United States v. Maury

PRECEDENTIAL UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT Nos. 09-2305/09-2306/09-2345/09-2346/09-2356 UNITED STATES OF AMERICA v. JEFFREY MAURY, Appellant No. 09-2305 _____ UNITED STATES OF AMERICA v. CRAIG DAVIDSON, Appellant No. 09-2306 _____ UNITED STATES OF AMERICA v. JOHN PRISQUE, Appellant No. 09-2345 _____ UNITED STATES OF AMERICA v. SCOTT FAUBERT, Appellant No. 09-2346 ____ UNITED STATES OF AMERICA v. ATLANTIC STATES CAST IRON PIPE COMPANY, a Division of McWane, Inc. Appellant No. 09-2356 _______________ On Appeal from the United States District Court for the District of New Jersey (D.C. No. 3-03-cr-00852-001 through 004 and 006) District Judge: Hon. Mary L. Cooper _______________ Argued March 29, 2012 2 Before: FUENTES, SMITH, and JORDAN, Circuit Judges. (Filed: September 17, 2012) _______________ Michael N. Pedicini (ARGUED) 60 Washington Street Courthouse Plaza Morristown, NJ 07960 Attorney for Defendant-Appellant Jeffrey Maury Hilary L. Brunell (ARGUED) Vincent J. Nuzzi Nuzzi & Mason 50 Nelson Street Dover, NJ 07801 Attorneys for Defendant-Appellant Craig Davidson Michael D. Critchley (ARGUED) Critchley, Kinum & Vazquez 75 Livingston Ave 3rd Floor Roseland, NJ 07068 Attorney for Defendant-Appellant John Prisque Michael D’Alessio, Jr. (ARGUED) Walder, Hayden & Brogan 3 5 Becker Farm Road 3rd Floor Roseland, NJ 07068 Timothy I, Duffy Mark. K. Silver Michael J. Sullivan Coughlin Duffy 350 Mount Kemble Avenue P.O. Box 1917 Morristown, NJ 07962 Attorneys for Defendant-Appellant Scott Faubert John J. O’Reilly (ARGUED) Day Pitney One Jefferson Road Parsippany, NJ 07054 Attorney for Defendant-Appellant Atlantic States Cast Iron Pipe Co. Mark E. Coyne Office of the United States Attorney 970 Broad Street Room 700 Newark, NJ 07102 Glenn J. Moramarco (ARGUED) Office of United States Attorney Camden Federal Building & Courthouse 401 Market Street Camden, NJ 08101 4 John L. Smeltzer United States Department of Justice Environment & Natural Resources Division P.O. Box 7415 Washington, DC 20044 Attorneys for Plaintiff-Appellee the United States of America Table of Contents I. Facts & Posture………………………………………...8 A. Factual Background………………………………...8 1. The Defendants…………………………………8 2. The Plant………………………………………9 B. Clean Water Act Violations……………………….10 1. Unlawful Discharge of Wastewater…………...11 2. Resulting Oil Spills & Ensuing Investigations...15 C. Clean Air Act Violations………………………….19 1. The Plant’s CAA Permits……………………...20 2. Burning Excess Paint………………………….21 D. OSHA Incidents…………………………………...23 1. The Coxe Fatality……………………………...24 2. The Marchan Incident…………………………26 3. The Owens Incident…………………………...28 5 4. The Velarde Incident…………………………..29 E. Indictment…………………………………………31 F. Pre-Trial Issues……………………………………34 G. Trial & Sentencing………………………………...34 H. The Present Appeal & the Parties’ Arguments……37 II. Discovery Issues………………………………………39 A. Pre-Trial Discovery in a Criminal Case…………...39 1. Jencks Material………………………………...39 2. Fed. R. Crim. P. 16…………………………….40 3. Brady and Giglio Material…………………….43 B. Pre-Trial Discovery in this Case…………………..43 C. The Scope of Fed. R. Crim. P. 16(a)(1)(C)……….46 1. Applicable Standard of Review……………….47 2. Rule 16(a)(1)(C)(ii)……………………………50 III. Jury Instructions on the Clean Water Act Violations.......................................................................55 A. Negligence Instruction Under the Clean Water Act............................................................................55 1. Invited Error Doctrine…………………………61 B. The District Court’s Refusal to Define “Recklessness”…………………………………….68 1. Accuracy of the Court’s Instructions on “Knowing” Conduct………………………………70 2. Abuse of Discretion in Rejecting Proposed Language…………………………………………..73 6 IV. Mutually Exclusive Verdicts…………………………76 V. Conclusion…………………………………………….82 FUENTES, Circuit Judge: Following an eight-month criminal trial, a jury convicted Atlantic States Cast Iron Pipe Company and four of its managers of various crimes. These included conspiring to commit a host of environmental pollution and worker safety violations, attempting to cover up or impede federal investigation of those violations, and substantive violations of the Clean Water Act and the Clean Air Act. Specifically, the Defendants were found to have illegally pumped contaminated water into storm drains and, as a result, into the Delaware River; to have unlawfully burned 50-gallon drums of paint waste in a cupola and emitted the fumes from those activities into the air; and to have attempted to cover up several work-related accidents at its facility, one of which resulted in the death of an employee. The jury also found that the Defendants engaged in a conspiracy to commit these acts—and to impede the resulting federal investigation—in order to maximize productivity and profits at the Plant. The Defendants appealed from the jury’s verdict, raising a litany of issues relating to pre-trial discovery, the District Court’s handling of the trial itself, the propriety of certain jury instructions, and the District Court’s sentencing determinations. For the reasons that follow, and in light of the District Court’s fine handling of these extraordinarily complicated proceedings, we will affirm the final judgments of conviction and sentence in this case. 7 I. Facts & Posture A. Factual Background 1 1. The Defendants Atlantic States Cast Iron Pipe Company (the “Company”), owned by McWane, Inc., operates a pipe foundry in Phillipsburg, New Jersey. The Plant, which produces ductile iron pipes used as municipal water pipes, sits on a 33-acre facility located just one mile from the Delaware River. Prior to 2002, the facility had several large storm drains that flowed through the municipal storm sewers to an outfall pipe that fed into the Delaware River. 2 During the periods in question, the Plant was overseen by John Prisque, who became the Plant Manager in 1998. Prior to that, Prisque served as the Production Superintendent and the Production Manager. Jeffrey Maury, like Prisque, had a long career at the Plant. After serving as the Maintenance Foreman from 1995 1 A more detailed discussion of the evidence introduced at trial can be found in the District Court’s comprehensive opinion. United States v. Atl. States Cast Iron Pipe Co., No. 03-852, 2007 WL 2282514, at *69-133 (D.N.J. Aug. 2, 2007). 2 After 2002, the Company overhauled its internal sewage and drainage systems and completely removed itself from the city’s sewer system. The Company says that this eliminated any risk of future discharges into the Delaware River and allowed them to conserve water for use at the Plant. 8 to 1997, Maury was promoted to Maintenance Superintendent in 1998. In this position, Maury oversaw the maintenance of all equipment used in the casting and finishing of the Company’s pipes and of construction equipment, such as forklifts. He also supervised all of the Plant’s maintenance foremen. Maury reported directly to Prisque. All finishing processes at the Plant, including the cement lining and pipe painting operations, were overseen by Craig Davidson, who served as the Finishing Superintendent beginning in March 1998. Like Maury, Davidson reported directly to Prisque. Scott Faubert served as the Plant’s Human Resources Manager from 1996 to September 2000. His job responsibilities included overseeing the Plant’s health and safety programs and supervising the Plant’s Safety Director. Faubert also reported directly to Prisque. 2. The Plant As part of the Company’s production process, employees at the Plant melt scrap iron and steel at an extremely high heat in a cupola, or furnace. The molten metal is then poured into a pipe cast and passed through an oven that makes the pipes ductile. From there, the pipes are sent to the finishing department, where they are cooled with water, grinded, pressure tested, lined with cement, rinsed and painted. Much of the machinery used during this process relies on hydraulic cylinders and petroleum-based hydraulic fluid. This process produces contaminated water and air pollutants, which are governed by the Clean Water Act, 33 U.S.C. § 1251 et seq., and the Clean Air Act, 42 U.S.C. § 7401 et seq., respectively. Discharges of any such contaminated water and emissions of any such air pollutant 9 are controlled by strict permits issued to the Company by the relevant policing agency. As an industrial facility, the Plant is also subject to the requirements of the Occupational Health and Safety Act, as administered by the Occupational Health and Safety Administration (“OSHA”). See 29 U.S.C. § 651 et seq. OSHA has promulgated volumes of regulations designed to ensure that workers can perform their jobs without substantial risk of industrial accident. Under OSHA’s regulations, the Plant is required to comply with certain health and safety standards, educate employees about workplace safety, and report any on-the-job injuries. Certain categories of workplace injuries, including those involving a fatality or the risk of an ongoing imminent danger at a facility, are subject to investigation by a regional OSHA inspector. The Defendants in this case were charged with and convicted of conspiring to violate these regulatory schemes and to impede federal investigations into those violations. They were also separately charged with and convicted of several of the underlying violations. The sections that follow describe these violations in turn. B. Clean Water Act Violations Congress enacted the Clean Water Act (the “CWA”) to preserve the environmental and physical integrity of the national waterways. 33 U.S.C. § 1251. To that end, the CWA “prohibits the discharge of any pollutant into waters of the United States except as expressly authorized.” United States v. Allegheny Ludlum Corp., 366 F.3d 164, 171 (3d Cir. 2004). Chemical or industrial waste constitutes a pollutant. 33 U.S.C. § 1362(6). An individual wishing to discharge such a substance must first obtain a National Pollution 10 Discharge Elimination System permit (a “discharge permit”) from the relevant agency overseeing the program—in this case, the New Jersey Department of Environmental Protection (“NJDEP”). 3 Any discharge must comply with the terms of the permit, and the restrictions set out in the CWA. Id. § 1342(k). The Company held a discharge permit that fundamentally allowed the Plant to discharge two types of substances: (1) storm water runoff (e.g., rainwater); and (2) “non-contact cooling water”—i.e. water that has not come into contact with any industrial machinery or its associated pollutants. Atl. States, 2007 WL 2282514, at *14. To qualify as “non-contact cooling water,” the discharged water could have no visible sheen, had to pass tests designed to measure petroleum hydrocarbon levels, and could contain no solid debris. At no point was the Company permitted to discharge wastewater created from the cleaning of equipment, the pipes produced in the Plant, or the Plant’s facilities. 1. Unlawful Discharge of Wastewater The Plant uses large volumes of water at several stages of its manufacturing process, producing both noncontact 3 Pursuant to 33 U.S.C. § 1342(b) and (c), the Environmental Protection Agency (the “EPA”) may authorize the issuance and enforcement of discharge permits by analogous state agencies so long as the state program meets federal requirements set forth under the Act. New Jersey has administered its own discharge permit program since 1982. See 47 Fed. Reg. 17,331 (Apr. 22, 1982). 11 cooling water and wastewater. During the relevant time periods, water was pumped from large containment tanks through the Plant via a closed-loop system, and was used to cool the pipes after casting, to cool the casting machines themselves and to cool the cupola. The Plant also used water to rinse the pipes during the cement lining process and afterward, prior to painting. Both processes yielded non- dischargeable wastewater. a) The Number Four Pit While some of the cooling water burned off as steam, the Plant’s cooling processes—the casting process in particular—also generated wastewater. During the casting process, the water used to cool the pipes dripped from the closed-loop recirculation system into a pit—referred to as the Number Four Pit—located beneath the casting machine. 4 Because the machines also dripped hydraulic oil into the pit, the water in the Number Four Pit was often contaminated. The Number Four Pit was positioned adjacent to the “bull ladle pit,” in which molten iron from the cupola was held, and had to be watched carefully to ensure that it did not overflow 4 Robert Bobinis was a millwright and foreman at the Plant and worked in the casting area. He testified that the water system in the casting area was designed to be a closed-loop, sealed system that recirculated the water used in the production process. However, the system did not function this way. Though the Plant also had a wastewater treatment process for this closed-loop system, employees testified that the water treatment facility was inoperable for much of the period in question. 12 into the bull ladle pit and cause an explosion. Workers testified that the pit was typically full and that the need to drain the pit varied day to day, depending on the severity of the leaks from the machinery above. To drain the Number Four Pit, workers were supposed to pump the wastewater through trenches that ran alongside the casting equipment and into large holding tanks located outside the building. However, both during production and during the night-shift maintenance activities, the workers were instructed to pump into the tanks even if the tanks were too full to handle more wastewater. The result of this pumping was that the tanks overflowed, sending “dark, hot, muddy water” flowing down the roadway alongside the Plant, and into the nearby storm drains. Id. at *118. Night-shift employees were also sometimes instructed to pump the water out of the pit and directly onto the ground outside the Plant. Though the frequency of this pumping depended on how much the machinery leaked in a given day, employees estimated that the discharges from the Number Four Pit occurred once or twice a week. Robert Bobinis, who worked in the casting area of the Plant, noted that he mentioned the overflow problem to Maury one day and told Maury, “I’m not going to jail for this.” Maury responded only with, “Shhh.” Id. at *120. b) The Cement Pit Like the Number Four Pit, the Cement Pit suffered from too much wastewater and too little storage. Water from the Plant’s cement washing process gathered in containment pits below the cement lining equipment and then flowed, via trenches or troughs, to the Cement Pit located outside the building. Along the way, this water collected the cement debris and hydraulic oil that had accumulated during the 13 manufacturing process. Once in the pit, the debris sunk to the bottom and the oil floated to the top. As a result of these pollutants, water in the cement pits had a greenish, oily appearance. During production, this wastewater remained in the pit. The oil was supposed to be skimmed from the top of the water, and the water itself recycled through the Plant and reused for cement washing. However, over time, the debris at the bottom of the pit would accumulate and displace the water in the pit, until it began to interfere with production. This required that the pit water be removed, usually at the end of the production week, so that maintenance workers could scrape the bottom of the pit of debris. Workers were supposed to clean the pit by pumping the water out and into hazardous materials tanks that were maintained and emptied by outside contractors. In 1999, the Plant altered this system, installing next to the pit a large, brown tank specifically intended to hold the water during cleaning and allow for its reuse afterward. When the tank was full, workers were to pump any excess water into the preexisting hazardous materials tank, located across the roadway, for disposal. In reality, however, these practices were routinely disregarded in favor of illegally pumping the wastewater onto the roadway and, as a result, into the facility’s storm drains. 