Selected Case Law » General Motors Corporation

Occupational Safety and Health Review Commission and Administrative Law Judge Decisions

04/26/1995 - 91-2973 - General Motors Corporation, Delco Chassis Division

  • Docket Number: 91-2973
  • Standard Number: 1910.147
  • Case Citation: 17 BNA OSHC 1217
  • Company: General Motors Corporation, Delco Chassis Division
  • Information Date: 04/26/1995




 General Motors Corporation, Delco Chassis Division

OSHRC Docket No. 91-2973



At issue is whether former Commission Judge Edwin G. Salyers erred in vacating citations that alleged failure by General Motors Corp., Delco Chassis Div. ("GM") to deenergize and lockout machines under the lockout/tagout standard, 29 CFR 1910.147.(1) We find that, by its plain meaning, the standard applies only to those machines and pieces of equipment for which energization or start up would be unexpected by employees. The Secretary has argued for an interpretation that would have the Commission ignore this requirement. We agree with the judge, however, that the standard requires the Secretary to establish that a cited piece of equipment or machinery presents the hazard of unexpected energization, and that he failed to do so in each of the three cases here. We thus affirm the judge's decision.(2)

FOOTNOTE(1)  In each of these cases, GM was cited for failure to require its employees to lockout electrically-powered machines prior to servicing or maintenance. The machines were used to manufacture automotive parts at GM's plants at Vandalia and Dayton, Ohio.  In Docket No. 91-2973, the cited standard is section 1910.147(d)(4)(i).  That section provides:

(d) Application of control.  The established procedures for the application of energy control (the lockout or tagout procedures) shall the cover the following elements and actions and shall be done in the following sequence:

(4) Lockout or tagout device application.  (i) Lockout or tagout devices shall be affixed to each energy isolating device by authorized employees.

Section 1910.147(b) defines an "energy isolating device" as a "mechanical device that physically prevents the transmission or release of energy." Examples are "a manually operated circuit breaker [or] a disconnect switch."  That definition also provides that "[p]ush buttons, selector switches and other control circuit type devices are not energy isolating devices."  A "lockout device" is a lock or other positive means that holds an energy isolating device in a safe position so that a deenergized machine or equipment remains deenergized.  Id.  In Docket Nos. 91-3116 & 91-3117, the cited standard is section 1910.147(c)(4)(i).

That section provides:

(c) General--(1) Energy control program....
(4) Energy control procedure. (i) Procedures
[including a lockout procedure] shall be developed, documented and utilized for the control of potentially hazardous energy when employees are engaged in the activities covered by this section.

The Secretary has withdrawn the other alleged violations that were directed for review.  Those items alleged deficiencies in GM's written lockout program under section 1910.147(c)(4)(ii), in Docket Nos. 91-3116 and 91-3117.  We accept the Secretary's withdrawel of those items.

FOOTNOTE(2)  The pre-enforcement challenges to the validity of the lockout/tagout standard recently were rejected.  International Union, UAW v. OSHA, 37 F.3d 665 (D. C. Cir. 1994) ("UAW v. OSHA II").  The court found "that OSHA's current interpretation of its statutory authority to issue safety standards is consistent with the nondelegation doctrine and that its explanations of the other disputed decisions are adequate[.]" Id. at 668.  In 1991, the court had remanded the standard to OSHA for a supplemental statement of supporting reasons, on the ground that the agency's interpretation of its authority was not consistent with the nondelegation doctrine.  International Union, UAW v. OSHA, 938 F.2d 1310 (D. C. Cir. 1991) ("UAW v. OSHA I").


I. The application of the standard

The lockout/tagout standard begins with a scope provision, the first sentence of which reads as follows:

This standard covers the servicing and maintenance of machines and equipment in which the unexpected energization or start up of the machines or equipment, or release of stored energy could cause injury to employees.

29 CFR 1910.147(a)(1)(i) (emphasis in original). The Secretary's case is premised upon his "official interpretation" of this sentence. Transposing the operative language, the Secretary frames the applicability inquiry as follows: "whether injury could occur in the event of an unintended energization, start up, or release of stored energy." Having thus interpreted the scope provision, the Secretary reasons that the lockout/tagout standard applies regardless of "how likely or remote the chances of" unexpected energization are and argues that any other conclusion "constitutes nothing less than a prohibited challenge to the wisdom of the standard."

Like all other Commission judges whose decisions have been cited to us,(3) Judge Salyers concluded that the language in question renders the standard inapplicable whenever employees are given sufficient notice of energization to allow them to vacate the zone of danger. In situations where the meaning of regulatory language is "not free from doubt," the Commission, like any reviewing court, must give effect to the Secretary's reasonable interpretations of his regulatory language. Cf., e.g., Martin v. OSHRC (CF & I Steel Corp.), 499 U.S. 144, 150 (1991).

FOOTNOTE(3) Armco Steel Co., OSHRC Docket No. 93-641 (December 20, 1993) (ALJ Loye); Caterpillar, Inc., 92 OSAHRC 67/C8 (Docket No. 92-127, 1992) (ALJ Barkley). See also Metal Shredders, Inc. 92 OSAHRC 17/A2 (Docket No. 90-2273, 1992) (ALJ Burroughs) (work that was not performed under lockout in that case was not servicing or maintenance, and all affected employees expected equipment to start up).

