/traffieking was 17, moving up 18 levels would add 12 to 15 years to the base offense exposure of 2 to 2^ years and the range stayed below the statutory maximum. But when the base offense level in § 2G2.2 was increased by 5 levels up to level 22, adding another 18 levels piled up the exposure much higher than an extra 12 to 15 years, because the levels jump higher the more serious the punishment becomes. Piling 18 levels on top of level 22 increased the base offense level by 21 to 26 years instead of 12 to 15 years. The Commission fashioned § 2G2.2 so that this draconian impact occurs, seemingly without discussion. The § 2G2.2 Enhancements Are Promoting Sentencing Disparity As the Supreme Court wrote in Kim-brough v. United States, — U.S. -, 128 S.Ct. 558, 169 L.Ed.2d 481 (2007), it is unquestioned that uniformity remains an important goal of sentencing. As we explained in Booker, however, advisory Guidelines combined with appellate review for reasonableness and ongoing revision of the Guidelines in response to sentencing practices will help to “avoid excessive sentencing disparities.” In the last decade and a half, congress has increased penalties for child sexual exploitation over and over, and law enforcement has increasingly focused on criminal prosecutions. The Office of Justice Programs in the Department of Justice wrote in a December 2007 bulletin entitled “Federal Prosecution of Child Sex Exploitation Offenders, 2006” that sex offenses are among the fastest growing crimes handled by the federal justice system, and that child pornography matters accounted for 82% of the growth in sex exploitation matters referred from 1994 to 2006. The December 2007 bulletin reports that the conviction rate is high for sex offenders generally. In 2006, 95% of child pornography defendants were convicted, 92% of them through a plea. Of those who were not convicted, 9 out of 10 had then-cases dismissed. The vast majority— 96% — of convicted sex offenders are sentenced to prison. Child pornography defendants represented the largest increase in persons receiving a prison sentence. In 1996, only 77% of them received a prison sentence, compared to 97% in 2006. The median prison term for federal sex exploitation offenses more than doubled from 1996 to 2006, with child pornography sentences increasing faster than that: they quadrupled from 15 months in 1996 to 63 months in 2006. From these statistics, particularly the mean sentence of 63 months, it would appear that courts are following the avowed intent of congress in punishing offenders with significant prison terms, but nowhere close to what the enhancements in § 2G2.2 produce. Instead, federal prosecutors cut deals with defendants such as the one originally offered to David Grober. Behind the scenes of such a plea, had it gone through in this case, lie the untold stories of all the many images of child pornography and all the victims. But as Professor Berman testified, the judge does not learn about them and the sentencing statistics do not reflect them. Assuming this more benign means of punishing the downloader is the norm, it reflects a choice to punish way below the guidelines mandate if the defendant pleads guilty. A system like that imposes a very severe trial penalty. Finally, as the recent cases show, judges are varying from the guidelines in typical downloading cases when they measure the guidelines against the § 3553(a) factors. Significantly, these judges are not varying because a given case is atypical. The Commission was prescient when, in 1996 (as sentences began ramping up), it wrote as follows in its report, Sex Offenses Against Children: Findings and Recommendations Regarding Federal Penalties (June 1996): The amount of prison time that is appropriate for these offenses is a policy judgment to be made by Congress and the Sentencing Commission. This judgment can be informed, however, by data on how the range of punishments available under the present guidelines are being used. Increasing offense levels far above that which is believed to be appropriate by practitioners in the courts could lead to improper charges or departures intended to result in lower sentences. Sex Offenses Against Children at 6 (emphasis added). This Court knows from its own experience about the problematic nature of charging discretion and plea arrangements — issues about which Professor Ber-man also testified. This year, the Court sentenced an offender under a plea agreement similar to the one offered to David Grober. The government and the defense used the sentencing guidelines in effect on November 2003, which meant that the applicable guideline was § 2G2.4, which, as Troy Stabenow has documented, was later collapsed into § 2G2.2. The defendant (“the earlier defendant” for purposes of this discussion) entered a guilty plea to possession of child pornography, which then carried a base offense level of 15. The plea stipulations were to the number of images (plus 2), the use of a computer (plus 2), and the pre-pubescent age of the children involved in the images (plus 2). The government dismissed the receipt/distribution count. The adjusted offense level became 21. The earlier defendant had challenged the search and seizure of the computer, the Court held a suppression hearing and ruled against him, and he pleaded guilty close to the trial. As such, he received a two rather than three level adjustment downward for acceptance of responsibility. The government did not supersede the original indictment. The parties stipulated that a reasonable sentencing range was at level 19. The Court sentenced the earlier defendant to 30 months, the bottom of the guideline range of 30 to 37 months for level 19. The earlier defendant had viewed child pornography on his computer between October 2002 and June 2004. David Grober’s active period according to the government was between October 2002 and his arrest date in December 2005. It is a chilling exercise to take the earlier defendant’s plea deal and apply it to David Grober’s case. His original indictment charged him with the same two offenses the government charged against the earlier defendant. But by December 2005, the changes in the guidelines created by direct congressional action and the Commission’s collapse of § 2G2.4 into § 2G2.2 were in place — the very changes that are discussed by Stabenow and Berman. These changes add