5 5 Testimony at trial indicated that the large brown tank, installed to hold this excess water, clogged frequently as a result of the debris in the water. Thus, shortly after it was installed in 1999, it became largely inoperative and was used only sporadically. 14 During the night-shift and, in particular, on Friday nights, maintenance workers were instructed to “pump the pit” by emptying the wastewater onto the nearby roadway. Id. at *102. After the pit was emptied, workers put the pumps away and cleared the debris from the bottom of the pit. They then re-filled the pit with water so that production could resume during the morning shift. Employees testified that some of these orders came directly from Davidson, the head of the Finishing Department. Other employees testified that they received their orders from their shift foreman. As with the discharges from the Number Four Pit, this water flowed from the roadway into the nearby storage drains and, eventually, into the Delaware River. This practice appears to have continued until February 2000, when environmental inspectors began looking into possible spills from the Plant. At that point, workers began pumping the cement pit water into the hazmat tanks. However, because those tanks were frequently full, the cement water still occasionally spilled over the sides of the tank, down the roadway to the storm drains and, eventually, into the river. 2. Resulting Oil Spills & Ensuing Investigations The unpermitted discharges from the Plant resulted in at least three separate oil spills on the Delaware River. Statements made by the individual defendants to the state and federal investigators looking into these spills would later serve as the basis for some of the false statements and obstruction charges in the indictment. 15 a) March 1998 On March 19, 1998, New Jersey residents reported seeing an oil slick on the Delaware River. An employee of the Phillipsburg Department of Public Works investigated and traced the spill to a large outfall pipe in the vicinity of the Plant. The discharge was eventually traced to the facility itself. The local agency sent the Company a letter, copied to the NJDEP, requesting that the facility take measures to avoid such spills going forward. b) December 1999 On December 5, 1999, another local resident noticed a large sheen of oil—later estimated to contain approximately 200 – 300 gallons of liquid petroleum 6—floating on the Delaware River. The resident reported the sighting to local authorities, which again triggered an investigation. Local authorities contacted Bruce Doyle, an emergency response specialist for the NJDEP, who began by visiting the location at which the sheen was spotted. He took photographs and collected samples of the water, and noted that he believed the discharge had been continuing for some time based on the amount of oil that he observed in the river. Doyle traced the discharge to an outfall pipe near the Plant. Later that same day, Doyle and the Chief of the Phillipsburg Fire Department, Richard Hay, went to the 6 Estimates were that this amount of petroleum, mixed with approximately 20 times that amount in wastewater, had been discharged into the river. 16 Company’s facility, where they were initially denied entry. While waiting outside the facility’s gates, Doyle observed large, oily puddles on the ground inside the gates. He could also see discharge flowing into two storm drains, one of which was near what he later learned to be the cement pit. He also observed a “partially-submerged sump pump that was not operating at the time.” Id. at *111. Doyle testified that it appeared to him that the water flowing into the drains had come from the exposed end of the sump pump lying on the roadway. Once allowed into the facility, Doyle confirmed that the cement containment pond was the source of the discharge he had observed at the outfall. Doyle also had an opportunity to speak with Joseph Maddock, the Plant’s safety director, who reported directly to Faubert. Maddock said he had no information on the sump pump or hose in the pit and told Doyle that “he had no idea what had caused the discharge.” Id. at *112. Doyle told Maddock that the Company was directed to clean up the spill at the Plant and at the outfall. NJDEP officials obtained a search warrant to examine the facility, and returned to the Plant on February 24, 2000. They were accompanied by officers from the New Jersey Department of Law and Public Safety, Christopher Fernicola and Jeffrey Hill; and FBI Special Agent James Spence. Doyle was also present. At that time, the cement pit’s condition was again full of debris, oil and milky-green water, and was fuller than it had been on December 5. Investigators collected a sump pump and hose, not in use, from around the pit and took those items into evidence. Investigators also took photographs of the area and sampled water from the pit. Later tests confirmed that the liquid in the pit matched the 17 liquid that had flowed through the storm drains and into the Delaware River. During this visit, officials interviewed Maury, Davidson and Prisque. Agents Fernicola and Hill interviewed Maury. Maury told them that he thought the leak had been caused by a leak in the hydraulic line on one of the street sweeper trucks used at the Plant. Agent Fernicola testified that Maury fidgeted in his chair and wrung his hands throughout the interview. 7 Davidson told Agent Hill that the discharge was caused by a hole in the hose that led from the sump pump to the storage tank where the water was held before being recirculated through the system. 8 Like Davidson, Prisque attributed the discharge to a hole in the sump pump hose. c) April 2000 On April 16, 2000, Doyle was again called to investigate reports of a floating oil sheen on the Delaware River. This sheen was smaller, but was still visible from the road. Doyle again traced the spill back to the outfall pipe near the Company’s facility, this time observing oil and a “cement-like material” in the discharge. Id. at *113. Doyle went to the Plant later that day and noted that, though there had been rainfall that day, the roads surrounding the Plant did not have puddles and were dry, indicating that the oil 7 Fernicola also testified that he believed that Maury was lying at the time of the interview. 8 Davidson’s pager went off at some point during the interview. Afterward, he refused to be interviewed further. 18 discharge had not been caused by surface rainfall washing the oil from the ground into the drain. He also observed a large oil sheen surrounding a grated drain at the facility and concluded that drain had been the source of the sheen he observed on the river. 9 The NJDEP again issued a directive and notice of violation against the Company. In light of these discharges and the oil spills, the Defendants—the Company, Maury, Davidson and Prisque— were eventually charged with violations of the Clean Water Act § 1319(c). They were also charged with lying to investigators about the source of the 1999 spill. C. Clean Air Act Violations The Clean Air Act (the “CAA”), overseen by the EPA but administered at the local level by the NJDEP, provides that “certain stationary sources of air pollution must obtain federal operating permits” prior to any emissions. Ocean Cnty Landfill Corp. v. United States EPA, 631 F.3d 652, 654 (3d Cir. 2011) (citing 42 U.S.C. §§ 7661-7661f, 7661a(d)(1), 40 C.F.R. § 70, App. A). Because the Company’s manufacturing process involves the smelting of metals in a cupola, it was required to obtain and comply with an NJDEP- issued emissions permit. 9 The District Court speculated that this smaller discharge may have been caused by “passive weekend leaking onto the roadway of cement pit water pumped to the hazmat pit, enhanced by the rain event of that day, and perhaps swept from the blacktop at Atlantic States before NJDEP arrived to check.” Atl. States, 2007 WL 2282514, at n.141. 19 1. The Plant’s CAA Permits The Plant’s cupola, which was approximately 70 to 80 feet high, required the use of various fuels in order to heat or “charge” the furnace to a temperature high enough to melt scrap metal to make the cast iron pipes. The Company held operating permits at all times relevant to this case. In addition to regulating how the emissions were to be measured and reported to the NJDEP, the operating permits also governed what fuels could be sent into the cupola and what amount of pollutant emissions could be released into the air. Under the relevant permits, the Company could use only certain types of fuel. The Plant typically used coke, but was also permitted to use non-hazardous waste paint in limited amounts. At all times relevant to this case, the Plant was permitted to burn no more than 55 gallons of non- hazardous waste paint per day. 10 No material other than a permissible fuel and the scrap metal being melted were to be burned in the cupola. In particular, hazardous waste paint— essentially, fresh, liquid paint—was never a permitted fuel, since it releases higher levels of volatile organic chemicals when burned. As for the Plant’s emissions levels, the Company was permitted to release from the cupola’s stack no more than 2,500 parts per million of carbon monoxide per hour, averaged over the course of a day. Three times per quarter, the Company was permitted to burn up to 4,000 parts per million of carbon dioxide per hour, averaged over the course 10 Prior to 2001, the Plant was permitted to burn 55 gallons of non-hazardous waste per hour. 20 of a day. Any amount beyond these levels, termed an “exceedance,” was in violation of the Plant’s operating permit and had to be reported to the NJDEP. Atl. States, 2007 WL 2282514, at *124. The Plant had a long, documented history of failure to comply with these limitations and, as of late 2000, was facing fines of approximately $600,000 for documented exceedances. Prisque would have known about these exceedances and any others because, as Plant Manager, he was responsible for certifying the Plant’s compliance and exceedance reports to the NJDEP. 2. Burning Excess Paint The pipe finishing process, during which the pipes were painted inside and/or outside, generated large volumes of waste paint. The Plant’s disposal method for this waste, at least in part, was to burn it in the Plant’s cupola in violation of the Company’s operating permit. As automated machines spray painted each pipe, excess paint dripped and collected on the floor below the paint machine, forming a thick, tarry coating. During the night shift, once production had stopped, workers used flat shovels to scoop the waste paint into 55-gallon drums, generally filling them 2–3 inches from the top of the drum. Some employees testified that they also placed paint-covered plastic drop cloths—not permitted in the cupola—in the drums. The workers then placed small pieces of iron called “chill” in each drum so that the crane, which used a magnet to lift things, could lift them. The workers then sealed the drums, duct taped the seam of the lid and moved the drums to the scrap yard. Employee testimony at trial demonstrated that these drums of waste paint frequently were burned in the cupola. 21 After placing the drums in the scrap yard, cranes would lift the drums into the cupola and burn them. In addition to waste paint, the drums often contained the plastic tarps on which the paint had gathered under the painting station. Employees of the Plant testified that Davidson instructed them to handle the paint in this way, that he did so specifically when the drums started to build up in the scrap yard, and that workers filled and moved somewhere between four and ten drums per night. The Plant disposed of waste paint this way as early as 1999, burning an average of four drums a day. NJDEP officials first spotted large collections of 55- gallon drums in the scrap yard in February 2000, while they were executing the search warrant to investigate the 1999 wastewater discharge into the Delaware River. Officials photographed the drums, some of which were sealed and others of which were open, exposing a black, tarry substance. Going forward, the Company made efforts to hide any existing drums from state officials when they visited the Plant, or to dispose of them prior to the officials’ arrival. One crane operator testified that, in anticipation of an expected NJDEP visit in 2003, his supervisor told him to put one 55- gallon drum of paint per hour into the cupola. When the crane would not lift some of the drums, he was told to bury them in scrap metal so that they would not be visible. George Shepherd, the former General Electrical Foreman and a key government witness at trial, testified that this incident occurred around April 2003. He explained that this was fairly standard practice any time “visitors” came to the facility. Prisque would instruct them to “make sure there’s nothing out 22 in the plant they could find.” 11 Id. at *128. Shortly after the NJDEP inspectors visited, Prisque directed Shepherd to get rid of the drums that had been buried. Shepherd said that the Plant burned approximately 20 55-gallon drums of waste paint in the cupola that day, and another 15-20 drums the next. Moreover, both the workers and their supervisors understood that burning the paint in the cupola was causing the Plant to exceed the emissions cap in its operating permit. Shepherd testified that he reported this concern to Prisque in 2003, after the aforementioned incident, but that Prisque told him that the paint had nothing to do with the meter readings and said, “Don’t go there.” Atl. States, 2007 WL 2282514, at *128. In light of these practices, Prisque and the Company were eventually charged with violating the Clean Air Act between February and August 2003. D. OSHA Incidents Between 1999 and 2002, several of the Company’s employees suffered severe injuries at the Plant due to unsafe working conditions. The evidence introduced at trial demonstrated that the Company’s management took steps to conceal the cause of those injuries and to obstruct OSHA investigators’ inquiries. 11 The crane operator also testified that Prisque had given him similar instructions. 