In determining whether the language of a standard is ambiguous, we first look to its text and structure. When the statute speaks with clarity, in all but the most extraordinary circumstances, judicial inquiry is ended. E.g., Estate of Cowart v. Nicklos Drilling Co., 112 S. Ct. 2589, 2594 (1992). Only if we can make no initial determination need we refer to contemporaneous legislative histories of the standard. If the question remains unsettled, we look to the reasonableness of the interpretation of the agency that administers the challenged standard. Kiewit Western Co., 16 BNA OSHC 1689, 1693, 1993 CCH OSHD paragraph 30,396, p. 41,940 (No. 91-2578, 1994) (citing Unarco Commercial Prods., 16 BNA OSHC 1499, 1502-03, 1993 CCH OSHD paragraph 30,294, p. 41,732 (No. 89-1555, 1993)).

Examining further the text of the standard, not only does the very first provision emphasize that its scope is limited to "unexpected" energization, startup, or release of stored energy, but the definition of covered maintenance and servicing contains the same limitation:
Servicing and/or maintenance. Workplace activities such as constructing, installing, setting up, adjusting, inspecting, modifying, and maintaining and/or servicing machines or equipment. These activities include lubrication, cleaning or unjamming of machines or equipment and making adjustments or tool changes, where the employee may be exposed to the unexpected energization or startup of the equipment or release of hazardous energy.

Section 1910.147(b) (emphasis in original).(4) OSHA literally underscored the importance of the standard's limitation to "unexpected" energization, etc., by italicizing that word in those two provisions--a form of emphasis that OSHA rarely uses. The same limitation is restated in the general requirement for an energy program, section 1910.147(c)(1), which the Secretary describes as "[p]erhaps the clearest summary of the requirements of 1910.147."

FOOTNOTE(4)  GM has not conceded that its employees were engaged in "servicing and/or maintenance within the meaning 1910.147(b), because in its view that definition is limited to activities where inadvertent activation of the machine or equipment could occur and cause injury.

The employer shall establish a program consisting of energy control procedures, employee training and periodic inspections to ensure that before any employee performs any servicing or maintenance on a machine or equipment where the unexpected energizing, start up or release of stored energy could occur and cause injury, the machine or equipment shall be isolated from the energy source, and rendered inoperative.
(Emphasis added.)(5)

FOOTNOTE(5)  Other provisions are to the same effect.  The stated purpose of the standard is as follows:

This section requires employers to establish a program and utilize procedures for affixing appropriate lockout devices or tagout devices to energy isolating devices, and to otherwise disable machines or equipment to prevent unexpected energization, start-up or release of stored energy in order to prevent injury to employees.

Section 1910.147(a)(3)(i) (emphasis added).  Similar language appears in certain other provisions:  sections 1910.147(a)(2)(iii)(A)(c)(4)(i) (exception to requirement of energy control procedure), (f)(4), and Appendix A.

We find that the standard clearly and unambiguously applies only where the Secretary shows that unexpected energizing, start-up or release of stored energy could occur and cause injury. Under these circumstances, we find it unnecessary to look outside the standard itself for guidance as to its meaning. E.g., Brown v. Gardner, 115 S. Ct. 552, 556 (1994) (long-standing Veterans' Affairs Department interpretation of statute overruled--"the text [of the statute] and reasonable inferences from it give a clear answer against the Government, and that, as we have said, is `the end of the matter"') (quoting Good Samaritan Hosp. v. Shalala, 113 S.Ct. 2151, 2157 (1993)); Unarco, 16 BNA OSHC at 1503 & n.3, 1993 CCH OSHD at p. 41,732 & n.3. See generally 2A Singer, Sutherland Stat. Const. 46.01 (5th ed. 1992) (citing Caminetti v. United States, 242 U.S. 470 (1917) ("[i]t is elementary that the meaning of a statute must, in the first instance, be sought in the language in which the act is framed, and if that is plain [and if the statute is constitutional], the sole function of the courts is to enforce it according to its terms")).

However, even if we were to look beyond the standard's terms, we find any ambiguity is clarified in the legislative history and that the Secretary's interpretation is unreasonable. The preamble to the standard expressly limited its applicability to situations where "unexpected energization," etc.,(6) is a hazard. For example, the opening summary of the preamble states:

FOOTNOTE(6)  The term "energization, etc.," is our shorthand for the separate concepts of "energization, startup, or release of stored energy."

This standard addresses practices and procedures that are necessary to disable machinery or equipment and to prevent the release of potentially hazardous energy while maintenance and servicing activities are being performed.

Control of Hazardous Energy Sources (Lockout/Tagout): Final Rule, 54 Fed. Reg. 36,644 (1989), as corrected by 55 Fed. Reg. 38,677 (1990) (emphasis added). The preamble explains the standard's scope as follows:

The standard covers servicing and maintenance in general industry where the unexpected energization or start-up of machines or equipment or the release of stored energy could cause injury to employees.

54 Fed. Reg. at 36,659. There are many other statements to the same effect in the preamble. As we read it, the preamble can only support the plain language of the standard that limits its application to machines or equipment where an unexpected energization could cause injury to employees.

