[DOCID: f:rock.wais] ROCK OF AGES CORPORATION February 24, 1998 YORK 94-76-RM FEDERAL MINE SAFETY AND HEALTH REVIEW COMMISSION 1730 K STREET NW, 6TH FLOOR WASHINGTON, D.C. 20006 February 24, 1998 SECRETARY OF LABOR, : MINE SAFETY AND HEALTH : ADMINISTRATION (MSHA) : : v. : Docket Nos. YORK 94-76-RM : through 94-83-RM ROCK OF AGES CORPORATION : BEFORE: Jordan, Chairman; Marks, Riley, and Verheggen, Commissioners[1] DECISION BY THE COMMISSION: This consolidated contest and civil penalty proceeding, arising under the Federal Mine Safety and Health Act of 1977 ("Mine Act" or "Act"), 30 U.S.C. § 801 et seq. (1994), involves citations issued by the Mine Safety and Health Administration ("MSHA") against Rock of Ages Corporation ("ROA"). The citations at issue arose from an investigation into a fatal accident at ROA's Smith Quarry and involve 30 C.F.R. Part 56, Subpart E, which governs the use and handling of explosives at surface metal/nonmetal mines. Administrative Law Judge Jerold Feldman upheld the citations and assessed penalties totaling $180,000. 17 FMSHRC 1925 (November 1995) (ALJ). The Commission granted the petition for review filed by ROA in which it challenges the judge's decision that ROA violated the Mine Act as charged in the citations and that its violations were due to its unwarrantable failure. In addition, ROA challenges the penalty assessment and contends that it did not receive a fair trial because the judge was biased against ROA and its counsel. For the reasons that follow, we affirm the judge's findings of violations and determinations of unwarrantability; we reject ROA's allegations of bias; and we vacate the penalty assessments and remand them for further consideration. I. Factual and Procedural Background A. How Granite is Quarried ROA is a granite-quarrying company with twelve quarries and over 300 employees. 17 FMSHRC at 1927; Tr. II 453.[2] The Smith Quarry, where the fatal accident occurred, is a massive granite formation where granite is quarried through the removal of large benches in the shape of cubes. 17 FMSHRC at 1927. A typical bench is 40 feet wide, 35 feet deep and 16 feet high. Id. at 1929. In the beginning of the quarrying process, an automatic torch is used to burn a channel or seam, approximately 6 inches wide, around the top of a bench to separate the bench for blasting and quarrying. Id. The channel burning begins at the front face on top of a bench, proceeds across one side, across the back, across the opposite side, and ends back at the front face. Id. Burning the channels on the bench takes approximately 15 days. Id. After the channels are burned, a series of holes are drilled horizontally at intervals along the bottom of the base of the bench. Id. at 1930. The holes are approximately 1 7/8 inches in diameter and 32 feet long. The horizontal "lift" or detonation holes are completed after about 26 days into quarrying the bench. Id. Next, a series of vertical holes are drilled every 5 1/2 feet in the top of the bench to within about 1 foot of the horizontal lift holes. Id. The vertical holes are completed after about 34 days into the quarrying process. Id. The bench is separated from the quarry floor by detonating explosives loaded into the horizontal lift holes. Id. Not all the holes are loaded and different loading patterns - explosives in every third or fourth hole - are used. Id. With the exception of seven pyrodex shots from February to July 1993, ROA used seismic cord, which is a continuous rope-like cord of blasting material, placed in various lift holes, connected to a trunk line and detonated by blasting caps. Id. Following a detonation, the powderman, his assistant and a foreman go to the face of the bench to evaluate the success of the lift and check for any misfires. Id. They look for proper cracking from lift hole to lift hole, signs of discoloration from blast residue, and any evidence of non-detonated blasting materials. Id. The greatest danger with undetonated seismic cord arises when it breaks as it is being unrolled and pushed into the lift holes; however, the powderman generally can detect such a break because the roll of cord usually stops before reaching the back of the bench. Id. at 1931. The blasting process occurs after 35 days into quarrying the bench. Id. at 1930. After the detonation, the bench is further divided by driving shims and wedges into the vertical holes drilled at 5 1/2 foot intervals on top of the bench. Id. Slabs are split off from the main bench by a front-end loader with a boom. Id. As each slab is removed, new stone is exposed and the powderman conducts a further examination of the newly exposed area. Id. at 1930-31. When a lift is clean, the top half of the hole becomes part of the lifted bench, and the bottom half becomes the top of the bench to be quarried on the next level below. Id. at 1927. If the lift is not clean, "caprock," or a layer of rock over the lift hole, may remain in place on the surface after the bench is removed with the lift hole intact, sometimes with undetonated explosives inside. Id. at 1927-28. Similar to caprock, "underbreak" occurs when the rock underneath the lift holes cracks, leaving lift holes intact that are removed as the vertical slabs of the bench are quarried. Tr. II 528-29. The slabs are split into smaller pieces of granite that are transported from the bench to a loading area where they are lifted by a derrick out of the quarry. 17 FMSHRC at 1931. This process is repeated until the entire bench is removed, a process that takes about 10 to 12 days after the blasting. Id. In all, the entire bench is quarried in a period of about 47 days from the initial channel burning. Id. After the bench has been removed, the surface is cleaned and scraped with a loader and examined further. Tr. II 843-45. The granite surface underneath the removed bench then becomes the top of the bench that will be quarried in future years as the level of the quarry descends. B. The Use of Pyrodex as an Explosive Beginning in 1986, ROA tested the use of pyrodex as a substitute for seismic cord. 17 FMSHRC at 1932. The Hodgdon Powder Company manufactures pyrodex, which has the same chemical ingredients as black powder plus additional binders and burn rate modifiers. Id. at 1931. Pyrodex is generally referred to as a "replica" of black powder and is classified as a Class B explosive by the United States Department of Transportation ("DOT"), while black powder is a Class A explosive.[3] Id. at 1931, 1932. Black powder is a detonating explosive, while pyrodex is a propellant explosive. Id. at 1931. A propellant explosive generates gas and energy as it burns; a detonating explosive generates gas, energy and shock when it detonates. Id. at 1931-32. Pyrodex is regarded as having superior qualities in blasting rock without radial fracturing, which causes more waste. Id. at 1932. ROA performed several test detonations with pyrodex in 1986, but decided against further use of it because of the labor needed to pressurize the hole in which the pyrodex was used. Id. Pressurization, which could be accomplished through stemming the collar of the lift hole with rags or paper, was essential for proper lift. Id. Pressurization also channeled gas and energy to the back of the hole to ensure that other bags of pyrodex in the middle and rear of the hole would ignite following ignition of pyrodex in the front of the hole, which was accomplished with an electric "squib," sometimes referred to as a match. Id. A further test detonation in 1987 did not adequately split the granite bench, and ROA ceased experimenting with pyrodex. Id. Hodgdon Vice President Glenn Barrett supervised the test firings in 1986 and 1987. Id. In