23 1. The Coxe Fatality On March 24, 2000, Al Coxe, an employee at the Plant, was run over and killed by a forklift driven by Juan De Los Santos. De Los Santos had never been trained, as required, on the forklift and this was his second accident. OSHA officials were notified of the accident, as required by law. Safety Compliance Officer Carol Tiedeman was assigned to investigate the facility and arrived that same morning. Faubert, the Human Resources Manager and the individual in charge of overseeing safety at the Plant, met with Tiedeman and showed her the scene of the accident. Faubert informed Tiedeman that the forklift driver had not been properly trained on the equipment. He also showed her the forklift—referred to as forklift #24—which appeared to be working properly, and assured Tiedeman that it had been untouched since the accident. As she was leaving the facility that day, Tiedeman instructed Faubert to gather any inspection sheets for the forklift. She recommended that Faubert have the forklift completely inspected before it be used again. When Tiedeman returned to the Plant on April 7, 2000, she met with Faubert and his subordinate, Joe Maddock, the employee charged with maintaining the Plant’s OSHA injury log. She was given an inspection report, dated the day after the Coxe accident, which noted that Maury and Yukna, the garage mechanic, had inspected forklift #24 and found it to be in “perfect operating condition.” Id. at *77. In later visits to the Plant, however, Tiedeman was given the opportunity to review several drivers’ maintenance checklists for forklift #24 from immediately before and immediately after the accident. Almost all of those checklists noted extensive problems with 24 the forklift, including that the brakes were defective. Moreover, the forklift’s repair logs showed no work performed on the forklift, suggesting that it could not have been in “perfect” condition during or immediately after the Coxe accident. Id. Indeed, despite Faubert’s representations to the contrary, later investigations demonstrated that the Company’s management had concealed the fact that the brakes on forklift # 24 had not been functioning at the time that Coxe was run over, and that this was the cause of the accident. As a general matter, testimony and equipment records from the Plant demonstrated that the Plant’s forklifts were often in disrepair, with brake failure being one of several recurring maintenance problems. Forklift operators repeatedly reported problems with the machines, but were told to drive the equipment or risk losing their jobs. These forklift operators testified that, in order to make do, they routinely brought their own brake fluid to work, and many learned how to drop the forklift’s load in order to stop the machine without brakes. In fact, forklift #24 had been reported as having brake problems during the shift immediately prior to the fatal accident. Shepherd testified that, just after Coxe’s death, Prisque told Maury to have Shepherd take forklift #24 to the maintenance area for inspection. There, Shepherd and Maury confirmed that the brakes had malfunctioned. Shepherd instructed Yukna to fix the problem. 12 When Shepherd saw the forklift later that morning, around the time of the OSHA 12 Yukna did not testify at trial. 25 visit, he noted that the brakes were working again and assumed they had been fixed. At some point during the morning, Prisque also held a meeting with Maury, Davidson, Shepherd, and two others. Shepherd said that, at this meeting, he was instructed to make preparations for the inevitable OSHA visit, which might include a walk through the facility. Robert Rush, another employee who testified at trial, stated that Prisque told him to lie about the condition of the forklift, and to tell OSHA investigators that “the forklift was fully operational, it was safe, and [De Los Santos] was driving recklessly.” Id. at *89. When Rush told Prisque that he did not want to lie, Prisque threatened him and said, “You’re going to tell them this because your job depends on it. It’s in the best interest of your employment. In the best interest of your job somewhere down the line.” Id. 2. The Marchan Incident Gabriel Marchan, a plant supervisor, was struck by a forklift driven by De Los Santos on April 27, 1999. Though the accident occurred in 1999, OSHA first learned about Marchan’s injury in 2000, while investigating the Coxe accident. There had been no prior investigation into the accident because Marchan’s injury had not been recorded, as required, in the Plant’s OSHA 200 logs—logs in which employers are required to record workplace injuries. In April 2000, while investigating the Coxe accident, Tiedeman read a news article which stated that the forklift driver who had run over Coxe had been involved in a similar incident at the Plant the year before. When Tiedeman asked Faubert about this incident, he chuckled and stated that Marchan’s leg had been broken. Tiedeman requested a copy 26 of Marchan’s accident report. The evidence presented at trial demonstrated that the Defendants thereafter took steps to conceal the extent of Marchan’s injuries from OSHA. In May 2000, Faubert provided Tiedeman with a statement from Marchan that stated falsely that he had come back to work the day following the accident. When Tiedeman, confused, asked about the broken leg, Faubert changed his story and said that there had been no broken leg. When Tiedeman asked why the injury had not been recorded, as required, Faubert again stated that this was because the leg had not been broken and because Marchan had returned to work the next day. In light of these discrepancies, Officer Tiedeman opted to interview Marchan directly. Marchan testified that, just before his interview with Tiedeman in July 2000, Faubert threatened to fire him unless he lied about his injuries. Marchan thus told Tiedeman that he had been struck and run over by a forklift, but that he had only sustained a scratch and a bruise as a result. He denied having been seriously injured. That same day, Tiedeman also interviewed Faubert and his subordinate, Maddock, regarding the Plant’s failure to log the Marchan accident. Maddock stated that he understood that OSHA required employers to report employees’ on-site injuries and any restrictions on their ability to work, and said that it was his job to decide whether to include an injury in the OSHA log, based on the description of that injury in the Plant’s own first aid log. Maddock denied having been at the Plant on the day that Marchan was run over. When Tiedeman asked Maddock why Marchan’s injury had not been recorded and whether Marchan had missed any work because of it, Faubert interjected that Marchan had not been treated at the plant and that, as a result, no OSHA log entry was made. 27 It was only later, in reviewing subpoenaed company records that Officer Tiedeman learned that Faubert, Maddock and Marchan had lied to her. Despite their representations to the contrary, Marchan had been seriously injured and placed on restricted duty for 44 days following his accident. Indeed, Marchan testified that, immediately after the accident, it was Maddock who had driven him to the hospital for treatment. After his leg had been placed in a splint, and despite his pleas that he be allowed to go home because he was in pain, it was also Maddock who drove him back to the Plant and forced him to sit in a chair and paint from a bucket for the rest of the day. During the remainder of the time that his leg was in a cast, another employee drove him to and from work. The Company had an obligation to report Marchan’s injury, and the fact that he was limited to restricted duty for the month and a half after his accident, to OSHA but had failed to do so. 3. The Owens Incident On June 25, 1999, employee Robert Owens was injured and lost his eye when a piece of a rotating blade from the cut saw he was using broke off from the blade and struck him in the face. OSHA compliance officer Alex Salerno was assigned to investigate the incident and to assess whether adequate safety features had been in place at the time of Owens’ injury. Salerno visited the Plant roughly two weeks after Owens’ accident, met with Faubert and was escorted to the site of the accident. At that time, he witnessed a worker operating the machine from behind a sliding plexiglas shield with a wooden frame and a wire mesh screen that was designed to protect the worker. Salerno observed that the shield appeared to have been newly built, but Faubert assured him that it had existed in that condition for 16 years. The 28 employee at the machine told Salerno the shield had been in place for 10 years. Despite these representations, evidence and testimony at trial indicated that no wire mesh safety shield had been in place on the day of Owens’ accident. Rather, the plexiglass and wire mesh covering had been built by a carpenter a few days after Owens’ injury. The employee who told Salerno otherwise later testified at trial that Prisque instructed him to say that the shield “was always there” and that he lied to Officer Salerno because he wanted to keep his job. Id. at *84. 4. The Velarde Incident Attempts to conceal deficient safety measures and employee injuries were made again in December 2002, after Hector Velarde lost three fingers in a cement mixer accident. Velarde was cleaning the mixer on December 7, 2002, when his co-worker activated the mixer without alerting him. When the mixer’s blades activated, Velarde’s fingers were amputated. Officer Tiedeman was the OSHA inspector assigned to investigate Velarde’s injury. On Tiedeman’s first visit to the Plant to investigate this incident, she met with Prisque, Don Harbin, the Plant’s maintenance manager, and Mark Neetz, then the Plant’s safety director. All three individuals escorted her to the mixer, and Harbin explained what had happened during the accident. When Tiedeman asked whether the machine had come with a safety switch or interlock switch—designed to shut down the machine if the doors were open—she was assured that it had not, but was told the Plant would attempt to install one itself. In truth, the mixer had come from the manufacturer with an interlock switch installed, but Prisque had instructed 29 Shepherd to bypass the switch because it slowed production. 13 Shepherd testified that prior to Tiedeman’s arrival, at Prisque’s instruction, he had removed all evidence that he had bypassed the switch in the first place and had rigged the machine to look as though no interlock switch ever existed. Before Tiedeman’s second visit to the Plant, Shepherd reinstalled the original interlock switch in order to make it look like the Plant had taken steps to improve the safety of the mixer. Tiedeman only learned that the mixer actually had been manufactured and shipped to the Plant with a functioning interlock switch in February 2003, when she requested and reviewed the owner’s manual for the equipment. The manual explicitly warned of the risks of bypassing the safety feature. When she confronted Harbin with this discovery, he confessed that there had originally been a switch, but said that it had only functioned for a day or two before it got too dirty to operate properly. He also speculated that the switch had been removed by an employee no longer at the Company and said he could not provide a name for that individual. In light of these incidents, the Company, along with Faubert, Prisque, and Maury, were charged with lying to 13 The doors had to be opened frequently during production to determine whether the cement was the right consistency. However, because of the interlock safety switch, every time the doors were opened the machine had to be restarted. Shepherd testified that this delay caused the cement line to become backed up during production. 30 OSHA inspectors and with attempting to obstruct their investigation into the employees’ injuries. E. Indictment On December 11, 2003, a federal grand jury sitting in the District of New Jersey indicted Defendants Prisque, Faubert, Maury, Davidson and the Company14 on charges of conspiring, in violation of 18 U.S.C. § 371, to (1) knowingly discharge a pollutant into U.S. waters, without and in violation of a permit, in violation of the Clean Water Act, 33 U.S.C. §§ 1311(a) and 1319(c)(2)(A); (2) knowingly violate a requirement and prohibition of permits under the Clean Air Act, 42 U.S.C. § 7413(c); (3) defraud the United States by obstructing the lawful functions of the [OSHA] and the [EPA] in enforcing federal workplace safety and environmental laws and regulations; 14 Daniel Yadzinski, the Plant’s engineering manager and environmental manager, was also indicted, but was acquitted on all charges against him. He is, of course, not a party to this appeal. 31 (4) make false statements in matters within the jurisdiction of OSHA, EPA and the [FBI], in violation of 18 U.S.C. § 1001; and (5) corruptly influence and obstruct the administration of law under a pending proceeding before OSHA, in violation of 18 U.S.C. §§ 1505 and 1515(b). D. Ct. Op. 2. The indictment also charged the defendants with committing, or aiding and abetting the commission of, substantive violations of the aforementioned laws. Maury was charged with making false statements to investigators concerning the December 4, 1999 oil spill (Count 3), the Coxe forklift accident (Count 5), and with obstructing the OSHA investigation into Coxe’s death (Count 9). He was also charged with substantive violations of the Clean Water Act for the pumping of the cement pit that resulted in the sheen on the Delaware River on December 5, 1999 (Count 27), and the Number Four Pit from May to October 1999 (Counts 28-33). Davidson, in addition to the conspiracy charge, was indicted for substantive violations of the CWA for the pumping of the cement pit from December 1998 to February 2000 (Counts 12-26), and for the pumping of the cement pit that resulted in the sheen on the Delaware River on December 5, 1999 (Count 27). Davidson was also charged with making false statements to investigators regarding that spill (Count 4). Prisque was charged with violating the CWA based on the pumping of the cement pit that resulted in the sheen on 32 the Delaware River on December 5, 1999 (Count 27) and with a substantive violation of the Clean Air Act for his participation in the burning of the Plant’s waste paint in the cupola between February and August 2003 (Count 34). Prisque was also charged with making false statements to OSHA officers concerning the Owens accident (Count 8), and with obstruction of a federal investigation for his role in concealing evidence regarding the Coxe forklift fatality, Marchan’s forklift injury, and the Velarde accident (Counts 9- 11). Faubert was charged with making false statements to OSHA officials concerning Marchan’s forklift accident (Count 7), obstructing the investigation of Marchan’s forklift accident (Count 10), and obstructing the investigation of the Coxe fatality (Count 9). The Company was also charged with numerous crimes, including making false statements regarding the 1999 oil spill (Counts 3-4), the Coxe fatality (Count 5), and the Marchan accident (Count 7). It faced obstruction charges for the attempts to conceal evidence of the Owens accident (Count 8) and the cause of Coxe’s death (Count 9), the extent of Marchan’s injury (Count 10), and the cause of Velarde’s cement mixer accident (Count 11). Prosecutors also charged the Company with violating the CWA for the pumping of the cement pit from December 1998 to February 2000 (Counts 12-26), the pumping of the cement pit that resulted in the sheen on the Delaware River on December 5, 1999 (Count 27), and the pumping of the Number Four Pit in 1999 (Counts 28-33). Finally, the Company was charged with violating the Clean Air Act (Count 34). 33 F. Pre-Trial Issues Pre-trial discovery in this case was highly contentious. 