In addition to the preamble, other contemporaneous documents make clear that the standard is limited to risks of unexpected energization, etc. For example, OSHA Publication No. 3120, Control of Hazardous Energy (Lockout/Tagout) 2 (1991), states that if employees are performing servicing or maintenance tasks that do not expose them to the unexpected release of hazardous energy, the standard does not apply. The 1994 edition of that publication makes the same statement. OSHA's internal instructions for administering section 1910.147 use comparable language. OSHA Instruction STD 1-7.3, Control of Hazardous Energy Lockout-Tagout)--Inspection Procedures and Interpretive Guidance, I.1.b., d., e. (September 11, 1990). Thus, the Secretary's contemporaneous publications and the statements therein consistently reaffirm that the Secretary meant what he said in the standard--it applies only where unexpected energization, etc., could occur and cause injury to employees.

The plain and unambiguous terms of the standard, and the support of the legislative history, is so overwhelming that the Commission need not even consider the Secretary's current interpretation. We note, however, that the Secretary's interpretation is inconsistent with the standard's terms. He interprets it to apply to every piece of machinery and equipment regardless of whether it could start up unexpectedly. The Secretary would require lockout/tagout even if the record shows there is no possibility of "unexpected" energization, etc. As GM states, the Secretary would "inquire only into whether an employee could be injured if unexpected energization [etc.] were to occur (even if, in fact, it could not)."

This interpretation fails to give effect to the entire clause that defines the standard's application. We cannot ignore the term "unexpected" as a limitation on the application of the standard. Regulations are to be read so as to give effect to all their terms, if possible. E.g., United States v. Menasche, 348 U.S. 528, 538-39 (1955). See Brown v. Gardner, 115 S. Ct. at 556 (maxim of statutory interpretation held relevant to whether agency had properly interpreted its governing statute) (citing Russello v. United States, 464 U.S. 16, 23 (1983)). Again, the Secretary not only emphasized that limitation, but wrote it in repeatedly and purposely.

The Secretary may change his standard, but his interpretation of his current standard could only be "reasonable" if it "sensibly conforms to the purpose and wording of the regulations." CF & I, 499 U.S. at 151 (citing Northern Indiana Pub. Serv. Co. v. Porter County Chapter of Izaak Walton League of America, Inc., 423 U.S. 12, 15 (1975)); Martin v. American Cyanamid Co., 5 F.3d 140, 144 (6th Cir. 1993). Here, the Secretary's interpretation is plainly inconsistent with the wording of the regulations. His reading would essentially remove all limits to the standard's applicability.

The Secretary's interpretation in this case also contrasts sharply with portions of his supplemental statement of reasons supporting the standard. In that document, which the D.C. Circuit required during the pre-enforcement challenge to the standard, see supra note 2, the Secretary sought to reassure the court that he interpreted the standard to contain a common-sense limit to situations where unexpected energization, etc., could cause injury to employees.

The standard [does not apply] to servicing and maintenance that present minimal and readily controlled risk .... [E]ach covered employer's burden is determined by the frequency and complexity of servicing actually undertaken.... Machines and equipment that present no hazard are excluded from coverage.

Final rule: supplemental statement of reasons, 58 Fed. Reg. 16,612, 16,621 (1993) (emphasis added). The Secretary further emphasized before the D. C. Circuit that the standard applies only to machines and equipment that pose a significant risk of harm and to employees exposed to that risk. See UAW v. OSHA II, supra note 2, 37 F.3d at 670 ("[i]f, as OSHA asserts and NAM [the National Association of Manufacturers] appears not to dispute, the regulation applies simply to machines that pose a significant risk and to workers subjected to that risk, we see no reason why OSHA should be concerned with industry classifications that appear essentially irrelevant to its task") (emphasis in original).

Yet, in this enforcement action the Secretary contends that GM should not be allowed to prove that its machines and equipment present no hazard or no significant risk of harm. The Secretary essentially argues that he alone may decide whether such machines and equipment present a hazard of unexpected energization, etc., or a significant risk of harm, and that employers and the Commission may not question that determination. We again do not believe that the standard can be reasonably read to require the result the Secretary seeks.

Finally, we find no merit in the Secretary's claim that Judge Salyers's reading of the standard: (1) violates the requirement that the authorized employee have exclusive control over his/her safety, and (2) rewrites the definition of "energy isolating device." That claim presumes that there always is a hazard of unexpected energization, etc., on every industrial machine and piece of equipment during servicing and maintenance. The terms of the standard clearly place the burden on the Secretary to show that there is such a hazard as to the cited machines and equipment. The Secretary must show that there is some way in which the particular machine could energize, start up, or release stored energy without sufficient advance warning to the employee. However, the Secretary seeks to disallow reliance on even the most failsafe control circuit devices--even where the employees as well as employers favor them. We find that this unreasonable approach is flatly inconsistent with the unambiguous terms of the standard, as well as the preamble and the Secretary's other contemporaneous explanations of the standard.(7)

FOOTNOTE(7)  The standard contains certain specific exceptions to lockout/tagout requirements, as the dissent notes.  However, the Secretary must show that the standard applies before the exceptions become relevant. Because the Secretary failed to show that unexpected energization, etc., could occur on the cited machines, he has not shown that the standard applies, and thus we need not consider the exceptions.