January 1993, ROA informed Hodgdon that it was interested in resuming use of pyrodex as an explosive. Id. 1933. Hodgdon Vice President Barrett sent ROA information concerning the use and handling of pyrodex. Id. Barrett recommended that pyrodex be used by spacing several bags of it in a pressurized 30 to 40- foot lift hole without using any connecting detonating cord. Id. The bag of pyrodex in the front of the lift hole was ignited by using the electric squib. Id. Succeeding bags were ignited by the first detonation if there were no obstructions in the hole. Id. From February to May 1993, ROA used pyrodex at the Smith Quarry and a second quarry. Id. Because ROA did not generally use pyrodex in quarrying, it prepared detailed blasting reports, specifying the location and dimensions of the bench at which pyrodex was used, the number and loading sequence of bags of pyrodex in each lift hole, and the number and location of lift holes that were loaded. Id. at 1933-34. On June 22, 1993, the pertinent report indicates that the 42-foot bench had 80 lift holes that were 37 feet deep and spaced at 6-inch intervals across the face. Id. at 1934. Four bags of pyrodex were placed in each fourth hole.[4] Id. One bag was placed in the front of each hole, one in the middle, and two in the back, approximately 32 to 37 feet away from the mouth of the hole. Id.; Ex. R-7. Neither blast foreman Earl Kelty nor powderman Bud Reynolds pressurized the lift holes in which pyrodex was used. Tr. I 588- 89. Following the blast, Kelty and Reynolds inspected the bench and reported their observations of the condition of the bench in the blast report. 17 FMSHRC at 1934. Other blast reports indicated significant variation in the number and sequence of lift holes loaded and the number of bags of pyrodex loaded into each hole. See Ex. R-7. C. Eventsz Surrounding the Fatal Accident Around July 1, 1993, derrick operator Earnest Batchelder was loading blocks of granite to be lifted out of the quarry when he saw four unexploded bags of pyrodex that were shaken loose and hanging from one of the blocks. 17 FMSHRC at 1934. There were no detonators attached to the bags. Id. at 1935. He reported the presence of the bags to Foreman Kelty. Id. Subsequently, Kelty noted on the report for the June 22, 1993 pyrodex blast that four bags of pyrodex had failed to detonate. Id.; Ex. R- 7. However, Kelty did not conduct any further search to check for the presence of more undetonated pyrodex. 17 FMSHRC at 1935-36. MSHA Inspector Donald Fowler testified without contradiction that he questioned Kelty during the investigation into the accident as to why he did not follow up on the discovery of the undetonated bags of pyrodex and Kelty responded that he "forgot." Id.; Tr. I 586-87. On May 20, 1994, channel burner operator Michael Bassett had completed a channel on the side of a bench and was initiating the burning of a channel at the rear of the bench. 17 FMSHRC at 1936. Unbeknownst to Bassett, his torch passed within two feet of a lift hole with two bags of unexploded pyrodex that remained from the detonation and quarrying of the bench directly above the new bench. Id. His torch then passed over three empty holes. Id. at 1937. Just before 11:00 a.m., Bassett's torch ignited two unexploded bags of pyrodex as he passed over another lift hole. Id. at 1936-37. Bassett was thrown 10 feet into the air and killed instantly. Id. at 1936. A subsequent search for more unexploded pyrodex on the bench where the accident occurred revealed seven more lift holes, each with two bags. Id. at 1937. A total of 22 bags of pyrodex were discovered - the four bags uncovered in 1993, the four bags that Bassett encountered on the day of the accident, and the 14 additional bags uncovered after the accident.[5] Id. Based on the loading sequence of the lift holes and the loading pattern within the holes, MSHA concluded that the accident location was at the base of a removed bench where ROA had used pyrodex on June 22, 1993. Id. The 22 misfires represented 26 percent of the 84 bags of pyrodex loaded into the lift holes for the June 22 detonation. Id. Following its accident investigation, MSHA issued four citations against ROA. Id. at 1928. MSHA charged ROA with violating 30 C.F.R. § 56.6311(b), for permitting work other than work necessary to remove a misfire in the blast area; 30 C.F.R. § 56.6306(g), for permitting work to resume prior to an adequate post-blast inspection following the June 22, 1993 detonation; 30 C.F.R. § 56.6904, for permitting an open flame within 50 feet of explosive material; and 30 C.F.R. § 56.6300(a), for inadequately training blasting personnel. Id. at 1928-29. MSHA alleged that the violations were significant and substantial ("S&S")[6] in nature and a result of ROA's unwarrantable failure. D. Judge's Decision Following an eight-day hearing, the judge issued his decision in which he affirmed the four citations and determined that the violations were S&S and occurred as a result of ROA's unwarrantable failure. 17 FMSHRC 1938-54. The judge assessed penalties totaling $180,000 based on the four violations and special findings. Id. at 1954. II. Disposition A. Whether Pyrodex Misfires Are Governed by Part 56 Regulations Before the judge, ROA argued that pyrodex was not an explosive and, therefore, not covered by the Part 56 regulations. 17 FMSHRC at 1938; ROA Post-Trial Br. at 9-11. ROA has abandoned this argument on review, contending instead that, because pyrodex burns rather than detonates, it does not fit within the definition of "misfire" in 30 C.F.R. § 56.6000.[7] ROA Br. at 26-27 n. 21. In response, the Secretary contends that pyrodex is an explosive under DOT regulations, 49 C.F.R. § 173.88 (1996), and to interpret the misfire regulations to exclude pyrodex would defeat their purpose. S. Br. at 47 n. 19. The Secretary argues that the regulatory history of the Part 56 regulations demonstrates her intent to include all explosives that fail to perform as planned. Id. at 47-48. Where the language of a regulatory provision is clear, the terms of that provision must be enforced as they are written unless the regulator clearly intended the words to have a different meaning or unless such a meaning would lead to absurd results. Dyer v. United States, 832 F.2d 1062, 1066 (9th Cir. 1987); see also Utah Power & Light Co., 11 FMSHRC 1926, 1930 (October 1989); Consolidation Coal Co., 15 FMSHRC 1555, 1557 (August 1993). If, however, a regulation is ambiguous, courts have deferred to the Secretary's reasonable interpretation of it. See Energy West Mining Co. v. FMSHRC, 40 F.3d 457, 463 (D.C. Cir. 1994); accord Secretary of Labor v. Western Fuels-Utah, Inc., 900 F.2d 318, 321 (D.C. Cir. 1990) ("agency's interpretation is `of controlling weight unless it is plainly erroneous or inconsistent with the regulation'") (quoting Bowles v. Seminole Rock Co., 325 U.S. 410, 414 (1945)). The Secretary's interpretation of a regulation is reasonable where it is "logically consistent with the language of the regulation[s] and . . . serves a permissible regulatory function." General Elec. Co. v. EPA, 53 F.3d 1324, 1327 (D.C. Cir. 1995) (alteration in original) (quoting Rollins Environmental Services, Inc. v. EPA, 937 F.2d 649, 652 (D.C. Cir. 1991). The Commission's review, like the courts', involves an examination of whether the Secretary's interpretation is reasonable. Energy West, 40 F.3d at 463 (citing Secretary of Labor on behalf of Bushnell v. Cannelton Indus., Inc., 867 F.2d 1432, 1439 (D.C. Cir. 1989)); see also Consolidation Coal Co., 14 FMSHRC 956, 969 (June 1992). Because the regulation at issue here is ambiguous, we defer to the Secretary's reasonable interpretation. The Secretary correctly notes that the regulatory definition of explosives includes pyrodex. The MSHA regulations defining "explosives" incorporate by reference regulations issued by DOT and its classification of explosives. 