15 In May 2004, the Government sought (and eventually obtained) a protective order allowing it to defer production of otherwise discoverable statements by current and former employees of the Plant until 30 days before the start of trial. The Government based this request on allegations of workplace harassment and intimidation against Plant employees and their concern that witnesses would be penalized for cooperating with the prosecution or would be too intimidated to testify at all. G. Trial & Sentencing Trial began on September 27, 2005. The United States called 50 witnesses—many of them employees of the Company—and three rebuttal witnesses. Among the employees who testified were five unindicted co-conspirators: Joseph Maddock, the former safety and human resources director of the Plant; Donald Harbin, the Plant’s maintenance manager; George Shepherd, the general electrical foreman; Tom Dalrymple, who supervised emissions controls at the cupola and reported directly to Prisque; and Steven Wayda, the union representative at the Plant. The Defendants began presenting their case on February 8, 2006 and called another 52 witnesses. All of the individual Defendants testified. Defendants moved for a judgment of acquittal at the close of the Government’s case, 15 For a more detailed account of the discovery proceedings in this case, see infra Part II. 34 and at the end of the evidence. The Court reserved judgment and submitted all counts to the jury. The Court also submitted to the jury, at the Defendants’ request, the lesser included offense (not charged in the indictment) of a negligent—as opposed to knowing—violation of the Clean Water Act, 33 U.S.C. § 1319(c)(1)(A). Negligent violations of the Act constitute misdemeanors, rather than felonies. Following these proceedings and after 10 days of deliberation, on April 26, 2006, the jury convicted the Defendants of almost all of the charges levied against them. 16 Notably, though the Company was convicted of felony violations of the CWA for the 1999 spill on the Delaware 16 The jury failed to reach a verdict on Count 2 of the indictment, which charged Faubert and the Company with making false statements to OSHA inspectors when Faubert told them that he was unaware of a pit excavated in the casting department that had supposedly collapsed on an employee’s leg. Maury was acquitted on Count 5, which concerned his representation to OSHA inspectors that the forklift had been found to be in “perfect operating condition” immediately after the Coxe fatality. The Company was acquitted on Count 6, which concerned alleged false statements to the NJDEP regarding the accuracy of NJDEP’s testing procedures. And Prisque was acquitted on Count 10, which alleged that he had been part of efforts to convince Marchan to lie to the OSHA inspectors about the extent of his injuries from the forklift accident in 1999. And, finally, though not relevant to this opinion, Daniel Yadzinski, the Plant’s engineering manager and environmental manager, was acquitted of all charges against him. 35 River and the unpermitted pumping of the cement pit (Counts 12-26, 27), the jury concluded that Prisque, Maury and Davidson were guilty only of negligent violations for those incidents. The jury found the Company and Maury guilty of felony violation of the CWA for the unpermitted pumping of wastewater from the Number Four Pit (Counts 28-33). All Defendants were convicted of having knowingly and willfully engaged in a conspiracy to achieve the purposes set out in the indictment. Following the jury verdicts, the Defendants filed an omnibus post-trial brief that raised a multitude of challenges to the jury’s verdicts and to the District Court’s handling of the case. In August 2007, the District Court ruled on the Defendants’ post-trial motions, rejecting the majority of those challenges, but granting Rule 29 judgments of acquittal for insufficient evidence on the following: (1) one false statement charge against Faubert and the Company, for the jury’s failure to reach a verdict (Count 2); (2) one CWA charge against Davidson and the Company, concerning an alleged unlawful discharge of wastewater in September 1999 (Count 21); and (3) one CWA charge against Maury and the Company concerning an alleged unlawful discharge of wastewater in October 1999. The Court denied the Rule 29 motions as to all other counts of conviction. The District Court held sentencing hearings in April 2009. It sentenced Prisque to 70 months’ imprisonment, Faubert to 41 months’ imprisonment, Maury to 30 months’ imprisonment, and Davidson to 6 months’ imprisonment. As for the Company, the Court opted to apply the Alternative Fines Act (the “AFA”), 18 U.S.C. § 3571(c)(1), rather than the CWA and CAA, in imposing criminal penalties. Applying the AFA, the Court fined the Company the 36 maximum penalty of $500,000 per violation on Count 1 (conspiracy), Counts 8-11 (obstruction), Counts 12-16 (CWA—cement pit discharge), Counts 28-32 (CWA— Number Four Pit discharge), and Count 34 (CAA) for a total fine of $8 million dollars. It also sentenced the Company to 4 years’ probation, with a court-ordered monitor to ensure regulatory compliance going forward. The Defendants timely appealed to this Court in May 2009. H. The Present Appeal & the Parties’ Arguments 17 Jointly, the Defendants raise a multitude of arguments, principal among them that (1) they were afforded inadequate pre-trial discovery under Rule 16 of the Federal Rules of Criminal Procedure, (2) the Government withheld substantial Brady and Giglio material and engaged in numerous acts of prosecutorial misconduct, and (3) the District Court erred in instructing the jury on the appropriate standard of mens rea for a negligent violation of the CWA. They also raise numerous challenges to the District Court’s handling of the proceedings themselves, including its evidentiary rulings and the adequacy of the voir dire. The Defendants’ individual claims on appeal are equally numerous. The Company, in addition to challenging the adequacy of the Rule 16 discovery it received, argues that the District Court erred in denying its motions for acquittal 17 The District Court had jurisdiction under 18 U.S.C. § 3231. We have jurisdiction under 28 U.S.C. § 1291. 37 and for a new trial. It also challenges the reasonableness of the criminal penalties imposed by the District Court under the AFA. Faubert contests the sufficiency of the evidence used to convict him; argues that there was a fatal duplicity in the charged objectives of the conspiracy, and challenges the sentence imposed by the District Court. Davidson argues that the verdicts against him for conspiracy and for negligent violation of the Clean Water Act are mutually exclusive; that the Government failed to prove essential elements of the crimes charged against him, that there was a variance in one of the charges against him, that the government “released false discovery” that hampered his ability to defend himself, and that he was prejudiced by the “unfair charging practice of the government.” Davidson Br. 24, 26. Prisque, like Davidson, argues that the verdicts against him were mutually exclusive. He also claims that the Government engaged in a pattern of prosecutorial misconduct against him, that the evidence was insufficient to convict him, and that the District Court erred in imposing his sentence. Finally, Maury challenges the sufficiency of the evidence against him. He also argues that his false statement conviction stemming from the December 1999 oil spill (Count 3) should be dismissed because the question posed to him was too ambiguous to support the charge, that there was a variance between the indictment and the evidence against him on that same charge, and that the District Court erred in allowing prejudicial testimony against him. 38 II. Discovery Issues The Defendants raise many challenges to the degree of criminal discovery they were afforded throughout the proceedings in this case. Among the many arguments they raise, the Defendants argue that the District Court erred in its interpretation of Fed. R. Crim. P. 16(a), as it applies to organizational or, in this case, corporate defendants. We find the issue concerning the question of the extent of a corporate defendant’s right to discovery under Rule 16(a)(1)(C) to be the most deserving of attention and further discussion, and we therefore address it herein. Because we find that the discovery the Defendants received was adequate and in keeping with the standards of Fed. R. Crim. P. 16, we discern no error in the District Court’s handling of discovery in this case. A. Pre-Trial Discovery in a Criminal Case The Defendants’ arguments raised issues involving several overlapping aspects of pre-trial criminal discovery. We begin with a brief overview of a criminal defendant’s right to discovery of the evidence that the government intends to use against him at trial. In a criminal trial, the government is subject to three sets of disclosure obligations: those imposed under the Jencks Act, 18 U.S.C. §3500(b), those imposed by Fed. R. Crim. P. 16 (“Rule 16”), and those imposed by the Supreme Court’s holding in Brady v. Maryland, 373 U.S. 83 (1963). 1. Jencks Material Pursuant to the Jencks Act, implemented by Federal Rule of Criminal Procedure 26.2, any time that a government witness testifies on direct examination, the defendant is 39 entitled to a copy of “any statement . . . of the witness in the possession of the United States which relates to the subject matter as to which the witness has testified.” 18 U.S.C. § 3500(b)(2). Provided the government contests the relevance of any portion of the statement, it must submit the entire statement to the court for in camera review. Id. § 3500(c). A “statement” is defined as (1) any written statement made and signed or adopted by the witness; (2) a “stenographic, mechanical, electrical, or other recording, or a transcription thereof” that recites “substantially verbatim”, and was made contemporaneously with, the witness’s statement; or (3) a recording or transcript of grand jury testimony. Id. § 3500(e). The purpose of Jencks disclosure is to provide the defendant with an opportunity to review the witness’s statements for any possible inconsistencies that he might use to impeach the witness. United States v. Rosa, 891 F.2d 1074, 1076-77 (3d Cir. 1989). Importantly, in light of this purpose, the government has no obligation to produce Jencks material until the witness has testified. 18 U.S.C. § 3500(a). 18 This limitation does not apply to testimony by the defendant himself. Id.; see also Fed. R. Crim. P. 16. 2. Fed. R. Crim. P. 16 The bulk of the government’s pre-trial disclosure obligations stem from Rule 16, which requires that the government, “[u]pon a defendant’s request,” allow the 18 Despite this limitation, many federal prosecutors routinely turn over Jencks material a few days before the witness testifies. 40 defendant access to certain categories of information that the government has collected over the course of its investigation. With respect to an individual defendant, the government must disclose (1) “[A]ny relevant oral statement made by the defendant, before or after arrest, in response to interrogation by a person the defendant knew was a government agent if the government intends to use the statement at trial.” Fed. R. Crim. P. 16(a)(1)(A). (2) Any relevant “written or recorded statement by the defendant” if the statement is in the government’s possession or control, and the government knows of the statement. Id. 16(a)(1)(B)(i). (3) Any written record containing the substance of a defendant’s oral statements provided the statement is made in response to interrogation by a person the defendant knew to be a government agent. Id. 16(a)(1)(B)(ii). (4) Any recordings of the defendant’s grand jury testimony, if related to the offense. Id. 16(a)(1)(B)(iii). Organizational or corporate defendants are also entitled to pre-trial discovery. Fed. R. Crim. P. 16(a)(1)(C). However, in this context, the meaning of Rule 16 shifts. An organization has no self-knowledge of its own conduct, since it acts through its agents, and must be afforded an opportunity to learn what of its employees’ conduct is being used against it at trial. Thus, under Rule 16(a)(1)(C), an organizational defendant is entitled to any statement described in Rule 16(a)(1)(A) and (B), above, so long as the government contends that the person making the statement: 41 (i) was legally able to bind the defendant regarding the subject of the statement because of that person’s position as the defendant’s director, officer, employee, or agent; or (ii) was personally involved in the alleged conduct constituting the offense and was legally able to bind the defendant regarding that conduct because of that person’s position as the defendant’s director, officer, employee, or agent. Fed. R. Crim. P. 16(a)(1)(C)(i), (ii). It is only by learning what statements can be attributed to it as an organization that a corporate defendant can defend itself at trial. Fed. R. Crim. P. 16(a)(1)(C) Advisory Committee Notes to 1994 Amendments. Moreover, the government must allow any defendant to review any “books, papers, documents, data, photographs, tangible objects, buildings or places” that are material to the defense, provided that the government intends to use that evidence in its case-in-chief or that the item was obtained from the defendant. Fed. R. Crim. P. 16(a)(1)(E). These obligations are subject to a carveout for attorney or agent work product “made . . . in connection with investigating or prosecuting the case.” Fed. R. Crim. P. 16(a)(2). Rule 16 also makes clear that it does not authorize “discovery or inspection of statements made by prospective 42 government witnesses except as provided” under the Jencks Act, 18 U.S.C. § 3500. Fed. R. Crim. P. 16(a)(2). 3. Brady and Giglio material Finally, independent of these obligations, under Brady v. Maryland, a prosecutor has an obligation to disclose “evidence favorable to an accused” individual so long as it is “material either to guilt or to punishment.” 373 U.S. 83, 87 (1963). Evidence is “material” if there is a reasonable probability that, “had the evidence been disclosed to the defense, the result of the proceeding would have been different.” United States v. Friedman, 658 F.3d 342, 357-58 (3d Cir. 2011) (quoting United States v. Bagley, 473 U.S. 667, 682 (1985)). “The prosecution must also disclose evidence that goes to the credibility of crucial prosecution witnesses.” Buehl v. Vaughn, 166 F.3d 163, 181 (3d Cir. 1999) (citing Giglio v. United States, 405 U.S. 150, 154 (1972)). Referred to as Giglio material, this evidence is a subset of Brady material insofar as it addresses situations in which certain evidence about a witness’s credibility or motivation to testify exists, and where “the reliability of a given witness may well be determinative of guilt or innocence.” Giglio, 405 U.S. at 154 (citing Napue v. Illinois, 360 U.S. 264 (1959)). B. Pre-Trial Discovery in this Case Fed. R. Crim. P. 16(d)(1) allows a court to, “for good cause, deny, restrict, or defer discovery or inspection, or grant other appropriate relief.” In this case, the District Court granted a protective order allowing delayed disclosure of certain employees’ statements that the government sought to use in its case against the Company—in other words, statements that the Court believed fell within the reach of 43 Rule 16(a)(1)(C). In granting that request, it reviewed in camera the statements themselves, and the Government’s affidavits in support of their motion. It determined that delayed production was the only way to “protect the safety of prospective witnesses and the likelihood that they will appear and testify without intimidation or improper influence at trial.” (J.A. 652.) The Court also determined that alternative, less drastic proposals, such as producing the materials with redactions of the employees’ names, or ordering disclosure of a list of the sought after statements and allowing the full documents to be produced later, would “undermin[e] the effect of the protective order to be issued.” (J.A. 652.) In the wake of the District Court’s initial decision granting the protective order, the Defendants filed several motions to compel discovery on a host of grounds. 