II. Whether the Secretary proved that unexpected activation could occur and cause injury

It is undisputed that the machines had extensive precautions to protect servicing and maintenance employees. An electronically interlocked gate surrounded the machine area in each case. Once an employee opened that gate or pushed an emergency stop button, a time-consuming series of eight to twelve steps were required before any hazardous movement of the machine could occur. The evidence indicated that the restart procedures would provide plenty of warning to the employees, in the form of alarms and visible motions, so that they could avoid any hazardous movement of the machinery.

The Secretary presented testimony at the hearing, however, that unexpected activation (startup or release of stored energy) actually could occur on GM's machines because they were not routinely deenergized or locked out. The judge found the Secretary's evidence insufficient. On review, the Secretary does not specifically contest the judge's factual findings or present an argument for finding unexpected activation on the particular facts of these cases. In fact, the Secretary does not dispute GM's assertion that he has abandoned any objections to the judge's factual findings. On the other hand, the Secretary summarizes the facts and relies on the injury described in Docket No. 91-3116 as evidence of the general risk of injury from failure to lockout.

We affirm the judge's factual findings and his conclusions. GM's expert witness, engineer Richard Parry, testified convincingly that unexpected activation could not occur on any of the machines under GM's procedures. The judge properly relied on Parry's testimony. He did not discredit the contrary testimony of OSHA's compliance officers ("CO's"), but found that Parry had "superior knowledge and experience" regarding the machines at issue. Thus, he concluded that in each case, "Parry's testimony is entitled to greater weight."

Parry testified that the machine involved in each case would shut down immediately if the employee first: (1) opened the interlocked barrier gate around it, as the employees were trained to do, or (2) pushed an emergency stop button. In either event, there could be no unexpected activation. The basic reason is that a time-consuming startup procedure of at least eight steps, most of them obvious to the servicing or maintenance employee, would be required before any hazardous activation of the equipment could occur.

The restart procedures could be performed by an employee other than the one performing servicing or maintenance, but the latter could not help being aware that they were taking place. In Docket No. 91-2973, eight to eleven startup steps were required, and numerous warning bells would sound during the process. In Docket No. 91-3116, twelve startup steps were required and were all performed within 4 feet of the servicing or maintenance employee. In Docket No. 91-3117, eight startup steps, including many movements by the robot arm that would be obvious to the servicing or maintenance employee, would be required before any hazardous movement could occur.(8) As a result, the servicing or maintenance employee would know of, and have plenty of time to avoid, the hazards. Parry's testimony about the multi-step startup procedure completely disposes of the alleged hazards in Docket No. 91-2973. We vacate the citation in that case accordingly.

FOOTNOTE(8)  The startup process in Docket No. 91-3117 was typical of the others. The judge summarized it as follows:

(1) All interlock gates would have to be closed;
(2) The start button on the control console would have to be pushed, which activates the robot system and instructs it to orient itself as to location. This takes some time and, if not in the home position, the robot returns to the home position;
(3) The robot then proceeds to the conveyor area where the bushings are located and picks one up;
(4) The robot then rotates to the press area;
(5) The robot places a bushing into a funnel;
(6) The robot next picks up a dog bone [metal piece into which bushings are inserted] and the other bushing;
(7) The robot returns to the funnel, drops a bushing, and places the dog bone into a fixture; and
(8) Only after the above sequence of tasks is completed will the solenoid valve be signaled to activate the compressed air power to initiate movement of the press.

The CO's testified in Docket Nos. 91-3116 & 91-3117 that one of the hazards was inadvertent activation of the equipment in the event of a short circuit or ground. However, Parry contradicted that testimony. He testified that the electrical model on which the CO's relied, a 2-wire, 110-volt system, was not used by GM. He further testified that a short circuit or ground in GM's system (using 3-wire, 220-volt connections) would result in shutting the machinery down, not starting it up. Parry's testimony establishes that no hazard existed in either case due to possible electrical failure. We vacate the citation in Docket No. 91-3117 based on Parry's testimony regarding that issue and regarding the multi-step startup procedure in that case.

Finally, in Docket No. 91-3116, the Secretary relies on the fact that an employee actually was injured while servicing the machine.(9) The union safety and health representative at the plant, Thomas Ashburn, testified that based on his investigation of the incident, the employee had followed proper procedures. However, Parry testified that the employee could not have followed proper procedures and must have entered the machine area without using the interlocked barrier gate as required. He testified that GM thoroughly investigated and that its engineers could not cause any movement of the machinery to occur once the barrier gate was opened. Parry testified that once that gate was opened or an emergency stop button was hit, there could be no startup of any equipment until another time-consuming, multi-step startup procedure was completed, which inevitably would alert the servicing or maintenance employee.

FOOTNOTE(9)  GM employee Kaye Lowe was injured by a robot while servicing the A-7 module.  It is undisputed that Lowe's injury happened when she was at the dial table servicing the machinery.  She was attempting to release a ball joint that had become jammed at the point where they are greased. Apparently her movement triggered the grease unit on the dial table and it cycled.  Lowe was struck on the back of her forearm with grease and had to be taken to the hospital to have it all removed.

Again, the judge credited Parry's expert testimony about the machine. The judge found that a preponderance of the evidence failed to show that there was a hazard of unexpected energization, etc., in Docket No. 91-3116 because: (1) if the employee had used the barrier gate as required, the 12-step restart procedure would be necessary and would alert the employee in time to avoid any activation of machinery, and (2) if the employee did not use the barrier gate as required, he or she would know that activation could occur at any time--hence, it could not be unexpected. We find that the evidence supports the judge's finding and thus we vacate the citation in Docket No. 91-3116.