30 C.F.R. § 56.6000. These DOT regulations clearly include propellants as explosives. 49 C.F.R. § 173.59. Although ROA concedes that pyrodex is an explosive, it nonetheless argues that because pyrodex burns rather than detonates, the definition of misfire does not include pyrodex. ROA Br. at 26 n. 21. This interpretation, however, would create glaring inconsistencies in the regulations - pyrodex would be covered for some purposes of the MSHA regulations but not for others. Further, the regulatory history of the definition of "misfire" at 30 C.F.R. § 56.6000 indicates that the choice of the word "detonate" was not intended to carry with it any technical meaning but was substituted for the word "explode" to maintain "consistency" in terminology. See 56 Fed. Reg. 2070, 2073 (1991). Failing to interpret the regulations to include pyrodex would defeat the Mine Act's purpose of enhancing miner safety through the regulations. See Emery Mining Corp. v. Secretary of Labor, 744 F.2d 1411, 1414 (10th Cir. 1984). For these reasons, we affirm the judge's determination that pyrodex misfires are covered by the Secretary's Part 56 regulations. B. Section 56.6311(b)[8] 1. Violation ROA argues that it complied with subsection (b) when it properly disposed of the four misfires located by derrick operator Batchelder, noting that it was not cited under subsection (a) of the standard for failing to properly examine the blast site. ROA Br. at 27-28. ROA further argues that the judge misinterpreted the language of the regulation by requiring it to seek and remove misfires throughout the bench removal process. Id. at 28-29. ROA contends that the judge's interpretation and application of the regulation in requiring ROA to search for further misfires once the four misfires were found is contrary to the plain language of the regulation. Id. at 29- 32. Finally, ROA concludes that, because the citation was not based upon any work performed at the quarry prior to the disposal of the four bags of pyrodex discovered by Batchelder, the citation must be vacated. Id. at 32. The Secretary responds that the judge correctly accepted her interpretation of section 56.6311(b) that operators must properly handle and dispose of misfires about which they know or should know. S. Br. at 48. The Secretary further contends that the standard must be read in its entirety because the obligation to properly handle a misfire does not arise unless there has been an examination for misfires. Id. at 47-50 n. 19. Finally, the Secretary argues that her interpretation is consistent with the regulatory history and the protective purposes of the Act and that it would defeat the purposes of the Act and the standard if ROA were only required to dispose of known misfires when it has reason to believe there are other misfires. Id. at 50-51. The judge held that ROA violated the standard when it did not make a reasonable effort to locate and remove misfires at the quarry after Batchelder's discovery of the four bags of pyrodex. 17 FMSHRC at 1948. We agree with the judge that ROA violated the standard, but for reasons different than those relied on by the judge. The "language of a regulation . . . is the starting point for its interpretation." Dyer v. United States, 832 F.2d at 1066 (citing Consumer Product Safety Comm'n v. GTE Sylvania, Inc., 447 U.S. 102, 108 (1980)). Where the language of a regulatory provision is clear, the terms of that provision must be enforced as they are written unless the regulator clearly intended the words to have a different meaning or unless such a meaning would lead to absurd results. Id.; Utah Power & Light Co., 11 FMSHRC at 1930; Consolidation Coal, 15 FMSHRC at 1557. The citation at issue charged ROA with violating section 56.6311(b) by permitting "work not necessary to remove a misfire and protect the safety of miners" on the accident bench. Citation No. 4282251. The clear language of the standard restricts the work that can be performed in an area containing misfires. Barrett was killed while he was burning a channel when his torch made contact with unexploded bags of pyrodex. Channel burning does not constitute work "necessary to remove a misfire or protect the safety of miners," and therefore cannot be carried out in an area that contains misfires. Thus, ROA's authorization of Bassett's channel burning violated the plain terms of the regulation. ROA argues that the standard only applies "where a misfire has been found." ROA Br. at 32. The operator further contends that it fully complied with the standard because Kelty's inspection of the blast site immediately after the June 22 detonation disclosed no misfires,[9] and ROA properly disposed of the misfires that Batchelder subsequently found on July 1. Id. However these contentions are without merit. As the Commission has previously stated, "[t]he Mine Act is a strict liability statute and an operator may be held liable for violations without regard to fault." Wyoming Fuel Co., 16 FMSHRC 19, 21 (January 1994). The standard prohibits all but a specific category of work in areas containing misfires. ROA's interpretative gloss on the standard - that it only applies once a misfire has been found - is contrary to the clear language of the standard.[10] Whether or not ROA knew or had reason to know about the presence of the misfires at the accident bench, the prohibition applied to Bassett's activity. 2. Unwarrantable Failure[11] The unwarrantable failure terminology is taken from section 104(d) of the Act, 30 U.S.C. § 814(d), and refers to more serious conduct by an operator in connection with a violation. In Emery Mining Corp., 9 FMSHRC 1997 (December 1987), the Commission determined that unwarrantable failure is aggravated conduct constituting more than ordinary negligence. Id. at 2001. Unwarrantable failure is characterized by such conduct as "reckless disregard," "intentional misconduct," "indifference," or a "serious lack of reasonable care." Id. at 2003-04; Rochester & Pittsburgh Coal Co., 13 FMSHRC 189, 194 (February 1991); see also Buck Creek Coal, Inc. v. FMSHRC, 52 F.3d 133, 136 (7th Cir. 1995) (approving Commission's unwarrantable failure test). In addition, the Commission has relied upon the high degree of danger posed by a violation to support an unwarrantable failure finding. See BethEnergy Mines, Inc., 14 FMSHRC 1232, 1243-44 (August 1992) (finding unwarrantable failure where unsaddled beams "presented a danger" to miners entering area); Warren Steen Constr., Inc., 14 FMSHRC 1125, 1129 (July 1992) (finding violation aggravated and unwarrantable based upon "common knowledge that power lines are hazardous, and . . . that precautions are required when working near power lines with heavy equipment"); Quinland Coals, Inc., 10 FMSHRC 705, 709 (June 1988) (finding unwarrantable failure where roof conditions were "highly dangerous"). We agree with the judge that the violation of the standard was the result of ROA's unwarrantable failure. See 17 FMSHRC at 1948-49. The discovery of the four unexploded bags of pyrodex should have alerted foreman Kelty to the possibility of additional misfires, particularly in light of the experimental nature of the pyrodex blasting. As a supervisor, Kelty had to appreciate that any pyrodex misfires would pose an extreme danger to the miners in the area. As the judge noted (17 FMSHRC at 1948), unexploded bags of pyrodex would not lie harmlessly under piles of rock but would eventually be exposed as benches were removed and pose a hazard as channel burning resumed. As a supervisor, Kelty is held to a high standard of care. Midwest Material Co., 19 FMSHRC 30, 35 (January 1997). We agree with the judge that Kelty's failure to order any meaningful search for unexploded pyrodex, such as probing caprock at the blast site, "evidenced a callous disregard for the hazards associated with misfires." 