five levels to how the guidelines operated under the terms of the earlier defendant’s plea deal: the possession base offense level rises from 15 to 18, and there are an extra three points added for number of images (assuming the stipulation was in excess of 600 images). So the earlier defendant’s deal in the age of Grober would be modified and the adjusted offense level for possession would be 26 instead of 21. The final adjusted offense level after deducting the two levels for pleading guilty would be 24, calling for a sentencing range of 51 to 63 months. (Interestingly, that range encompasses the mandatory minimum in the heavier count of the original indictment.) What happened in David Grober’s case vividly demonstrates the danger in pursuing trial instead of grabbing a plea offer pre-indictment or pre-motion practice. When plea negotiations apparently broke down and the trial date was imminent, the government abandoned its original and longstanding offer to a count of possession, superseded, and argued passionately for a sentencing range of between 235 and 292 months! This Court knows enough of the facts behind the earlier defendant’s offense to be unpersuaded that there is an intellectually honest basis to distinguish what the earlier defendant did from what David Grober did. What valid sentencing function transforms David Grober’s core conduct from warranting a sentencing range of 51 to 63 months to warranting a sentencing range of 292 months to 365 months merely because he did not plead guilty? In this way, and the others discussed above, the harshness of § 2G2.2 is fostering a capricious and dangerous environment. There Is a Paucity of Direct Judicial Experience to Use in Fashioning Fair Sentences The prior discussion focused on the twin problems of sentencing disparity and the trial penalty. There is another pernicious problem embedded in the prevalence of pleas as the vehicle for disposition of downloading offenses. The practice of avoiding the offense characteristics in § 2G2.2 (or having colossal warfare over them) breeds an artificial, unrealistic environment for sentencing; encourages a failure on judges’ part to make a valid contribution to the development of reasonable guidelines; and fosters a skewed empirical base. This became apparent when the government sought in this case to justify the sentence of more than 20 years by describing David Grober as a particularly egregious internet consumer. Using the testimony of SA Chase, whom it recalled specifically for this purpose, the government argued that the pictures and videos in his storage facilities were particularly bad ones and so the 20-year sentence was appropriate. A review of all of SA Chase’s testimony indicates that the government failed to make its point. Reviewing that testimony: SA Chase located child pornography on three of the four hard drives seized, and on three CDs. She recovered almost 1700 still images and 200 videos of child pornography, mixed in with family photos, nature shots, pictures of sports figures, and an unquantified but plentiful amount of adult pornography. In analyzing the media, SA Chase used a forensic software tool that allows her to run a slideshow and pinpoint child pornography. After locating it, she inspected the child pornography. She testified that she did not catalogue any images on the drives and discs that were not child pornography and was markedly reluctant to estimate the ratio of child pornography to adult pornography. Dr. Katz, the government’s mental health expert, reviewed the media and testified that there was considerably more adult than child pornography on the drives and discs. It is one of the disturbing aspects of the government’s proofs that without Dr. Katz’s testimony, the Court would never have learned the ratio of child pornography to legal pornography. SA Chase opined, over the objection of the defense, that the child pornography in Grober’s possession was exceptional in quantity and egregious in content as compared to other child pornography she has analyzed. As to quantity, she said this was the second largest collection of the 180 collections she has analyzed, and as to content, she testified that more children were involved in actual sexual activity, as opposed to posing, than appeared in other collections she analyzed. The Court has reviewed SA Chase’s earlier testimony in which she reported the specific breakdown of child pornography content. The Court is left with the impression from that testimony that the majority of the sizeable numbers of images of prepubescent children did not involve infants and depictions of sadistic and masochistic conduct, which does not quite square with her opinion testimony. The point needs to be made here that the adversary system suffers at the point where a judge is asked to determine whether the child pornography is “material that portrays sadistic or masochistic conduct or other depictions of violence,” which if found, dramatically increases the offense level by 4 levels. For the most part, the proofs of this specific offense characteristic come in by way of little descriptive paragraphs written by the PSR writer about particular images presumably selected by the government. Or a judge personally reviews the images, which from this Court’s anecdotal information, occurs rarely. And when the judge does review the images, the ones they see are chosen by the government. Things get more than one-sided, they get unreliable when, as here, the forensic agent seeks to evaluate one collection as compared to others. SA Agent Chase testified that in the majority of all the child pornography collections she has reviewed, there is at least one image and one video of sadistic and masochistic conduct. (SA Chase Test., Dec. 1, 2008, 82:20-24). To make the point that a collection is particularly repugnant — which is the point the government was pressing — there had to be more of these images than usual. But what is usual? How does defense counsel test the conclusion that what everyone knows is repugnant, is really really repugnant? SA Chase has reviewed 100,000 images in her 4 year career. If she is entitled to be an expert, what exactly is her expertise? No one else is in a position, or wants to be, of testing the soundness of her expert conclusion. And assuming someone wanted to gain the expertise to do that, it would be extremely diff