Accordingly, we vacate the citations to GM in each of these cases.

Dated: April 26, 1995

Weisberg, Chairman, dissenting:

My quarrel with my colleagues in this case is fundamental. After extensive research, review and comment, OSHA promulgated a standard reflecting its finding that, with certain enumerated exceptions, it is necessary to deenergize machinery and equipment, and to lock or tag it out, prior to servicing or maintenance to effectively protect employees from the hazards of the unexpected energization, startup, or release of stored energy. Locking or tagging out the machinery or equipment is mandatory under the standard if the energy involved is strong enough to result in injury if released unexpectedly. It is undisputed that the energy involved here was sufficient to be considered hazardous if released. Further, GM acknowledges that in the cited instances it has not adhered to the lockout or tagout procedures required under the standard but claims such devices are unnecessary because it has devised a better way of protecting employees. GM does not claim that compliance with the standard would have been infeasible. Nor does GM claim that it lacked fair notice of the Secretary's interpretation of the standard.

In vacating the instant citations my colleagues, through myopic reliance on the so-called "plain meaning" of the term "unexpected" energization, reject the Secretary's interpretation of his own standard in favor of their own interpretation, one that negates the required use of lockout/tagout procedures where the employer has devised an alternative means of protection.(1) Additionally, my colleagues would require that the Secretary prove a negative in every case for each cited machine or equipment, i.e., that in myriad circumstances alternative measures fashioned by the employer would not be completely effective in preventing unexpected energization. The majority's action will encourage a lack of conformity with this important standard, which Elizabeth Dole, President Bush's Secretary of Labor, considered "one of my top safety and health priorities."(2) This will place more workers in danger for, as my colleagues well know, OSHA lacks the resources to inspect even a respectable fraction of the workplaces subject to this standard much less to evaluate all of the alternative measures employers may devise under the exception to the standard the majority creates today. Accordingly, I must dissent.

FOOTNOTE(1)  Nothing in the standard, however, prevents an employer from adding whatever warning mechanisms it feels will further protect employees.  Nor does the standard discourage employers from doing so.

FOOTNOTE(2)  BNA Occupational Safety and Health Reporter, Sept. 6, 1989, at 620.

I find the Secretary's interpretation of his standard to be reasonable and consistent with the protective intent of the standard as explained in the preamble. The Secretary's reasonable interpretations of his standards are entitled to deference. E.g., Martin v. OSHRC (CF & I Steel Corp.), 499 U.S. 144, 150 (1991)), cited in Hackney, Inc., 16 BNA OSHC 1806, 1808, 1994 CCH OSHD paragraph 30,486, p. 42,113 (No. 91-2490, 1994); Martin v. American Cyanamid Co., 5 F.3d 140, 144 (6th Cir. 1993) (reviewing court is to uphold Secretary's interpretation of his standard "unless it is arbitrary, capricious, an abuse of discretion or otherwise not in accordance with law.") Therefore, I would find a violation in each of the three cases. In addition, the facts surrounding the serious injury to the employee who was performing servicing on the energized machinery in Docket No. 91-3116 demonstrate to me that unexpected activation is a hazard in such operations.


The intent of this standard is to reduce, so far as possible, the severe toll of death and injury to servicing and maintenance employees by requiring the most complete control of hazardous energy that is feasible. Some 7.1 percent of all fatalities occurring in general industry relate to failures to adequately control hazardous energy, according to the standard's preamble. Control of Hazardous Energy Sources (Lockout/Tagout): Final Rule, 54 Fed. Reg. 36,644, 36,652, as corrected by 55 Fed. Reg. 38,677 (1990).

The required level of control includes, at a minimum, deenergizing the machine or equipment whenever feasible and locking or tagging it out, before servicing or maintenance work begins:(3) 

FOOTNOTE(3)  The term "control of  hazardous energy" in the standard means lockout or tagout.  The very title of the standard is "[t]he control of   hazardous energy (lockout/tagout)."  The preamble states:

It should be noted that locks and tags by themselves do not control hazardous energy.  It is the isolation of the equipment from the energy source and the following of the established procedures for deenergization and reenergization of the equipment that actually controls the energy.

54 Fed. Reg. at 36,655.  Control circuit devices, on which GM relied here, do not control hazardous energy as the Secretary requires, because they do not isolate the equipment from the energy source.  See 1910.147(b), which defines "energy isolating device" and states that "control circuit type devices are not energy isolating devices."

OSHA believes that the safe performance of activities such as repair, maintenance and servicing, requires the deenergization of machines or equipment whenever feasible. Further, in order to ensure that maintenance or servicing activities are conducted [safely], a lockout or tagout procedure must be utilized.

Id. at 36,654 (emphasis added).(4) Where the energy is insufficient to result in injury if released unexpectedly, the energy is not considered hazardous and the machine or equipment need not be locked out or tagged out. Otherwise, lockout/tagout is mandatory, absent the specific exceptions set forth in the standard.

FOOTNOTE(4)  The Secretary's intent is clearly stated in the opening summary of the standard:

This standard addresses practices and procedures that are necessary to disable machinery or equipment and to prevent the release of potentially hazardous energy while maintenance and servicing activities are being performed.  The standard requires that lockout be utilized for equipment which is designed with a lockout capability except when the employer can demonstrate that utilization of tagout provides full employee protection.