19 In October 2004, the Defendants also argued that, despite the Court’s protective order, they were immediately entitled to any and “all statements made by the ‘co-conspirators’ referenced in the Superseding Indictment, as the United States is clearly attempting to bind the [C]ompany by the acts and/or statements of those co-conspirators,” as well as “all statements of employees whose alleged conduct binds Atlantic States.” ((J.A. 695.) (citing Fed. R. Crim. P. 16(a)(1)(C) (the “Rule 16 motion”)). They argued that even if the employees’ identities were protected under the Court’s 19 The Defendants filed motions to compel what they claimed was Brady material, Giglio material, and requesting early Jencks disclosure and rough notes from the OSHA and NJDEP inspectors, whom the government had not treated as “agents” within the meaning of Rule 16(a). 44 order, any and all statements by the employees should be turned over immediately. Importantly, these statements concerned not so much the scope of Rule 16(a)(1)(C), but whether the Defendants were entitled to anything under Rule 16(a)(1)(C) prior to 30 days before trial. See J.A. 695. The Defendants’ continuing demands for information concerned the timing of the Government’s disclosures, not the scope thereof. The Court reserved decision on the Rule 16 motion, pending further in camera review of the evidence, and noted that it would decide the Rule 16 motion in conjunction with the Defendants’ Motion for Reconsideration of the Protective Order. The Court intended to use that additional in camera review to determine whether the scope of the protective order could be narrowed. Following additional arguments and the Court’s in camera review, on February 17, 2005, the Court denied the motion to reconsider the protective order. On August 12, 2005, consistent with the Court’s protective order, the Government produced the previously withheld Rule 16(a)(1)(C) materials to the Defendants. In all, the production consisted of documents for 24 witnesses. Portions of these documents were heavily redacted, in keeping with the Government’s understanding of the scope of Rule 16(a)(1)(C). For individuals who could bind the Company by virtue of their position alone, the Government produced all oral statements that it intended to use at trial, and all written statements discussing issues on which the employee had the authority to bind the Company. See Rule 16(a)(1)(C)(i). For all employees who were capable of binding the Company by virtue of their participation in the charged events, i.e. employees who had participated directly in the charged conduct, the Government produced all oral and 45 written statements discussing that specific conduct. See Rule 16(a)(1)(C)(ii). All other statements from these individuals, such as statements unrelated to the charged conduct or observations of the conduct of others, were redacted. Following this production, the Defendants sent a letter to the Court complaining about the Government’s redactions, noting that “the disclosures lack[] any indication to substantiate” the need for the protective order. (J.A. 1533.) The Defendants “request[ed] that the summaries and transcripts be produced in unredacted form, so that the necessity for the Protective Order is clear.” Id. They did not object to the scope of the Government’s reading of Rule 16(a)(1)(C). 20 Full, unredacted versions of these documents were produced to the Defendants as Jencks material at least 3 days prior to the testimony of each witness. Of course, for those individuals who did not testify, no such Jencks disclosure occurred. See 18 U.S.C. § 3500(a). C. The Scope of Fed. R. Crim. P. 16(a)(1)(C) On appeal, the Defendants argue that the District Court erred in its application of Rule 16(a)(1)(C) and that it deprived the Defendants of “an inordinate amount of critical discovery” in the process. Joint Def. Br. 78. They concede that the Government’s application of Rule 16(a)(1)(C)(i), 20 The Defendants continued to file motions to compel discovery on other grounds. See, e.g., J.A. 1737-38 (“Motion to Compel Production of United States Attorney and Investigator Rough Notes”). 46 which governs individuals “legally able to bind the defendant regarding the subject of the statement” because of their position of authority, was in keeping with the requirements of that rule. See Joint Def. Br. 77. However, they argue that the discovery they received under Rule 16(a)(1)(C)(ii) was unduly narrow insofar as it was limited to statements by employees about the specific offending conduct. 21 The Defendants urge us to read Rule 16(a)(1)(C)(ii) to require that, “[o]nce an employee participated in the misconduct, all statements given by that employee should have been disclosed, even if unrelated to the misconduct.” Joint Def. Br. 80 (emphasis omitted). The Government responds that the Defendants’ argument is subject to plain error review because they failed to raise a challenge to the redactions under Rule 16(a)(1)(C)(ii) before the District Court. In any event, it argues, their redactions were in keeping with the scope and reach of Rule 16(a)(1)(C)(ii). 1. Applicable standard of review Though we review a district court’s application of the Federal Rules of Criminal Procedure, including its decisions 21 The Defendants raise this issue jointly and argue that they are all entitled to a new trial as a result of this supposed error. Rule 16(a)(1)(C), however, benefits organizational defendants alone, and an individual defendant is not entitled to discovery of the statements of his codefendants. Cf. United States v. Randolph, 456 F.2d 132, 135-36 (3d Cir. 1972). Therefore, even were we to find reversible error, only the Company would be entitled to the benefit of our ruling. 47 denying or limiting discovery, for an abuse of discretion, see Gov’t of Virgin Islands v. Fahie, 419 F.3d 249, 258 (3d Cir. 2005), our review of a court’s interpretation of the meaning or reach of a rule is plenary, see United States v. Toliver, 330 F.3d 607, 610 (3d Cir. 2003). Claims of error in criminal discovery that are raised for the first time on appeal are subject to plain error review. See Fed. R. Crim. P. 52(b). We agree with the Government’s contention that plain error review governs this argument on appeal. Despite the Defendants’ contentions to the contrary, the record demonstrates that they failed to directly challenge the appropriate scope of discovery available under Rule 16(a)(1)(C)(ii) before the District Court. In their reply brief, the Defendants point to three statements before the District Court which, they contend, adequately presented the issue for the District Court’s review: (1) their statement at the hearing for the Protective Order that said, “make no mistake about that—we want the entire statements and the rough notes,” J.A. 582; (2) their motion to compel discovery, in which they sought all statements from the relevant employees, even if the employees’ identities were not discoverable per the Court’s protective order, J.A. 696; and (3) their post-production letter in which they complained about the Government’s redactions and “request[ed] that the summaries and transcripts be produced in unredacted form, so that the necessity for the Protective Order is clear.” (J.A. 1533.) However, while these statements do reference Rule 16, they do so in the context of challenging the reach and necessity of the Court’s Protective Order, both prior to and 48 after the Government’s production. 22 None squarely presents the Court with the question of whether, under Rule 16(a)(1)(C)(ii), the scope of discovery is limited to only those statements that reference the offending conduct, or instead reaches any statement from an employee who has participated in that conduct. Therefore, because the Defendants never squarely presented the Court with this challenge, we hold that plain error review applies to the Defendants’ claim. Before an appellate court can correct an error not properly preserved before the District Court, it must find that “(1) there was error; (2) the error was clear or obvious; (3) the error affected the defendant’s substantial rights; and (4) the error seriously affected the fairness, integrity, or public reputation of the legal proceeding.” United States v. Tyson, 653 F.3d 192, 211 (3d Cir. 2011); see also United States v. Williams, 464 F.3d 443, 445 (3d Cir. 2006). We find no such “clear or obvious” error in this case. 22 We note that the Defendants have not directly challenged the appropriateness of the order itself. Had they, we would be hard pressed to find that the Court abused its discretion given the circumstances presented in this case and the evidence before the Court at the time of its decision. United States v. Hedaithy, 392 F.3d 580, 605-06 (3d Cir. 2004) (explaining that a district court abuses its discretion only where its “decision rests upon a clearly erroneous finding of fact, an errant conclusion of law, or an improper application of law to fact” (citation and internal quotation omitted)). 49 2. Rule 16(a)(1)(C)(ii) We know of no circuit court opinion that has confronted the precise scope of discovery available under Rule 16(a)(1)(C)(ii). This dearth in law alone makes it particularly challenging for the Defendants to prove that the discovery they received was plainly erroneous under the law as it stands. See United States v. Riley, 621 F.3d 312, 323 (3d Cir. 2010) (stating that an error is “clear or obvious” where the law at issue has been clarified at the time of the appeal). Moreover, because the Defendants never challenged the Government’s application of Rule 16(a)(1)(C)(ii) directly, there is no clear ruling by the District Court on this issue. We are left only to consider whether the discovery the Defendants received was clearly at odds with the language of Rule 16(a)(1)(C). We conclude that it was not. Under Rule 16, an individual is entitled to essentially (1) any oral statement made in response to interrogation by a government agent, provided that the government intends to use that statement at trial; (2) copies of any written or recorded statement by the defendant—or any record containing the substance of that statement—provided it is within the control of the government; and (3) the defendant’s grand jury testimony relating to the offense. Fed. R. Crim. P. 16(a)(1)(A), (B). Such statements are often admissions and can be used to bind the defendant at trial. Pre-trial discovery allows an individual defendant the opportunity to seek suppression of these statements before they are introduced into evidence at trial, see Fed. R. Crim. P. 16(a)(1)(C) Advisory Committee Notes to 1994 Amendments, and to evaluate the weight of such direct evidence against him in deciding whether to take a plea or face trial. 50 Rule 16(a)(1)(C) imports the individual-defendant discovery rules in subsections (a)(1)(A) and (B) to the context of organizational defendants, such as the Company in this case. The Advisory Committee Notes pertaining to this portion of Rule 16 make it clear that the disclosure requirements “apply equally to individual and organizational defendants.” Fed. R. Crim. P. 16(a)(1)(C) Advisory Committee Notes to 1994 Amendments (emphasis added). This, in our view, indicates that the same limitations and driving principles which control in the individual context transfer, through incorporation, to the discovery rights afforded to organizational defendants under Rule 16(a)(1)(C). Of course, organizations are able to act or speak only through their agents. The statements of agents can bind the corporation. See, e.g., Fed. R. Evid. 801(d)(2)(C), (D). And, “[b]ecause an organizational defendant may not know what its officers or agents have said or done in regard to a charged offense, it is important that it have access to statements made by persons whose statements or actions could be binding on the defendant.” Fed. R. Crim. P. 16 Advisory Committee Notes to 1994 Amendments. Rule 16(a)(1)(C) thus entitles an organizational defendant to those very statements. See id. In terms of evidence, an organization such as a corporation can be bound by the words of its employees in two basic ways: first, by a statement from a person with authority to speak on the behalf of the organization on that subject; second, by a statement from an employee or agent of the organization on a matter within the scope of his assigned conduct. See Fed. R. Evid. 801(d)(2). Such statements are admissible against the principal because we view the speaker, in these situations, as speaking for or as the principal. 51 Rule 16(a)(1)(C) similarly limits the disclosure afforded to those “statements made by a person who the government contends,” falls in one of two categories. Those who (1) [were], at the time of making the statement, so situated as a director, officer, employee, or agent as to have been able legally to bind the defendant in respect to the subject of the statement; or (2) [were], at the time of the offense, personally involved in the alleged conduct constituting the offense and so situated as a director, officer, employee or agent as to have been able to legally bind the defendant in respect to that alleged conduct. See Fed. R. Crim. P. 16(a)(1)(C). Thus, two classes of individuals can make statements that are sufficient to constitute admissions against the corporation, which are thus discoverable under Rule 16(a)(1)(C): (1) representatives, or individuals who have the power to bind an organization by virtue of their authority to make statements on the subject on behalf of the organization, see Fed. R. Crim. P. 16(a)(1)(C)(i); and (2) employees who engage in illegal conduct within the scope of their jobs and then make some statement about having done so, see Fed. R. Crim. P. 16(a)(1)(C)(ii). Against this backdrop, we think the second category, which references “the conduct constituting the offense” and the ability “to bind the defendant in respect to that alleged conduct,” contemplates that the statements governed by Rule 16(a)(1)(C)(ii) are tethered to the conduct itself. In keeping with traditional notions of agency and vicarious liability, it is only in this context that the employee “speaks” on the behalf 52 of the Company as concerns the charged conduct against which the organizational defendant must defend itself. Indeed, and perhaps most convincingly, any broader reading begins to afford organizational defendants greater rights than those afforded to individual defendants. Rule 16 recognizes the dueling identities of an organization’s average employee: first, as a situational agent of the Company and second, as an average, run-of-the-mill fact witness. The further a general employee’s statements diverge from admissions about having engaged in a form of conduct on behalf of the Company, the more they begin to resemble general, fact-based, witness statements. Such statements are not discoverable to individual defendants prior to the witness’s trial testimony and until the Government’s Jencks obligations kick in. See Fed. R. Crim. P. 16(a)(2); 18 U.S.C. § 3500. To make such statements discoverable under Rule 16(a)(1)(C) would grant organizational defendants a windfall, by allowing such defendants to use Rule 16 as a means of gaining early access to witness’s inculpatory statements when non-organizational defendants have no such right. See United States v. Dessange, No. 99-Cr.