17 FMSHRC at 1948; see also Midwest Material Co., 19 FMSHRC at 35 (foreman's negligent conduct in the face of obvious hazard indicates serious lack of reasonable care). Finally, ROA's failure to provide any explanation for its negligence and the lack of safety precautions that resulted in this fatal accident supports an unwarrantable failure determination. Midwest Material Co., 19 FMSHRC at 35. In short, the record evidence overwhelmingly supports the judge's determination that ROA's violation was the result of a lack of reasonable care greater than ordinary negligence serious enough to constitute aggravated conduct. C. Section 56.6306(g)[12] 1. Violation The judge affirmed the citation charging a violation of section 56.6306(g) for exposing miners to hazardous conditions, because a qualified person failed to inspect the blast site. 17 FMSHRC at 1949-50. The judge rejected ROA's argument that the regulation did not apply to areas blasted before January 31, 1994, the effective date of the regulation. Id. at 1950. The judge also rejected ROA's argument that it had already "resumed work" at the blast site in 1993 when it continued to quarry the June 22, 1993 bench. The judge held that resumption of work is a "continuing process" and that ROA did not escape liability by resuming work prior to implementation of the standard. Id. ROA's main argument is that the provision did not go into effect until January 31, 1994, well after the June 22, 1993 pyrodex blast, and that the standard may not be applied retroactively. ROA Br. at 32. ROA further argues that the provision only requires a single post-blast examination and that the judge interpreted it to require multiple examinations for a blast that occurred well before the effective date of the standard. Id. at 32-34. ROA contends that the standard fails to inform a "reasonably prudent quarry operator" that it applies to events prior to its effective date. Id. at 35. ROA argues that, even if the standard is applied retroactively, it resumed work after the 1993 blast, following a properly conducted post-blast inspection. Id. at 35-38. The Secretary responds that the violative conduct occurred at the time of the fatal accident on May 20, 1994, after the effective date of section 56.6306, when ROA failed to conduct an adequate post-blast examination. S. Br. at 53. The Secretary contends that ROA's interpretation of the regulation is contrary to its history and purpose. Id. at 53-54. The Secretary further argues that in the context of granite quarrying, section 56.6306(g) requires continuing inspections for misfires as new rock is exposed. Id. at 54. The Secretary notes that the preamble to section 56.6306(g) explained that the subsection requires "post-blast examinations to minimize hazards to persons who will perform subsequent work in the area." Id. at 55 (citing 55 Fed. Reg. 69,596, 69,603 (1993)). The Secretary contends that the duty to inspect included a duty to probe lift holes under caprock. Id. at 57. Section 56.6306(g) was issued in January 1991, along with other regulations dealing with explosives at surface metal/nonmetal mines. The preamble to the 1991 Federal Register publication stated: Paragraph (g) is a new provision requiring post-blast examinations to minimize hazards to persons who will perform subsequent work in the area. . . . Trained and experienced persons conduct these examinations and would address all the potential hazards present at a blast area including ground conditions [and] undetonated explosives . . . . 56 Fed. Reg. at 2081.[13] If, as in this instance, a standard is ambiguous, we defer to the Secretary's reasonable interpretation of the regulation. See cases cited slip op. at 6. Contrary to ROA's argument, there is nothing in the plain language of the regulation that requires only a single inspection at a blast site. To the extent ROA relies on the reference in the rule to "a . . . post-blast examination" (ROA Br. at 32), the regulation is either silent or ambiguous on the issue of what may trigger the inspection. The preamble cited by the Secretary is fully consistent with her reading of the regulation that multiple inspections might be necessary depending on the hazards and the people in the area performing subsequent work. See also Coal Employment Project v. Dole, 889 F.2d 1127, 1131 (D.C. Cir. 1989). As the judge held, "the concept of resumption of work is a continuing process." 17 FMSHRC at 1950. Indeed, it is apparent from the practice at ROA's quarry that inspections in blast areas were an ongoing process. See id. at 1931. Thus, ROA's argument that resumption of work occurred only once while quarrying a bench and that, for the accident bench, it occurred in 1993 following the blast at the bench, is contrary to the record. As ROA noted in its post-trial brief to the judge: After the immediate post-blast examination, the nature of the ROA quarrying process necessarily produces subsequent examinations of the bench. As Mr. Murray explained, lines are split and toppled away from the blasted bench. This process exposes new stone as each line is removed. The ROA powdermen perform a safety inspection before each line is toppled. Then, after each new line is exposed, the powdermen examine the freshly exposed stone for misfires and other safety hazards in the same manner as they did the face. ROA Post-Trial Br. at 22. See also PDR at 7 n. 11; 17 FMSHRC at 1947. This process was repeated when removal of the bench was completed, the floor of the bench was completely exposed, and preparation for quarrying on the top of the adjacent bench was begun. See generally Tr. II 841-45 (testimony of ROA witness Bolio).[14] In short, ROA's practice belies the interpretation of section 56.6306(g) it urges before the Commission - that the regulation only requires one examination before work resumes on the bench following a blast. Application of the regulation to require ongoing examinations of the quarry floor, especially prior to resumption of channel burning on the adjacent bench, is reasonable.[15] The operative events to which the regulation applies are not, therefore, the June 1993 blast and subsequent inspection by foreman Kelty and Reynolds. Rather, ROA had an ongoing obligation to inspect the blast site as new granite was exposed during the quarrying process. Thus, ROA's retroactivity argument fails. 2. Unwarrantable Failure We agree with the judge that the violation was the result of ROA's unwarrantable failure. See 17 FMSHRC at 1950. In light of the discovery of the four unexploded bags of pyrodex in 1993, Kelty's failure to probe the caprock for more misfires evidenced a disregard for the hazards associated with misfires in the presence of torch flames. The danger of misfires was evident from the fact that they would eventually be uncovered as benches were removed in the quarrying process. This obvious danger supports the judge's unwarrantable failure determination. See BethEnergy Mines, 14 FMSHRC at 1243-44. The discovery of 40 bags of pyrodex after the fatal explosion further indicates the inadequacy of ROA's efforts to ensure that misfires were not present before miners resumed work on benches after detonations. Substantial evidence supports the judge's findings. Accordingly, we affirm the judge's unwarrantable failure determination with respect to this violation. D. Section 56.6904[16] 1. Violation The judge affirmed the citation charging a violation of section 56.6904, which prohibits the use of an open flame within 50 feet of explosive material. 