54 Fed. Reg. at 36,644.  See also, e.g., 54 Fed. Reg. at 36,656:

[This standard] requires that the employer develop and implement an energy control program and procedure for servicing and maintenance of machinery and equipment, using lockout or its equivalent on the great majority of energy isolating devices, namely those which are currently capable of being locked out.

Reliance on push buttons and other control circuit devices, however ingenious, is not acceptable under the standard. The reason is that the electrical or other power circuits remain connected (or can be reconnected by someone other than the servicing or maintenance employee). Such circuits are capable of transferring power and causing machine movement, however remote the possibility may seem. As the preamble to the standard makes clear,

OSHA believes that the least desirable situation is to allow employees to perform maintenance, repair, or service activities while the machine or equipment is energized and capable of performing its normal production function....

The vast majority of servicing or maintenance activities can safely be done only when the machine or equipment is not operating and is deenergized; therefore, these activities are covered by this standard.

.... [S]hutting down a machine or equipment usually is not the total solution to the problem. Once the machine or equipment has been stopped, there remains the potential for employee injury from the unanticipated movement of a component of the machine or equipment, or from movement of the material being handled.

.... The generally accepted best means to minimize the potential for inadvertent activation is to ensure that all power to the machine or equipment is isolated, locked or blocked and dissipated at points of control, using a method that cannot readily be removed, bypassed, overridden or otherwise defeated.

54 Fed. Reg. at 36,647-48 (emphasis added).

GM's machines were neither deenergized nor locked out, although it would have been feasible to do both. Thus, the machines remained capable of performing their normal production function while employees performed servicing or maintenance. That is a situation the Secretary seeks to avoid. The serious injury to the employee who was servicing the robot module in Docket No. 91-3116, due to unexpected activation of a robot illustrates the problem and shows that GM's control circuit devices and procedures were not failsafe.

The majority decision is based almost exclusively on the so-called plain meaning of the term "unexpected" energization as used in the standard. The majority appears to read the word "unexpected" as meaning "without warning." Yet Roget's Thesaurus lists the following words as synonyms for "unexpected": unusual, sudden, chance, unanticipated, and unforeseen. R. Chapman, ed., Roget's Intl. Thesaurus (4th ed. 1977). The majority fails to explain how it arrived at its meaning of "unexpected" and why that interpretation is not free from doubt. I do not view the terms "unexpected" and "without warning" as being the same. In the instant cases, for example, while the employees may receive a warning, nevertheless they are exposed to an unanticipated or chance startup. While the employees may be given a momentary "heads up" or warning, they are not necessarily protected against the hazards that the standard is aimed at.

To the extent that a literal reading of the standard by itself suggests that an employer can forego lockout/tagout on a particular machine unless "unexpected energization," as my colleagues define that term, appears to be a realistic possibility, such an interpretation is squarely at odds with OSHA's intent as expressed in the preamble. "[D]efects in ... a regulatory warning may be cured by authoritative judicial or administrative interpretations which clarify obscurities or resolve ambiguities." Diebold, Inc. v. Marshall, 585 F.2d 1327, 1338 (6th Cir. 1978). A standard's preamble "is the best and most authoritative statement of the Secretary's legislative intent." E.g., American Sterilizer Co., 15 BNA OSHC 1476, 1478, 1991-93 CCH OSHD paragraph 29,575, p. 40,016 (No. 86-1179, 1992).

The preamble passages quoted above illustrate the Secretary's rulemaking finding that unexpected energization, etc., always is a hazard in servicing and maintenance work on power machinery and equipment unless employees follow lockout/tagout procedures. The Secretary's supplemental statement of reasons supporting the standard makes the same basic point. It states that "workers face a significant risk of material harm every time they perform service or maintenance work on powered industrial equipment." Final Rule: supplemental statement of reasons, 58 Fed. Reg. 16,612, 16,620 (1993) (citing 54 Fed. Reg. at 36,647-48, 36,652-53).

Thus, that supplemental statement is consistent with the preamble. They both find a significant risk of harm where the employer fails to use the energy control procedure the standard requires. Machines and equipment that in the Secretary's view present no hazard are not regulated, and risks that in the Secretary's view are "minimal and readily controlled" are not subject to lockout/tagout requirements. Numerous specific exceptions to the lockout/tagout requirements are provided in the standard. Examples are:

(1) certain routine, minor tool changes and adjustments during normal production operations, where effective alternative protection is used (section 1910.147(a)(2)(B)--exception); and

(2) work on cord- and plug-related equipment, where the protections mentioned in section 1910.147(a)(2)(iii)(A) exist. Thus, the Secretary's approach does not "essentially remove all limits to the standard's applicability," as the majority states. Rather, it provides a general requirement of lockout or tagout protection with specific, prudent exceptions.

All employees in occupations that perform servicing or maintenance on powered industrial equipment in general industry must use lockout/tagout. E.g., 54 Fed. Reg. at 36,684 ("OSHA classified `at-risk' occupations in the Final Rule as those being held by individuals who would actually perform lockout or tagout") (emphasis added). OSHA estimates that about 3 million employees in the affected industries will service or maintain powered industrial equipment (2 million in high-impact industries and 1 million in low-impact industries).(5) The decision whether to follow lockout/tagout procedures does not rest on whether the employee, supervisor, or compliance officer can think of a way that the particular machine could activate unexpectedly. The only question is whether the energy is strong enough to cause injury if released.