-1182, 2000 U.S. Dist. LEXIS 2557, at *8-9 (S.D.N.Y. Mar. 7, 2000); United States v. Lin Lyn Trading, Ltd., 911 F. Supp. 494, 497 (D. Utah 1996); United States v. Bhutani, No. 93-585, 1995 WL 632069 (N.D. Ill. Sept. 1, 1995) (explaining that Rule 16 reaches only those statements that are in the nature of admissions). To the extent a defendant wishes to use those statements to contradict the witness, or to draw the witness’s credibility into question at trial, that is the stuff of Jencks disclosures and is adequately accounted for by that requirement. The Defendants rely on United States v. Chalmers—a case from the Southern District of New York that was 53 decided after discovery in this case—in support of their proposition that once an employee has engaged in bad conduct sufficient to bind the organization, all statements of the employee are discoverable under Rule 16(a)(1)(C). 410 F. Supp. 2d 278 (S.D.N.Y. 2006). In Chalmers, the court held that under Rule 16(a)(1)(C) “the Government must disclose all statements made by persons whose actions it alleges bind the [organizational defendant].” Id. at 291. However, we note that other courts, including one other court from the same district, have reached conclusions similar to our own. See Dessange, 2000 U.S. Dist. LEXIS 2557, at *8; Lin Lyn Trading, Ltd., 911 F. Supp. at 494; Bhutani, 1995 WL 632069. Moreover, the Chalmers holding creates the very windfall we think must be avoided by closely focusing on the conduct that binds the organization under Rule 16(a)(1)(C)(ii). We have long held that Rule 16 is not to be used as a tool for general evidence-gathering prior to trial. See United States v. Randolph, 456 F.2d 132, 136 (3d Cir. 1972) (“Rule 16 does not require the prosecution to disclose ‘all the minutia of its evidence, to reveal its trial strategy, and to delineate with total specificity the case it intends to present.’”) (quoting United States v. Fioravanti, 412 F.2d 407, 411 (3d Cir. 1969) cert. denied, sub nom. Pannacione v. United States, 396 U.S. 837 (1969)). Because the Defendants’ interpretation of the rule would effectively allow that very result and, in so doing, give organizational defendants a substantial advantage in criminal proceedings, we must reject it as an incorrect reading of Rule 16(a)(1)(C)(ii). Rather, we conclude that where a statement itself is being used to bind the Company, Rule 16 affords an 54 organizational defendant the opportunity to access that statement—and only that binding statement—prior to trial so that the defendant can seek to have it suppressed and can attempt to estimate the damage that might be inflicted by it at trial. This is consistent with and equal to the discovery allowed to individual defendants under Rule 16(a). The discovery afforded to the Defendants in this case was in keeping with this reading, and therefore was not plainly in error. III. Jury Instructions on the Clean Water Act Violations Defendants raise two primary challenges to the District Court’s instructions to the jury. First, they argue that the District Court erred in defining the culpable mental state for a misdemeanor violation of the CWA as simple negligence, rather than gross negligence. Second, they argue that the District Court erred in refusing to include language stating that a showing of recklessness could not meet the mens rea for the charged offenses, all of which required that the Government prove a “knowing” or “willful” violation of the applicable law. 23 Both arguments fail. A. Negligence Instruction Under the Clean Water Act The Defendants were indicted on charges of felony, or “knowing,” violation of the CWA, 33 U.S.C. § 1319(c)(2)(A) 23 With the exception of the misdemeanor crime under the Clean Water Act, all of the charged crimes required that the government show that the Defendant knowingly engaged in the proscribed conduct. 55 (Counts 12-33). 24 However, late in the trial, the Defendants requested that the District Court also instruct the jury on the 24 Section 1319(c)(2) reads: (2) Knowing violations. Any person who— (A) knowingly violates . . . [this Act] . . . , or any permit condition . . . , or any requirement imposed in a pretreatment program . . . or in a permit issued under [this Act] . . . by the Secretary of the Army or by a State . . . . or (B) knowingly introduces into a sewer system or into a publicly owned treatment works any pollutant or hazardous substance which such person knew or reasonably should have known could cause personal injury or property damage or, other than in compliance with all applicable Federal, State, or local requirements or permits, which causes such treatment works to violate any effluent limitation or condition in a permit issued to the treatment works under [this Act] by the Administrator or a State; 56 lesser-included misdemeanor offense, which penalizes negligent violations of the Act, or of any permits issued under the Act. 25 Specifically, they requested that the District Court shall be punished by a fine of not less than $5,000 nor more than $50,000 per day of violation, or by imprisonment for not more than 3 years, or by both. 25 Section 1319(c)(1) reads: (1) Negligent violations. Any person who— (A) negligently violates [provisions of the CWA], or any permit condition or limitation . . . in a permit issued under [the CWA] by the Administrator or by a State, or any requirement imposed in a pretreatment . . . ; or (B) negligently introduces into a sewer system or into a publicly owned treatment works any pollutant or hazardous substance which such person knew or reasonably should have known could cause personal injury or property damage or, other than in compliance with all applicable Federal, State, or local requirements or permits, which 57 instruct the jury that a “person negligently violates the Clean Water Act by failing to exercise the degree of care that someone of ordinary prudence would have exercised in the same circumstances.” ((J.A. 5629) (Defendants’ proposed jury instructions (citing United States v. Hanousek, 176 F.3d 1116 (9th Cir. 1999))). The District Court agreed to charge the jury with the lesser included offense, as it was required to do. See Keeble v. United States, 412 U.S. 205, 208 (1973) (“[T]he defendant is entitled to an instruction on a lesser included offense if the evidence would permit a jury rationally to find him guilty of the lesser offense and acquit him of the greater.”). The Court’s instruction to the jury on the misdemeanor provision of the CWA read, in relevant part, as follows: [Y]ou may still consider whether the government has proven beyond a reasonable doubt that defendant is guilty of the offense of negligently violating that Act. . causes such treatment works to violate any effluent limitation or condition in any permit issued to the treatment works . . . ; shall be punished by a fine of not less than $2,500 nor more than $25,000 per day of violation, or by imprisonment for not more than 1 year, or by both. 58 . . “Negligence” may be defined as a failure to exercise, in the given circumstances, that degree of care for the safety of others which a person of ordinary prudence would exercise under similar circumstances. It may be the doing of an act which the ordinary prudent person would not have done, or the failure to do that which the ordinary prudent person would have done, under the circumstances then existing. (J.A. 7209.) The Court then made it clear that “[t]he Clean Water Act [was] the only offense in th[e] [i]ndictment that can be violated negligently. All of the other offenses require knowledge and/or willfulness.” Id. Despite having initially requested the lesser offense charge and recommended the simple negligence language to the Court, the Defendants now object to the Court’s instruction on the level of mens rea required to support a conviction under § 1319(c)(1) of the CWA, because they say the misdemeanor provision requires a showing of gross, rather than simple, negligence. They rely on the argument that the common law, the Model Penal Code and, as of 2004— after this trial was completed—the Third Circuit Model Jury Instructions all define criminal negligence to require a 59 showing of gross negligence. 26 In response, the Government argues that the Defendants are barred from bringing this challenge under the “invited error doctrine,” which provides that a “defendant cannot complain on appeal of alleged errors invited or induced by himself.” United States v. Console, 13 F.3d 641, 660 (3d Cir. 1993) (citation and internal quotation marks omitted). Because Defendants requested almost the exact instruction given, and made no request that the Court instruct on gross negligence, rather than simple negligence, the Government contends that the Defendants cannot now attack the District Court’s instructions. Alternatively, the Government argues—and the Defendants concede—that plain error review applies, since the Defendants did not challenge the instructions at trial. The Government asserts that, because other circuits have stated that simple negligence is the appropriate standard under the Clean Water Act, the District Court’s decision to rely on those cases was not plainly erroneous. 26 The Model Penal Code defines criminal negligence as a failure to perceive “a substantial and unjustifiable risk” of the harm that occurred. Model Penal Code, § 2.02(2)(d) (defining mental state for criminal negligence). The Third Circuit Model Criminal Jury Charges adopt the Model Penal Code definition, and via the commentary, note that the gross negligence instruction “should be used when the federal crime charged includes negligently as an element of the offense. This happens most often in regulatory type offenses, such as violations of environmental laws.” Third Circuit Model Criminal Jury Charges, § 5.09, Commentary (2004). 60 1. Invited Error Doctrine Under the invited error doctrine, “[a] defendant cannot complain on appeal of alleged errors invited or induced by himself.” Console, 13 F.3d at 660 (citation and internal quotation marks omitted). “[W]here a defendant makes a request in favor of certain instructions, he waives the right to complain of error in such instructions on appeal.” United States v. Andrews, 681 F.3d 509, 517 n.4 (3d Cir. 2012). However, we have held that “[w]here a defendant submits proposed jury instructions in reliance on current law, and on direct appeal that law is declared constitutionally infirm, we will not apply the invited error doctrine. Instead, we will review for plain error.” United States v. W. Indies Transp., Inc., 127 F.3d 299, 305 (3d Cir. 1997) (citing Johnson v. United States, 520 U.S. 461, 465-66 (1997)); Andrews, 681 F.3d at 517 n.4 (describing exception as applying to instances in which the law relied on by defendant is found to be “constitutionally problematic”). It is uncontested in this case that the District Court’s jury instruction defined the standard for negligence under § 1319(c)(1) of the CWA consistent with the Defendant’s own proposed language. Compare J.A. 5629 (Defendants’ proposal) with J.A. 7209 (District Court’s instruction). Thus, in order to except themselves from the application of the invited error doctrine, which otherwise bars their claim, Defendants must show that the law they relied on in proposing their instruction for simple negligence has since been found to be constitutionally problematic. See Andrews, 681 F.3d at 517 n.4. Defendants argue that they qualify for this exception and that, in the alternative, special circumstances warrant departure from the invited error bar. 61 Because we find no qualifying change in the law on this issue and no such special circumstances, we disagree. a) Simple Negligence Under the Clean Water Act The Clean Water Act does not, itself, define the appropriate standard for negligence under § 1319(c)(1). Nor has this Circuit weighed in on this issue. In proposing the simple negligence instruction to the District Court, the Defendants relied on the Ninth Circuit’s holding in Hanousek, which held that the appropriate mens rea under § 1319(c)(1) was simple, rather than gross, negligence. 176 F.3d at 1121. The Ninth Circuit began with the premise that Congress, in drafting statutes, intends to rely on the ordinary meaning of the words it uses. Id. at 1120. Noting that Congress, in the CWA’s civil provisions allowing increased civil penalties against violators who demonstrated “gross negligence” or “willful misconduct,” 33 U.S.C. § 1321(b)(7)(D), used the phrase “gross negligence” when it meant to do so, the Hanousek court concluded that Congress deliberately set forth a lesser standard in § 1319(c)(1). See id. at 1121. It also noted that Congress passed the Clean Water Act as a form of public welfare legislation, allowing it to “render criminal ‘a type of conduct that a reasonable person should know is subject to stringent public regulation and may seriously threaten the community’s health or safety.’” Id. at 1121 (quoting Liparota v. United States, 471 U.S. 419, 433 (1985)). Indeed, other Circuits have since concluded that simple, rather than gross, negligence is the appropriate mens rea for a misdemeanor violation of the CWA under § 1319(c)(1). In 2005, in United States v. Ortiz, the Tenth 62 Circuit relied on the plain meaning of the statutory text to determine that the use of “negligently” in § 1319(c)(1) referred to ordinary negligence—“a failure to exercise the degree of care that someone of ordinary prudence would have exercised in the same circumstance.” 427 F.3d 1278, 1283 (10th Cir. 2005). Though it noted that its decision was in accord with that of the Ninth Circuit in Hanousek, the Tenth Circuit believed that the plain meaning of the statute was clear on its face. Id. at 1283. More recently, in United States v. Pruett, the Fifth Circuit rejected a defendant’s argument that the district court had erred in defining the appropriate mens rea under § 1319(c)(1) as ordinary negligence, rather than gross negligence. 681 F.3d 232, 242 (5th Cir. 2012) (per curiam). Noting that negligence has a “plain and unambiguous meaning,” the Fifth Circuit found itself bound by § 1319(c)(1)(A)’s failure to specifically reference “gross negligence” as the appropriate mens rea, and therefore concluded that “§1319(c)(1)(A) requires only proof of ordinary negligence.” Id. The Pruett court, like the Ninth Circuit in Hanousek, noted that this plain text reading was bolstered by the fact that, where Congress had intended to require gross negligence, it had explicitly stated so in the plain language of the statute. Id. at 242 n.5. b) Whether Hanousek has been Constitutionally Invalidated The Defendants argue that the reasoning of Hanousek, and by extension, the reasoning of circuits that have reached similar conclusions, have been clearly invalidated by the Supreme Court’s 2007 decision in Safeco Insurance Company of America v. Burr, 551 U.S. 47 (2007). Thus, they argue, the invited error doctrine does not apply to bar their argument on appeal. 63 In Safeco, the Supreme Court held that, under the civil enforcement provisions of the Fair Credit Reporting Act (“FCRA”), 15 U.S.C. § 1681n(a), an insurer “willfully” disregards his notice obligations to consumers if he acts in “reckless disregard” of those obligations. 