17 FMSHRC at 1951-52. The judge found that ROA had constructive knowledge of the likelihood of the existence of unexploded pyrodex once derrick operator Batchelder discovered the four bags of pyrodex. Id. He further found that ROA had actual knowledge of the placement of the explosives and location of the bench from which those unexploded bags came. Id. at 1951. Thus, the judge concluded that permitting the use of the channel burning torch in light of that actual and constructive knowledge established a violation of the standard. Id. at 1952. ROA argues that the phrase "shall not be permitted" requires knowledge on the part of the operator before a violation can be established. ROA Br. at 38. ROA contends that, because it was not aware of the location of any misfire, the citation must be vacated. Id. at 39-40. ROA further argues that the record does not support a finding of constructive knowledge of misfires. Id. at 40-41. ROA also contends that application of section 56.6904 has been limited to storage or placement of explosives, not misfires. Id. at 38-39. In response, the Secretary contends that, as used in the regulation, the term "permit" does not require that the operator know that flames are being used within 50 feet of explosives. S. Br. at 58-59. Even if it did, the Secretary continues, knowledge includes constructive knowledge, deliberate ignorance, and reckless disregard. Id. Lastly, the Secretary notes that MSHA should not be estopped from citing a mine operator for having an open flame within 50 feet of a misfire because it has not applied the regulation in that circumstance before. Id. at 60 n. 25. We agree with the Secretary that the regulation does not require actual knowledge on the part of the operator. Such an interpretation of the standard would undermine the strict liability principles of the Mine Act. See generally Western Fuels-Utah, Inc., 10 FMSHRC 256 (March 1988). We note that other more commonly cited standards use the word "permitted" (see 30 C.F.R. § 75.400 ("Coal dust . . . and other combustible materials, shall be cleaned up and not be permitted to accumulate in active workings, or on electric equipment therein.")), and the Commission has not held knowledge to be a prerequisite to finding a violation of those regulations. Chairman Jordan and Commissioner Marks would affirm the judge's determination that ROA violated the standard and that the violation was the result of ROA's unwarrantable failure. Commissioner Riley and Commissioner Verheggen would reverse the judge's determination that there was a violation. Under Pennsylvania Electric Co., 12 FMSHRC 1562, 1562-63 (August 1990), aff'd on other grounds, 969 F.2d 1501 (3rd Cir. 1992), the effect of a tie vote is to allow the judge's decision to stand as if affirmed.[17] E. Section 56.6300(a)[18] 1. Violation The judge found that ROA violated section 56.6300(a) when it permitted miners who were not properly trained and experienced in the use of pyrodex to direct the June 22, 1993 blast. 17 FMSHRC at 1952-54. The judge relied on the 26 percent misfire rate of the June 22 detonation and foreman Kelty's failure to take any meaningful action following Batchelder's discovery of the four bags of unexploded pyrodex. Id. at 1953. The judge further held that the violation was due to ROA's unwarrantable failure in light of its apparent failure to follow proper loading procedures for pyrodex. Id. In addition, the judge noted the lack of training of ROA personnel during the six years after the initial use of pyrodex in 1986. Id. at 1954. Finally, in upholding the unwarrantable failure designation, the judge noted that it was the absence of a competent inspection following the discovery of the misfires that led to Bassett's death, rather than problems related to the loading procedures of pyrodex. Id. at 1954. ROA argues that it satisfied the training requirement of the regulation because its employees were trained in "explosives." ROA Br. at 41-42. Further, ROA argues that its employees were trained in the use of pyrodex as a result of the visits of Dean Barrett, vice president of Hodgdon Powder. Id. at 42. According to ROA, section 56.6300(a) does not require retraining. Id. at 43. Finally, ROA contends that the existence of misfires and an accident does not prove the training violation. Id. at 44. The Secretary counters that the regulation requires training in the particular explosive used at the mine, not simply training in handling any explosive. S. Br. at 61-62. The Secretary relies on the language of the preamble to support her contention that training in the specific explosive in use is required. Id. at 62-63. According to the Secretary, the record evidence supports the judge's finding that neither foreman Kelty nor powderman Reynolds were trained by Barrett in the use of pyrodex. Id. at 63-64. The Secretary further contends that neither were trained to look for misfires. Id. at 64-65. The judge concluded that the standard requires that blasting personnel be trained and experienced in the "particular explosive being used." 17 FMSHRC at 1952. We agree. The preamble to the standard in the Federal Register illustrates the Secretary's intent to require training in explosives actually utilized by employees: The training and experience needed to supervise or direct blasting operations in today's mines where the technology is continually changing and may exceed the training provided to miners who simply handle explosives under a supervisor's direction. As new explosive materials are introduced at the mine, the persons directing the blasting operations must be trained in the safe handling and use of the products. . . . These new products may create hazards in storage, transportation, loading, blast hook- up, blast area security, firing and post blast examinations. Training must address these areas where appropriate. 56 Fed. Reg. at 2079. We agree with the Secretary that, in order to avoid "absurd results" (see Consolidation Coal, 15 FMSHRC at 1557), the standard must be interpreted to require that individuals handling explosives be trained in the type of explosives they are actually utilizing. This is demonstrated here where ROA employees, who were trained and experienced in the use of black powder, were clearly ill prepared to properly handle and dispose of pyrodex. ROA's insistence that general training in explosives suffices to meet the standard's requirements is antithetical to the Mine Act's goal of protecting miner safety. See Freemen Coal Mining Co. v. Interior Board of Mine Operations Appeals, 504 F.2d 741, 744 (7th Cir. 1974). Moreover, the record shows that Hodgdon Vice President Dean Barrett did not recall that Kelty or Reynolds were present in 1986 or 1987 during the experimental testing of pyrodex.[19] 17 FMSHRC at 1953. Nor was there any evidence that anyone at ROA had been trained during the six-year interval between the test firings and the resumed use of pyrodex in 1993. Id. at 1953-54. Further, as the Secretary points out (S. Br. at 64), Barrett testified without contradiction that, during the 1986-87 experimental blasts, he did not participate with anyone at ROA in a post-blast examination. Tr. I 177-78. Finally, the judge relied on the 26 percent misfire rate of pyrodex, and foreman Kelty's failure to engage in a further search for additional misfires after Batchelder's discovery of the four bags, as strong circumstantial evidence of inadequate training in light of the sequential ignition process.