FOOTNOTE(5)  The Secretary's complete analysis, on which his comments in the preamble are based, so states.

OSHA has estimated that about 2 million workers in 340,451 high-impact establishments and almost 1 million workers in 291,034 low-impact establishments are employed in occupations that are at risk when equipment servicing and maintenance tasks are performed. This risk appears to be the greatest for those workers employed as craft workers, machine operators, and laborers.  Moreover, packaging and wrapping equipment, along with printing presses and conveyor belts, are associated with a high proportion of accidents.

OSHA, Regulatory Impact and Regulatory Flexibility Analysis of 29 CFR 1910.147, II-13, II-15 (August, 1989) (emphasis added).  That document also states flatly that "[u]nder the final rule, an employee must lock or tag machinery and equipment during servicing operations and equipment maintenance."  Id. at III-20.

The majority would allow GM to forego lockout/tagout in favor of control circuit devices, even though the Secretary has made clear that those devices do not actually control energy as he requires. See supra note 3. Furthermore, GM's reliance on control circuit devices meant that the safety of the servicing employee was not fully in his or her own control, contrary to the purpose of the standard. The protections provided by those control circuit devices could be removed by another employee, whereas "[l]ockout or tagout shall be performed only by the authorized employees who are performing the servicing or maintenance:" section 1910.147(c)(8).(6)

FOOTNOTE(6)  The term "authorized employee" is defined as follows:

A person who locks out or tags out machines or equipment in order to perform servicing or maintenance on that machine or equipment.  An affected employee becomes an authorized employee when that employee's duties include performing servicing or maintenance covered under this section.

Section 1910.147(b).  The term "affected employee" is defined in that section as follows:

An employee whose job requires him/her to operate or use a machine or equipment on which servicing or maintenance is being performed under lockout or tagout, or whose job requires him/her to work in an area in which such servicing or maintenance is being performed.

These [authorized employees] are the only employees who are required to be trained to know in detail about the types of energy available in the workplace and how to control the hazards of that energy. Only properly trained and qualified employees can be relied on to deenergize and to properly lockout or tagout machines or equipment which are being serviced or maintained, in order to ensure that the work will be accomplished safely.

54 Fed. Reg. at 36,676. Thus, the standard did not contemplate that the servicing employee would be subject to having to get out of the way because another employee could begin a restart procedure.

The majority states, however, that "we find it unnecessary to look outside the standard itself for guidance as to its meaning." I believe that the majority errs in that regard. First, the standard's meaning is "not free from doubt," especially in light of the preamble. See, e.g., CF & I, 499 U.S. at 150 (where meaning of regulatory language is "not free from doubt," reviewing court should give effect to Secretary's interpretation so long as it is reasonable) (citing Ehlert v. United States, 402 U.S. 99, 105 (1971)).

Second, even if the standard's language seemed "free from doubt," the Supreme Court has made clear numerous times recently that, particularly where, as here, the employer cannot claim lack of fair notice of the agency's interpretation, it is the actual intent of the regulator that controls, not necessarily the common meaning of the words used. "The circumstances of the enactment of particular legislation may persuade a court that Congress did not intend words of common meaning to have their literal effect." Watt v. Alaska, 451 U.S. 259, 266 (1981).

That the regulator's intent controls is clear from the following summary of the Supreme Court's requirements--relied on by the Commission in Unarco Commercial Prod., 16 BNA OSHC 1499, 1502-03, 1993 CCH OSHD paragraph 30,294, p. 41,732 (No. 89-1555, 1993).

Under the familiar principles enunciated by Chevron U.S.A. Inc. v. NRDC, Inc., 467 U.S. 837, 842, 104 S. Ct. 2778, 2781, 81 L.Ed. 2d 694 (1984), we first ask "whether Congress has directly spoken to the precise question at issue." We independently examine the language and, if necessary, the legislative history to determine whether the intent of Congress is clear. If congressional intent is unclear, we then inquire whether the agency's interpretation is "permissible," id. at 843, 104 S. Ct. at 2782, i.e., "rational and consistent with the statute." NLRB v. United Food & Commercial Workers Union, Local 23, [484 U.S. 112, 123], 108 S. Ct. 413, 421 [(1987).]

Securities Indus. Ass'n v. Federal Reserve Sys., 847 F.2d 890, 893-94 (D.C. Cir. 1988) (emphasis added). The analytical steps for interpreting an OSHA standard are the same as for a Congressional statute, as Unarco noted. 16 BNA OSHC at 1502-03, 1993 CCH OSHD at p.41,732. Thus, the Commission's job is to discern the intent of the Secretary in issuing the standard, not to decide independently what the language means. The Court cautioned judges "not to make a fortress out of the dictionary; but to remember that statutes always have some purpose or object to accomplish, whose sympathetic and imaginative discovery is the surest guide to their meaning." Public Citizen v. Department of Justice, 491 U.S. 440, 454-55 (1989), (citing Cabell v. Markham, 148 F.2d 737, 739 (2d Cir.), aff'd, 326 U.S. 404 (1945)).