551 U.S. at 52. The defendant insurance companies had argued against liability since, under the criminal provisions of the FCRA, willful conduct excludes mere recklessness and that the civil provisions should be read to include identical limitations. Id. at 60. The Supreme Court rejected this argument, stating that “willfully” for the purposes of the civil provisions, was broader and reached conduct that demonstrated a “reckless disregard” for its consequences. Id. In reaching that conclusion, the Court relied heavily on the structure of the FCRA and, specifically, on the fact that Congress often explicitly uses modifiers such as “willfully” in the criminal context in order to “narrow[] the otherwise sufficient intent, making the government prove something extra, in contrast to its civil law usage.” Id. It held that, on the facts and statutory scheme before it, “[t]he vocabulary of the criminal side of FCRA is consequently beside the point in construing the civil side.” Id. Whatever the import of the Safeco holding for future case law addressing the Clean Water Act, we are not persuaded that its holding so clearly overrules or undermines the Ninth Circuit’s decision in Hanousek so as to except the Defendants from the invited error doctrine. See W. Indies, 127 F.3d at 305 (applying the exception where “on direct appeal that law is declared constitutionally infirm” (emphasis added)). Safeco was not a case of constitutional dimension. It dealt exclusively with the FCRA, and with the structure, intricacies and legislative history of that statute. Nor did the 64 Court’s analysis indicate that it addressed broader principles of statutory construction relevant to this case. The Safeco Court addressed what it found to be the problems created by using the definition of “willful” conduct from the stricter, criminal side of the FCRA in order to interpret that same phrase, as it appeared in the civil, remedial side of the statute. 551 U.S. at 57 n.9. Noting that “willful” is a word of many meanings, the Court stated that those meanings are often different in the criminal and civil contexts. In the criminal law, it is often intended to heighten the burden on the prosecution in proving its case while, on the civil side, “use of the term . . . typically presents neither the textual nor the substantive reasons” for imposing such a high burden on the plaintiff. Id.; see also id. at 57 (“[W]here willfulness is a statutory condition of civil liability, we have generally taken it to cover not only knowing violations of a standard, but reckless ones as well.”). Here, however, we have a different statute, a different set of circumstances, and a different means of statutory interpretation. Notably, the Hanousek, Ortiz and Pruett courts reached their conclusions relying primarily on the plain text of § 1319(c)(1), in keeping with the canon of construction which provides that courts must follow the plain and unambiguous text of the statute when interpreting and applying criminal laws. Neither Hanousek, nor the cases which followed, attempted to import a definition from outside the criminal provision before them. Rather, each court concluded that negligence has an ordinary and well-accepted meaning, and that, where Congress intended to alter that meaning to demand a showing of “gross negligence,” it did so explicitly. See Pruett, 681 F.3d at 242. 65 Certain portions of the Defendants’ brief also imply that the Model Penal Code and our own Model Criminal Jury Instructions indicate that the reasoning of Hanousek has been invalidated. See Joint Reply 11. The Third Circuit Model Criminal Jury Instructions—published in 2004, after the District Court’s instruction to the jury in this case— recommend an instruction of “gross negligence” for crimes that include negligence as an element of the offense. See Third Circuit Model Criminal Jury Charges, § 5.09, Commentary (2004). However, the Model Instructions are not-binding on this, or any, court. They thus cannot invalidate the decisions of this Circuit or others. Moreover, we note that our recommended instruction at § 5.09 relies on the Model Penal Code § 2.02(2)(d). While our model charges may have post-dated the District Court’s handling of this case, the Model Penal Code existed at the time that the Defendants drafted their proposed instruction under § 1319(c)(1). Thus, Defendants could easily have relied on the Model Penal Code in arguing for a higher standard before the District Court. 27 They did not do so. This is the very situation that the invited error doctrine is intended to address. In short, Hanousek, and the cases which followed it, remain good law. We cannot say that they stand for 27 At argument, the Defendants’ counsel insisted that, because Hanousek was the only case law available at the time of the trial, the Defendants could not have “legitimately push[ed]” for a gross negligence charge under § 1319(c)(1). Oral Arg. Trans., No. 09-2305, at 32 (Mar. 29, 2012). Of course, the decisions of other circuits, while persuasive, are not binding on the district courts in this Circuit. 66 principles which have been “declared constitutionally infirm.” W. Indies, 127 F.3d at 305; see also Andrews, 681 F.3d at 517 n.4. We therefore conclude that the invited error doctrine should apply to bar the Defendants’ argument on appeal. Moreover, even were we to excuse the application of the invited error doctrine, the Defendants’ argument would still fail under plain error review. To find plain error in a court’s jury instructions, we must find that “(1) there was error; (2) the error was clear or obvious; (3) the error affected the defendant’s substantial rights; and (4) the error seriously affected the fairness, integrity, or public reputation of the legal proceeding.” Tyson, 653 F.3d at 211. An error is “clear or obvious” only where the applicable law at issue has been clarified by the time of the appeal. Riley, 621 F.3d at 323. As stated previously, Defendants cannot demonstrate that there was a clear error in this case. The Defendants asserted, at oral argument, that if one “analyze[s] the later developing law, and you look at it in the context of what the Model Penal Code was back then,” one realizes that it should have been plain to the Court that the simple negligence instruction was in error. Oral Arg. Trans., No. 09-2305, at 31, (Mar. 29, 2012). We disagree. At the time that the District Court issued its instruction, the Ninth Circuit had concluded that simple negligence was the appropriate standard of mens rea under § 1319(c)(1) of the CWA. Hanousek, 176 F.3d 1116. No circuit had held otherwise. Moreover, since then, relying on textual interpretation alone, the Tenth Circuit has reached the same conclusion. See Ortiz, 427 F.3d 1278; see also Pruett, 681 F.3d at 232. Most telling, however, is the fact that this Court has not yet reached a contrary conclusion. And though we need not reach this issue today, we note that we are now 67 confronted with a slowly expanding body of law from our sister circuits which indicates that simple or ordinary negligence may be the appropriate standard of mens rea under § 1319(c)(1). In sum, the supposed error, if any, was anything but “clear and obvious.” Tyson, 653 F.3d at 211. Thus, notwithstanding that they invited the error they now claim was plainly erroneous, the Defendants’ argument that the District Court erred in defining the appropriate level of mens rea for negligence under § 1319(c)(1) of the CWA must fail. B. The District Court’s Refusal to Define “Recklessness” The Defendants argue that the concept of “negligent conduct” in this case did “double duty” as “it was the basis for misdemeanor liability under the CWA, and at the same time, it served as a defense to all of the felony charges,” since they required a showing of knowing or willful conduct. Joint Def. Br. 61. To that end, after asking the Court to instruct the jury on the lesser-included negligence charge under the CWA, the Defendants requested that the Court caution the jury that “negligence is not a valid theory of liability” on the remaining felony counts. (J.A. 5646.) The District Court granted this request in light of the fact that negligence was the basis for conviction under § 1319(c)(1)—the misdemeanor offense under the Clean Water Act. The Defendants also asked the District Court to include language instructing the jury on the definition of recklessness, and explaining that a showing of recklessness was insufficient to demonstrate knowing conduct. Their proposed instructions therefore juxtaposed acts done “knowingly and willfully”—sufficient to render a felony 68 conviction—with acts done “negligently or recklessly.” 28 D. Ct. Op. 36. Though the District Court initially included the Defendants’ proposed language on recklessness in its jury instructions, it deleted that language from later drafts and, ultimately, from the final version. The Court reasoned that recklessness was not an element of any of the charged crimes, 28 For example, on the false statement charges, the Defendants proposed that the Court instruct the jury that “[t]he false statement at issue must have been a knowing and willful false statement instead of a mere negligent or reckless one.” D. Ct. Op. 36 (discussing Defendants’ proposed jury instructions). For the Clean Water Act violations, the Defendants proposed the following language: To find a “knowing” violation of the Clean Water Act, you must find . . . that a defendant knew that he was discharging petroleum-contaminated wastewater and knew that the discharge was in violation of the authorized limits of the water permits. Negligence or recklessness is not sufficient to satisfy the requirement of a knowing violation. Id. 69 and that discussing it risked confusing the jury in an already complex case and was not essential to the Defendants’ ability to argue their defense. See J.A. 5768. The Court also raised concerns that the instruction, were recklessness included, might actually misstate the applicable law on “willful” conduct. The Defendants now challenge the Court’s refusal to include their recklessness instruction, arguing that the Court’s instruction that negligence was insufficient to convict on the felony counts, combined with its refusal to instruct on gross negligence and recklessness, “erroneously defin[ed] ‘knowingly’ under the felony counts” and “infected the jury’s understanding and evaluation of [their] mens rea defense.” Joint Def. Br. 61, 71. 1. Accuracy of the Court’s Instructions on “Knowing” Conduct We review the legal accuracy of a district court’s jury instructions de novo. United States v. Hoffecker, 530 F.3d 137, 173-74 (3d Cir. 2008). Absent an affirmative misstatement of the applicable law, our review is for an abuse of discretion. Id. at 174. A trial judge retains broad discretion in this regard, so long as the court’s instructions “‘fairly and adequately submit [] the issues in the case to the jury.’” United States v. Petersen, 622 F.3d 196, 203 (3d Cir. 2010) (alteration in original) (quoting United States v. Hart, 273 F.3d 363, 373 (3d Cir. 2001)). In its preliminary instructions to the jury, the Court was clear that “a person acts ‘knowingly’ if that person acts voluntarily and intentionally and not because of mistake or accident or other innocent reason.” (J. A. 7149). In instructing the jury at the close of the trial, the District Court 70 expounded on its earlier definition, providing separate mens rea charges for each offense in the indictment, and defining “knowing” conduct in the context of each charge. For example, in defining “knowingly and willingly” for purposes of the conspiracy charge under 18 U.S.C. § 371 (Count 1), the Court explained: A person acts ‘knowingly’ if that person acts voluntarily and intentionally and not because of mistake or accident or other innocent reason. . . . It is also the law that a person cannot be convicted of conspiracy if the state of mind of the defendant was in the nature of negligence. . . . I repeat that a defendant cannot be convicted of conspiracy . . . based on a state of mind that does not rise to the level of knowing and willful participation in the conspiracy. A person acts ‘willfully’ if that person acts voluntarily and with the specific intent or purpose to do something the law forbids or with the specific intent to omit something the law requires that person do; that is to say, with bad purpose either to disobey or disregard the law. (J.A. 7195.) For the instruction on the false statement offenses, the Court explained that the government must prove 71 that the defendants “acted willfully, with knowledge of the statement’s falsity,” as “opposed to . . . innocently, unintentionally, or even negligently.” ((J.A. 7195, 7197-98) (charging the jury under 18 U.S.C. § 1001 (Counts 2-7))). On the obstruction of justice charges, it explained that a guilty verdict was appropriate only if the jury found that the defendants had acted “knowingly and dishonestly,” “with specific intent to subvert or undermine” a federal investigation. ((J.A. 7201) (charging the jury under 18 U.S.C. § 1505 (Counts 8-10)); see also J.A. 7203 (requiring a showing of “intent to impede, obstruct and influence” the OSHA investigation under 18 U.S.C. § 1519 (Count 11)). In instructing the jury on the felony CWA violation, the Court made clear that the evidence had to demonstrate that the Defendants discharged the wastewater “intentionally and not as the result of ignorance, mistake or accident;” that they “knew the nature of the material discharged;” and that they “knew the discharge was in violation of” their CWA permits. (J.A. 7208.) Similarly, in instructing the jury on the CAA counts, the Court instructed the jury that it must find that the defendants “knew that an amount of waste paint in excess of 55 gallons per day was being burned in the cupola,” and that they “knowingly participated in causing that amount to be burned.” ( J.A. 7213.) We note at the outset that these proffered instructions are consistent with our own case law and our recommended jury instructions concerning “knowing” conduct. See W. Indies, 127 F.3d at 310 (noting, in the context of the CWA, that “[a]n act is done knowingly if done voluntarily and intentionally, and not because of mistake or accident or other innocent reason.”); see also United States v. Flores, 454 F.3d 149, 160-61 (3d Cir. 2006) (approving similar “knowing” 72 instruction); Third Circuit Model Criminal Jury Instruction 5.02. Indeed, the Defendants do not contest that the instructions were accurate statements of the law governing “knowing” conduct. Rather, they argue that the District Court’s failure to include their additional proposed language regarding recklessness effectively lowered the standard for “knowing” conduct in the eyes of the jury. The Defendants argue that this entitles them to de novo review. However, fundamentally their argument is that the District Court’s refusal to include their proposed language confused the jury and undermined their ability to present their defense. We review such challenges for an abuse of discretion. Hoffecker, 530 F.3d at 176. 