[20] 17 FMSHRC at 1952-53, 1954. On this record, substantial evidence supports the judge's findings that Kelty and Reynolds had not been adequately trained in the use and handling of pyrodex. Accordingly, we affirm the judge's finding of violation. 2. Unwarrantable Failure We also agree with the judge that the violation was the result of ROA's unwarrantable failure. Even if Kelty and Reynolds were present at the 1986-87 test firings of pyrodex, it is apparent from the testimony of Barrett that no one from ROA was trained in how to conduct a post-blast examination. As MSHA Inspector Fowler testified, Reynolds stated during the accident investigation that he did not know how to search for a pyrodex misfire. 17 FMSHRC at 1953. Kelty clearly failed to understand the significance of the sequential ignition process, as evidenced by his lack of followup after Batchelder's discovery of the misfires. Substantial evidence supports the judge's determination. Therefore, we affirm the judge's ruling that ROA's failure to adequately train employees using pyrodex, given the extreme consequences of misfires in the presence of a channel-burning torch, was an unwarrantable failure to comply.[21] F. Whether the ALJ Denied ROA Due Process ROA argues that the judge was biased against ROA. ROA Br. at 57-62. ROA contends that the judge's decision itself evidenced a lack of impartiality, noting several of the judge's characterizations of ROA's position. Id. at 58. ROA asserts that the judge exhibited bias in the hearing and cites 15 instances of conduct that, according to ROA, support a finding of bias. Id. at 58-62. In response, the Secretary argues that ROA has not pointed to any direct evidence of bias but rather asks the Commission to infer bias from rulings of the judge. S. Br. at 77. The Secretary contends that the judge's actions in sustaining objections, questioning witnesses, and making findings adverse to ROA are insufficient to support a finding of bias. Id. at 78-80. The Secretary also notes that the judge made many rulings during the hearing favorable to ROA and that ROA's counsel engaged in provocative conduct at the hearing. Id. at 80-81. When reviewing matters involving a judge's conduct of trial, the Commission is governed by the abuse of discretion standard. See, e.g., In re: Contests of Respirable Dust Sample Alteration Citations, 17 FMSHRC 1819, 1843-44, 1853, 1864 (November 1995), appeal docketed sub nom. Secretary of Labor v. Keystone Coal Mining Corp., No. 95-1619 (D.C. Cir. Dec. 28, 1995); Big Horn Calcium Co., 12 FMSHRC 1493, 1496 (August 1990). With regard to the judge's role at trial, the Commission has stated, "a judge is an active participant in the adjudicatory process and has a duty to conduct proceedings in an orderly manner so as to elicit the truth and obtain a just result." Secretary on behalf of Clarke v. T.P. Mining, Inc., 7 FMSHRC 989, 993 (July 1985). In reversing a judge's sua sponte post-hearing joinder of a party, the Commission held, "[t]he role of the Commission and its judges is to adjudicate, not to litigate cases - a procedural axiom followed by this Commission from its formation." Lonnie Jones v. D&R Contractors, 8 FMSHRC 1045, 1053 (July 1986). Addressing allegations of judicial bias that deprived a party of a fair trial, one court has noted that a judge "has the prerogative to interrogate witnesses, and the duty to do so where necessary to clarify testimony, but the judge must maintain an air of impartiality." Deary v. City of Gloucester, 9 F.3d 191, 195 (1st Cir. 1993); see also Canterbury Coal Co., 1 FMSHRC 1311, 1312-13 (September 1979).[22] We have reviewed the judge's decision and the transcript of the hearing, particularly the transcript pages cited in ROA's brief. We conclude that the judge's actions were well within the bounds of his proper role in conducting the trial. See Liteky v. United States, 510 U.S. 540 (1994) (judicial rulings, routine trial administration efforts, and ordinary admonishments to counsel and witnesses inadequate grounds for disqualification); see also T.P. Mining, 7 FMSHRC at 993 ("A judge is an active participant in the adjudicatory process . . . ."). None of the cited pages, either individually or cumulatively, indicate that the judged harbored bias towards ROA or its counsel.[23] Cf. United States v. Donato, 99 F.3d 426, 434-39 (D.C. Cir. 1996) (jury conviction set aside for several reasons, including judge's prejudicial comments to defendant and "near constant criticism of defendant's counsel"). Accordingly, we reject ROA's argument that the judge's decision or his conduct of the trial require reversal due to the judge's lack of impartiality. F. Penalties Commission judges are accorded broad discretion in assessing civil penalties under the Mine Act. Westmoreland Coal Co., 8 FMSHRC 491, 492 (April 1986). Such discretion is not unbounded, however, and must reflect proper consideration of the penalty criteria set forth in section 110(i) of the Act.[24] Id. (citing Sellersburg Stone Co., 5 FMSHRC 287, 290-94 (March 1983), aff'd, 736 F.2d 1147 (7th Cir. 1984)). In reviewing a judge's penalty assessment, the Commission must determine whether the judge's findings with regard to the penalty criteria are supported by substantial evidence. The judge must make "[f]indings of fact on each of the criteria [to] not only provide the operator with the required notice as to the basis upon which it is being assessed a particular penalty, but also provide the Commission and the courts . . . with the necessary foundation upon which to base a determination as to whether the penalties assessed by the judge are appropriate, excessive, or insufficient." Sellersburg, 5 FMSHRC at 292-93. It does not appear that the judge considered all of the statutory criteria in determining an appropriate penalty for ROA's violations. The judge did not separately discuss facts relating to any of the criteria. Rather, the judge referred only to the extremely high negligence and serious gravity associated with each of the violations. 17 FMSHRC at 1949, 1950, 1952, 1954. This does not satisfy the requirements the Commission set out in Sellersburg. Accordingly, we vacate the judge's penalty assessment and remand for entry of detailed findings as to each of the six section 110(i) criteria, and assessment of an appropriate penalty. III. Conclusion For the foregoing reasons,[25] we affirm the judge's findings of violation and his determinations of unwarrantable failure, but vacate the penalty assessment and remand it for further consideration. Mary Lu Jordan, Chairman Marc Lincoln Marks, Commissioner James C. Riley, Commissioner Theodore F. Verheggen, Commissioner Distribution Robin A. Rosenbluth, Esq. Office of the Solicitor U.S. Department of Labor 4015 Wilson Blvd., Suite 400 Arlington, VA 22203 Henry Chajet, Esq. Patton Boggs, LLP 2550 M Street, N.W. Washington, D.C. 20037 Harry Tuggle, Esq. United Steelworkers of America Health & Safety Department Five Gateway Center Pittsburgh, PA 15222 David Gomo c/o United Steelworkers of America Amalgamated Local #4 P.O. Box 584 Barre, VT 05641 David Gomo Union Stewart & Safety Committeeman United Steelworkers of America P.O. Box 482 Barre, VT 05641 Administrative Law Judge Jerold Feldman Federal Mine Safety & Health Review Commission Office of Administrative Law Judges 5203 Leesburg Pike, Suite 1000 Falls Church, VA 22041 **FOOTNOTES** [1]: Commissioner Beatty assumed office after his case had been considered at a decisional meeting. A new Commissioner possesses legal authority to participate in pending cases, but such participation is discretionary. Mid-Continent Resources, Inc., 16 FMSHRC 1218, 1218 n. 2 (June 1994). In the interest of efficient decision making, Commissioner Beatty has elected not to participate in this matter. [2]: "Tr. I" references are to the consecutively numbered transcript volumes from the first four days of hearing conducted on January 10-13, 1995, and "Tr. II" references are to the consecutively numbered transcript volumes from the second four days of hearing conducted on April 25-28, 1995. Exhibits designated with the letter "R" are those introduced into evidence by the Secretary, and exhibits designated with the letter "C" are those introduced by ROA. [3]: "Explosive" is defined in 30 C.F.R. § 56.6000, as "[a]ny substance classified as an explosive by the Department of Transportation . . . . " [4]: The loading pattern began with the first and fourth holes and then went to each fourth hole thereafter, resulting in a sequential loading pattern of 1, 4, 8, 12, 16, 20, 24, 28, etc. 17 FMSHRC at 1934. The judge based this finding on ROA's initial accident report (Ex. R-8), testimony, and other exhibits, including a diagram, Ex. R-10, drawn by ROA foreman Earl Kelty indicating the loading pattern and bags of unexploded pyrodex located after the accident. Id. at 1939-41 & n 5. [5]: In addition, following the accident, 18 other pyrodex misfires were found at other blast sites at Smith Quarry and at a second quarry where pyrodex was tested, yielding a total of 40 unexploded bags of pyrodex. See 17 FMSHRC at 1941. [6]: The S&S terminology is taken from section 104(d)(1) of the Act, 30 U.S.C. § 814(d)(1), which distinguishes as more serious any violation that "could significantly and substantially contribute to the cause and effect of a . . . mine safety or health hazard." [7]: Part 30 C.F.R. § 56.6000 defines "misfire" as "[t]he complete or partial failure of explosive material to detonate as planned. The term also is used to describe the explosive material itself that has failed to detonate." [8]: Section 56.6311 provides: Handling of misfires. (a) Faces and muck piles shall be examined for misfires after each blasting operation. (b) Only work necessary to remove a misfire and protect the safety of miners engaged in the removal shall be permitted in the affected area until the misfire is disposed of in a safe manner. (c) When a misfire cannot be disposed of safely, each approach to the area affected by the misfire shall be posted with a warning sign at a conspicuous location to prohibit entry, and the condition shall be reported immediately to mine management. (d) Misfires occurring during the shift shall be reported to mine management not later than the end of the shift. [9]: Substantial evidence supports the judge's finding that the June 22, 1993 blast site was the location of the fatal accident, 17 FMSHRC at 1939-43. Particularly compelling is the blast report for that site, Ex. R-7, concluding that the four misfires Batchelder found originated from that site. In addition, the accident report which ROA submitted to MSHA, Ex. R- 8, identified the accident bench as the location of the June 22 blast. We also note that the loading pattern of pyrodex within the lift holes and the loading sequence of the lift holes at the June 22 blast site and the accident bench indicate that the blast site and the accident bench were the same. When reviewing an administrative law judge's factual determinations, the Commission is bound by the terms of the Mine Act to apply the substantial evidence test. 30 U.S.C. § 823(d)(2)(A)(ii)(I). "Substantial evidence" means "`such relevant evidence as a reasonable mind might accept as adequate to support [the judge's] conclusion.'" Rochester & Pittsburgh Coal Co., 11 FMSHRC 2159, 2163 (November 1989) (quoting Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229 (1938)). [10]: In support of its argument that section 56.6311(b) should be read to only apply to "found" misfires, ROA asserts that the present language of section 56.6311 was taken from 30 C.F.R. § 56.6106 (1990), which provided, in pertinent part: "undetonated explosives . . . found shall be disposed of safely." ROA Br. at 30 (citing 56 Fed. Reg. at 2081). However, ROA ignores language in the preamble which stated that section 56.6311 was also based on a second regulation, 30 C.F.R. § 56.6168 (1990) (Handling of Misfires), which prohibited work in a blast area where there were misfires without reference to those "found" or any other words of limitation. See 56 Fed. Reg. at 2082. Faced with the differing language of these two predecessor provisions, we find it more persuasive that the Secretary drafted section 56.6311(b) and the reference to misfires with no limiting language. Similarly, in the regulatory preamble to the standard, the Secretary rejected a more restrictive reference to the areas affected by misfires ("blast site") and opted for a broader reference ("affected area") in the standard. Id. See also id. ("The final standard retains `affected areas' because an accidental detonation during the removal of a misfire is likely to affect an area larger than the blast site."). [11]: We note that ROA raised in its PDR only the judge's determination of unwarrantability with regard to the standard limiting the handling of explosives to miners trained and experienced in the handling of explosives, 30 C.F.R. § 56.6300(a). See ROA PDR at 10,  29. Petitioner failed to specifically raise the judge's unwarrantability determinations with regard to the three other violations. ROA raised generally the judge's conclusions regarding ROA's negligence. Id. at 5,  7. We address the unwarrantability determinations with regard to these three violations that ROA has addressed in its brief, because those arguments are sufficiently related to the negligence issue in the PDR. However, we admonish petitioner and counsel to adhere to the requirements of the Mine Act and the Commission's procedural rules. See 30 U.S.C. § 823(d)(2)(A)(iii); 29 C.F.R. § 2700.70(d) and (f). [12]: Section 56.6306 states, Loading and blasting. * * * * (g) No work shall resume in the blast area until a post-blast examination addressing potential blast-related hazards has been conducted by a person with the ability and experience to perform the examination. [13]: MSHA stayed the effective date of several of the regulations, including section 56.6306(g), in order to reopen the rulemaking record. The stay was extended on several occasions, and on December 30, 1993, MSHA published the revised regulations with an effective date of January 31, 1994. See 58 Fed.Reg. 69,596 (1993). The final rule lacked any substantive changes but was "edited for clarity." Id. at 69,603-04. [14]: Testimony to the contrary, that a post-blast examination was a single inspection immediately after the blast, was elicited from ROA's expert witness in response to a leading question from ROA counsel. Tr. II 892. [15]: ROA's notice argument (ROA Br. at 35), that the regulation would not inform "a reasonably prudent person" of its requirements is undermined by its own conduct - ROA performed numerous inspections after the blast. See ROA Br. at 34 n. 28. Thus, the Secretary's interpretation is consistent with industry practice and ROA was on notice of the requirements of the regulation. See Ideal Cement Co., 12 FMSHRC 2409, 2416 (November 1990). [16]: Section 56.6904 provides: Smoking and open flames. Smoking and use of open flames shall not be permitted within 50 feet of explosive material except when separated by permanent noncombustible barriers. This standard does not apply to devices designed to ignite safety fuse or to heating devices which do not create a fire or explosion hazard. [17]: Chairman Jordan and Commissioner Marks conclude that the plain language of the regulation reaches ROA's conduct at issue - permitting an open flame within 50 feet of unexploded pyrodex. Furthermore, they do not view the reference to barriers as restricting application of the regulation solely to storage or transportation of explosives. The standard is located in a section of the regulations entitled "General Requirements-Surface and Underground" and is not found in the "storage" section (sections 57.6100-57.6133) nor in the transportation section (sections 57.6200-57.6205). Chairman Jordan and Commissioner Marks also agree with the judge that the violation was the result of ROA's unwarrantable failure. As the judge noted, once Batchhelder discovered the four misfires from the June 22, 1993, blast site, ROA was aware of the likelihood of the continued existence of unexploded pyrodex. 