2. Abuse of Discretion in Rejecting Proposed Language A court’s “refusal to give a proposed instruction” constitutes an abuse of discretion “only if the omitted instruction is correct, is not substantially covered by other instructions, and is so important that its omission prejudiced the defendant." United States v. Piekarsky, 687 F.3d 134, 142 (3d Cir. 2012) (citation and internal quotation marks omitted); see also Hoffecker, 530 F.3d at 176 (“[A] defendant is entitled to a theory-of-defense instruction if (1) he proposes a correct statement of the law; (2) his theory is supported by the evidence; (3) the theory of defense is not part of the charge; and (4) the failure to include an instruction of the defendant’s theory would deny him a fair trial” (citation and internal quotation marks omitted)). “[A] defendant is not entitled to a judicial narrative of his version of the facts, even [if] such a narrative is, in one sense of the phrase, a ‘theory of defense.’” United States v. Friedman, 658 F.3d 342, 353 (3d Cir. 2011) 73 (quoting Hoffecker, 530 F.3d at 176) (internal quotation marks omitted). The Defendants argue that the Court’s instructions, in defining “knowing” conduct and stating that negligence was not sufficient to convict under the felony counts, “failed to define the intermediate mental states between simple negligence and willful or knowing conduct.” Joint Reply 62. As a result, they argue, “the jury was forced to find [them] guilty of the felonies if their conduct rose even minimally above civil negligence.” Id. at 65. We disagree. We have never held that a court must define a requisite mental state by defining and explicitly excluding all of the mental states that do not meet the threshold. See Friedman, 658 F.3d at 354; Hoffecker, 530 F.3d at 177; cf. Hanousek, 176 F.3d at 1124 (holding under the CWA that the district court did not err in denying defendant’s request to further define the culpable mental state by explaining what it did not include). The jury need only receive those instructions necessary to its understanding of what conduct will suffice to support a conviction. Cf. Peterson, 622 F.3d at 203. That standard is met here. In light of the specificity of each instruction and the overall effect of these instructions read as a whole, we can only conclude that it was clear to the jury at the time of deliberation that only knowing, intentional conduct would suffice to render a conviction under each of these felony counts. Anything below that threshold required an acquittal. Moreover, even assuming that the Defendants can demonstrate that their proposed instruction was correct and supported by the record—assertions called into question by the District Court’s well-reasoned post-trial memorandum, D. Ct. Op. 85-86—they cannot substantiate their claims that the 74 Court’s omission of their proposed language foreclosed their theory of the defense or worked a substantial prejudice against them. See Piekarsky, 687 F.3d at 142; Hoffecker, 530 F.3d at 177 (considering whether the theory of the defense was implicit in the charge and whether the omission denied defendant of a fair trial). Because the District Court’s instructions made clear that a showing of anything less than “knowing” or intentional conduct would not support a felony conviction, any instruction on recklessness, as requested by the Defendants, would have been duplicative of the clear mens rea standards set forth by the District Court. See Hoffecker, 530 F. 3d at 177 (noting that instructions on “lack of intent” as a theory of defense were duplicative of the district court’s instructions that knowing and willful conduct was required). The Defendants were thus afforded ample room to argue to the jury that, “even if certain acts took place, at worse [sic] the activity was grossly negligent, or even reckless, but was not intentional,” Joint Def. Br. 65. See Friedman, 658 F.3d at 354; Hoffecker, 530 F.3d at 177; Hanousek, 176 F.3d at 1124. Recklessness, as a standard of mens rea, was never explicitly at issue in this case. The District Court decided not to introduce that concept out of fear of confusing the jury, because the judge doubted its soundness under the law and thought it was already addressed adequately by her discussion of “knowing,” “willful” and “intentional” conduct. That determination was well within the judge’s discretion. We therefore affirm the jury instructions provided by the District Court. 75 IV. Mutually Exclusive Verdicts Though charged with felony violations of the CWA, 33 U.S.C. § 1319(c)(2), Defendants Prisque and Davidson were convicted only of lesser-included misdemeanor or negligent violations under § 1319(c)(1)(A) (Counts 12-27).29 The jury also found that Prisque and Davidson had knowingly and willfully participated in a conspiracy with the specific objective of violating the CWA (Count 1). The Defendants now argue that these two verdicts are mutually exclusive and that we should thus vacate their conspiracy convictions (Count 1) on these grounds. We will affirm. A defendant seeking acquittal for inconsistent verdicts has a tough row to hoe. Consistency of verdicts is not necessary in a criminal trial. United States v. Vastine, 363 F.2d 853, 854 (3d Cir. 1996). Rather, where a jury convicts on one count and acquits on another, in most circumstances, “the most that can be said . . . is that the verdict shows that . . . the jury did not speak their real conclusions” on one of the convictions. Dunn v. United States, 284 U.S. 390, 393 (1932) (quoting Steckler v. United States, 7 F.2d 59, 60 (2d Cir. 1925)). As a matter of law, courts treat this situation as an instance of juror lenity, undeserving as a basis for overturning a conviction. Vastine, 363 F.2d at 855 (“That the verdict may 29 Defendant Maury was also charged with violations of the Clean Water Act (Counts 27, 28-33). Though the jury convicted him of the lesser charge concerning the December 1999 pumping and spill (Count 27), it convicted him of the felony counts concerning the pumping of the Number Four Pit (Counts 28-33). 76 have been the result of compromise, or of a mistake on the part of the jury, is possible. But verdicts cannot be upset by tears” into such matters. (citation and internal quotation marks omitted)); see also Harris v. Rivera, 454 U.S. 339, 346 (1981) (noting “the unreviewable power of a jury to return a verdict of not guilty for impermissible reasons”). In United States v. Powell, the Supreme Court determined that these same principles apply when a jury convicts on a conspiracy charge, but acquits on the underlying overt act, 469 U.S. 57, 65 (1984). Conceding that the rule yields sometimes unsatisfying results, the Court held that it applied nonetheless. Id. (citing Dunn, 284 U.S. at 393). To argue that such inconsistencies should receive special treatment, particularly where the district court’s instructions were correct, “simply misunderstands the nature of the inconsistent verdict problem.” Id. at 68. “Whether presented as an insufficient evidence argument, or as an argument that the acquittal on the predicate offense should collaterally estop the Government on the compound offense, the argument necessarily assumes that the acquittal on the predicate offense was proper—the one the jury ‘really meant.’” Id. While inconsistent verdicts might be the product of juror mistake, compromise, or lenity, determining which requires the sort of impermissible speculation into the reasoning of a jury’s decision that courts have historically avoided. Id. at 66. Moreover, the Court wrote, when inconsistent verdicts present themselves, “‘error,’ in the sense that the jury has not followed the court’s instructions, most certainly has occurred, but it is unclear whose ox has been gored.” Id. at 65. Just as the Government has lost on one count, a defendant has received the advantage of an acquittal on the other. Id. at 69. 77 However, though Powell explicitly foreclosed a defendant’s ability to challenge an inconsistent verdict involving a conviction of a conspiracy and an acquittal on a predicate act, the Court stated in a footnote that nothing in that decision was intended to decide the “proper resolution of a situation where a defendant is convicted of two crimes, where a guilty verdict on one count logically excludes a finding of guilt on the other.” Id. at 69 n.8 (citing United States v. Daigle, 149 F. Supp. 409 (D.D.C. 1957)). 30 Since Powell, this Circuit has had two occasions to determine the meaning of the Supreme Court’s footnote in Powell. In United States v. Gross, we were confronted with a defendant who challenged a jury verdict that acquitted him of knowingly causing the filing of a false statement with the Securities and Exchange Commission, but convicted him of insider trading and mail fraud. 961 F.2d 1097 (3d Cir. 1992). The defendant argued that Powell did not apply, since in his case the two verdicts both involved the same mental state: “knowingly and willfully.” Id. at 1106-07. We disagreed, 30 In Daigle, the district court invalidated a jury’s guilty verdict for larceny when the jury also convicted the defendant of embezzlement. Because embezzlement involves the conversion of property that a defendant rightfully had in his possession, but did not own, and larceny involves the carrying away of a good that the defendant had no right to possess in the first place, the two convictions could not be reconciled. The district court therefore struck the larceny verdict, stating that “where a guilty verdict on one count negatives some fact essential to a finding of guilty on a second count,” the two verdicts cannot both stand. Id. at 414. 78 noting that the verdicts were not necessarily inconsistent, since a jury could have decided to acquit because one element of the acquitted crime was not supported by sufficient evidence, even if the knowledge requirement was satisfied as to both. Id. at 1106 (“For example, . . . the government may have failed to make the required showing that [the defendant] ‘caused’ the false statements to be filed.”). We made clear that we interpreted Powell’s exception to narrowly apply to only “those situations where a jury has convicted a defendant of two crimes and those convictions are mutually exclusive”—or, put differently, where the defendant was “convicted of two crimes, at least one of which he could not have committed.” Id. at 1107. Several years later, in Buehl v. Vaughn, we relied on Gross for the proposition that “logically incompatible guilty verdicts may not stand” in considering the claims of a habeas petitioner who argued that his trial counsel had been ineffective in failing to object to purportedly inconsistent verdicts. 166 F.3d 163, 178 n.11 (3d Cir. 1999) (Alito, J.). The petitioner had been charged with first-degree murder, third-degree murder and involuntary manslaughter, all based on the same conduct. Id. at 177. The jury convicted the Defendant on all counts, and the petitioner argued on appeal that this demonstrated that the jury had been confused as to what level of mens rea was supported by the evidence. Id. at 177-78. Thus, the Buehl court considered whether, under Gross and Powell, a defendant could argue that two convictions based on the same conduct and same evidence could be “logically inconsistent” where one was essentially a lesser-included offense of the other. We held that he could not because, “although involuntary manslaughter requires only recklessness or gross negligence, that element may be satisfied by proof that the defendant intentionally killed the 79 decedent, as the first degree murder statute requires.” Id. at 179 (internal footnote omitted); see also id. at 178 (quoting Masoner v. Thurman, 996 F.2d 1003, 1005 (9th Cir. 1993) for the proposition that a verdict will be upheld if “based on the evidence presented to the jury[,] any rational fact finder could have found a consistent set of facts supporting both convictions”). Therefore, in light of Powell, as explicated in Gross and Buehl, a defendant may only challenge dual guilty verdicts that are inherently and fundamentally at odds with one another. Or, to put it differently, a conviction as to one of the crimes must negate an element of the other. Moreover, the rule remains that, under Dunn, a defendant cannot challenge inconsistent jury verdicts if he was acquitted on one count and convicted on another. With these principles in mind, it is difficult to see how Defendants Prisque and Davidson can succeed with their argument. The jury found Prisque and Davidson guilty of multiple objectives underlying the conspiracy charged in Count 1 of the indictment. In addition to the negligence charge under the CWA, the jury also found Davidson guilty of knowingly making materially false statements to NJDEP investigators following the December 1999 oil spill. Prisque was found to have committed every offense charged as an objective of the conspiracy. Thus, even were we to find some degree of inconsistency between the conviction under §1319(c)(1) of the CWA and the conspiracy under 18 U.S.C. § 371, the conspiracy conviction could still stand based on the jury’s verdict on the remaining underlying offenses. When a defendant is charged with a multi-object conspiracy, the conviction will stand so long as the verdict as to any one of the underlying objectives of the conspiracy is sound. See 80 United States v. Navarro, 145 F.3d 580, 590-91 (3d Cir. 1998). Even were we to ignore this fatal flaw in the Defendants’ argument, the argument would still fail. Insofar as Prisque and Davidson attempt to argue that their acquittal on the felony charge under the Clean Water Act is inconsistent with their conviction for the conspiracy, that argument is foreclosed by Powell, supra; accord United States v. Mathis, 579 F.2d 415, 418 (7th Cir. 1978) (“Any arguable inconsistency arises not from the verdict of guilty of unarmed assault, but rather from the implicit verdict of not guilty of the offense of assault with a deadly or dangerous weapon.”). To the extent their argument is that the conviction for negligently violating the CWA under § 1319(c)(1) is inconsistent with the conspiracy conviction, one need only compare the elements of these crimes to see that there is no necessary inconsistency between the two. Though the CWA instructions here required the jury to find a lower mens rea, the “failure to exercise in the circumstances that degree of care for the safety of others which a person of ordinary prudence would exercise under similar circumstances,” our case law indicates that a showing of a higher, “knowing” violation would suffice to meet this requirement. See Buehl, 166 F.3d at 179-80. Alternatively, the conspiracy itself could predate the actual negligent violation, or vice versa. Evanchyk v. Stewart, 340 F.3d 933, 942 (9th Cir. 2003) (relying on this logic to uphold dueling guilty verdicts). Defendants attempt to avoid this seemingly foregone conclusion by arguing that, in this case, unlike in Buehl, “conspiracy and the substantive CWA violations were not a unitary offense,” but instead were “different counts, . . . involved different facts, different elements, and entirely 81 different statutory schemes.” Davidson Br. at 24. But these arguments neither render Buehl inapplicable nor help the Defendants’ case. Instead, they serve to emphasize why the two convictions here—the first of negligent violation of the CWA, and the second, of knowingly participating in a conspiracy—are not mutually exclusive. To hold otherwise would require the very sort of speculation and hypothesizing about the jury’s verdict that our case law clearly forbids. Thus, here, as in Powell, Defendants have been “given the benefit of . . . acquittal on the counts on which [they were] acquitted, and it is neither irrational nor illogical to require [them] to accept the burden of conviction on the counts on which the jury convicted.” Powell, 469 U.S. at 69; see also Vastine, 363 F.2d at 854 (upholding conviction on conspiracy despite acquittal of defendants on the underlying substantive offenses). Accordingly, we see no error in the District Court’s handling of this issue. V. Conclusion Having carefully considered the Defendants’ various remaining arguments, we find them to be without merit. 31 31 We note that the Company filed a letter, pursuant to Rule 28(j) of the Federal Rules of Appellate Procedure notifying us of the Supreme Court’s recent decision in Southern Union Co. v. United States, 132 S. Ct. 2344 (2012). We find that case inapplicable to the Company’s challenge to the criminal penalties imposed by the District Court. 82 We therefore affirm the final convictions, judgments and sentences of the District Court, in all respects. 83