17 FMSHRC at 1951-52. Allowing Bassett's use of the channel burner with its open flame near the old blast site in light of the previous discovery of the unexploded pyrodex evinces a serious lack of reasonable care that well supports an unwarrantable failure determination. See Midwest Material Co., 19 FMSHRC at 35. Commissioners Riley and Verheggen would reverse the judge's finding of a violation because they find that section 56.6904 clearly does not apply to the conduct at issue. Dyer, 832 F.2d at 1066. There appears to be nothing in the language of the standard or its regulatory history which supports the Secretary's interpretation that would have the standard address the hazard of unexploded misfires. Regulations discussed earlier, section 56.6311(b) (handling of misfires) and section 56.6306(g) (loading and blasting), fully address the problem of allowing Bassett back onto the bench to work with a torch in the presence of unexploded pyrodex. In Title III of the Mine Act, which applies to underground coal mines, the standard that is analogous to section 56.6904 is clearly aimed at the safe storage of explosives, rather than the proper handling of misfires. See 30 U.S.C. § 873(g). Similarly, it is more reasonable to interpret the reference in section 56.6904 to "permanent noncombustible barriers" as referring to storage or transportation of explosives, but not to misfires. Commissioners Riley and Verheggen thus conclude that the Secretary's interpretation of section 56.6904 is unreasonable. Accordingly, they would reverse the judge's finding of a violation. [18]: Section 56.6300 provides: Control of blasting operations. (a) Only persons trained and experienced in the handling and use of explosive material shall direct blasting operations and related activities. (b) Trainees and inexperienced persons shall work only in the immediate presence of persons trained and experienced in the handling and use of explosive material. [19]: Neither Kelty or Reynolds testified at the hearing. [20]: ROA argues that Barrett's instructions for loading pyrodex were flawed, which was the reason for the fatal misfire. See ROA Br. at 43. However, it is more significant that there was an absence of any training, particularly in conducting post-blast inspections, than whether all of the information concerning the handling of pyrodex was completely accurate. [21]: Commissioner Riley dissents on the issue of unwarrantability, offering the following rationale: [T]he Secretary and the Commission interpret the words "unwarrantable failure" to require a culpability determination similar to gross negligence or recklessness. . . . Citing the Restatement (2d) of Torts § 283 (1965), the Secretary argues that this negligence-based definition of "unwarrantable failure" requires consideration of surrounding circumstances. ("[T]he standard of conduct to which [the actor] must conform to avoid being negligent is that of a reasonable man under the circumstances.") (emphasis added). Secretary of Labor v. FMSHRC, 111 F.3d. 913, 919-920. (D.C. Cir. 1997) (citations omitted) (alteration in original). For almost a century hardy souls around Barre, Vermont have been carving a living and a legend out of the granite of the Green Mountains. For Rock of Ages to continue to fulfill its historic role, it must remain profitable in a competitive industry. Realizing this, the miners at ROA also have the United Steelworkers of America ("USWA") to help them secure adequate wages, promote safe working conditions, and otherwise look out for their interests. While such relationships can be adversarial, enlightened labor relations have become more consultive and cooperative. Working together, ROA and the USWA developed and administered a joint training program. The program has operated successfully for a number of years and would seem to have achieved a measure of respect. Even the MSHA inspector, on site for several of the pyrodex blasts, was trained in the handling of explosives as an employee of ROA. In retrospect we know that, in fact, nobody had been adequately trained to guard against the particular threat that turned parts of the quarry into a mine field. Procedures that had proven adequate for primacord blasting were found to be insufficient to detect and eliminate numerous pyrodex misfires. Surely someone should have discovered deficiencies in the post blast inspection procedure which with the benefit of hindsight seem glaring. In fact, up until the day of this accident, neither the operator, the union, the workers or MSHA believed they had any reason to question the adequacy of training for anyone responsible for the post blast inspection regime. Unfortunately, even the discovery of unexploded ordinance did not seem to awaken those involved to the heightened danger posed by ROA's experiment with pyrodex. Under the circumstances, in the absence of proper instruction from the manufacturer on the safe use of pyrodex as well as the lack of intervention from MSHA, which was on site and aware of the use of experimental explosives, I am unable to characterize the now obvious deficiencies in ROA's training program as unwarrantable failure. [22]: In Canterbury, the Commission reversed the decision of an administrative law judge and remanded the case for reassignment to a new judge after concluding that the judge's conduct at the hearing was an abuse of discretion. While noting "the considerable leeway afforded administrative law judges in regulating the course of a hearing and in developing a complete and adequate record," the Commission concluded that the judge interjected himself in the proceedings "so often and so extensively that [the parties] were denied the opportunity to develop their case." 1 FMSHRC at 1312-13. [23]: We note that ROA did not allege bias of the judge until after it lost the case and petitioned the Commission for review. See ROA Post-Trial Br. at 8 n. 2. [24]: Section 110(i) sets forth six criteria to be considered in the assessment of penalties under the Act: [1] the operator's history of previous violations, [2] the appropriateness of such penalty to the size of the business of the operator charged, [3] whether the operator was negligent, [4] the effect on the operator's ability to continue in business, [5] the gravity of the violation, and [6] the demonstrated good faith of the person charged in attempting to achieve rapid compliance after notification of a violation. [25]: Following the Commission's October 10, 1997 Order in which it denied the Motion to Participate in Oral Argument by David Gomo, he submitted a written Oral Argument Statement to the Commission on October 17, which he requested the Commission to accept. We have considered Mr. Gomo's request, and we deny it. We have not considered the statement in our deliberations in this matter.