Third, the legislative history of a standard always may be consulted to determine its meaning. "When aid to construction of the meaning of words, as used in the statute, is available, there certainly can be no `rule of law' which forbids its use, however clear the words may appear on `superficial examination." Public Citizen (quoting United States v. American Trucking Assn., 310 U.S. 534, 543-44 (1940) (citations omitted)). Thus, reference to the preamble is always appropriate where the meaning of a standard is in dispute.(7) In my view the preamble here shows clearly that the Secretary intended to avoid, wherever feasible, having employees perform servicing or maintenance on powered industrial machinery while it is energized.

FOOTNOTE(7)  The cases the majority cites do not warrant a different result.  Estate of Cowart v. Nicklos Drilling Co., 112 S. Ct. 2589, 2594 (1992), acknowledges that in certain circumstances the inquiry goes beyond the text of a statute despite its seemingly plain meaning.  Cowart and the other recent Supreme Court cases modify the effect of the statement the majority quotes from Caminetti v. United States, 242 U.S. 470 (1917).

Brown v. Gardner, 115 S. Ct. 552, 556 (1994), involved no positive   legislative history on the disputed interpretation issue.  That case did  not mention, much less overrule, the Court's oft-repeated precedent that  legislative history always may be consulted if it is available.  Further,  Brown concerned the meaning of a veterans benefit statute, and the Court noted that interpretive doubt is to be resolved in the veteran's favor, not the Government's, in such cases.  Id. at 555 (citing King v. St. Vincent's Hosp., 502 U.S. 215, 220-21 n.9 (1991)).

The quotation in Brown from Good Samaritan Hosp. v. Shalala, 113 S. Ct. 2151, 2157 (1991), which the majority notes, supports my position strongly.  The Court's full statement actually was, "[t]he starting point in interpreting a statute is its language, for `[i]f the intent of Congress is clear, that is the end of the matter," quoting Chevron, 467 U.S. at 842 (emphasis added).  The intent of the lockout/tagout standard is what the Secretary says it is.  The preamble clarifies that intent.

The treatise which the majority cites also supports my position.  2A Singer, Sutherland Stat. Const., 48.01 (5th ed. 1992) (citing Train v. Colorado Public Interest Research Group, Inc., 426 U.S. 1, 9-10, 23-24 (1976) (plain meaning rule is not to be used to thwart or distort intent of Congress by excluding from consideration enlightening material from legislative history)).

The burden the majority places on the Secretary by superimposing its own interpretation on the standard likely will prove quite difficult and burdensome to meet. Putting aside the evidence of a recent injury on one of the machines during servicing (discussed below), two experienced OSHA compliance officers ("CO's") testified to specific ways in which they believed GM's machines could activate unexpectedly. They testified that the machines in question in Docket Nos. 91-3116 and 91-3117 could activate if a short circuit occurred in the line. The CO further testified in Docket No. 91-3117 that the press could stroke if an employee manually tripped the solenoid, or if the solenoid failed.

The judge did not discredit that testimony. Still, he found it insufficient based on strong expert testimony to the contrary by GM's engineer, Richard Parry. As the majority notes, the judge found that Parry had "superior knowledge and experience" regarding the equipment in question and thus that his testimony is entitled to greater weight. Parry testified that these particular machines were not subject to the specific hazards that the CO's raised. The majority fully affirms the judge's findings. There is no gainsaying that it will be difficult for the Secretary to prevail in hotly contested cases such as these involving sophisticated control circuit devices. In effect, the Commission is requiring the Secretary to reestablish in every case his reasonable rulemaking finding--that with the specific exceptions spelled out in the standard, unexpected energization, etc., poses a significant risk of harm if powered industrial machinery is not deenergized and locked out before servicing or maintenance is performed on it.


The evidence clearly shows that a hazard of unexpected startup was present on the machinery involved in Docket No. 91-3116, no matter which version of the accident one accepts. That injury occurred while an employee was performing servicing inside a robot module. She was attempting to unjam a ball joint when the adjacent grease unit cycled and struck her arm, injecting grease into it which had to be removed at a hospital.

The CO and the union's safety and health representative testified that, based on their investigations, the employee had followed proper procedures, including opening the electronically interlocked barrier gate before entering the module. GM's expert witness Parry testified, however, that based on his investigation, the employee must have circumvented that gate because when opened, it would shut down the power to the machinery in the module, and would necessitate a time-consuming, 12-step restart procedure before the machinery could activate again.

Even assuming Parry was correct, the employee's injury resulted from "unexpected" activation. The gate could be circumvented--an employee could readily climb through its horizontal rails without opening it. There is no basis in the record to conclude that the employee expected the machine to activate, even if she failed to follow GM's procedures by circumventing the gate. Nor did GM expect the machine to activate.

Employees do not forfeit the protection of the standard merely by making a mistake (such as failing to follow a set procedure). The standard clearly aims to protect such employees by requiring the deenergization of machines or equipment wherever feasible and locking or tagging them out, before the servicing or maintenance is performed. Commission and court precedent recognizes that employees sometimes attempt to circumvent control circuit devices on industrial machinery, and it holds that such employees do not thereby forfeit the protections of the standards. E.g., MRS Printing, Inc., 6 BNA OSHC 2025, 2026, 1978 CCH OSHD paragraph 23,102, p. 27,920 (No. 76-3113, 1978) (to comply with section 1910.212(a)(3), the general point of operation guarding requirement for machines, "an employer must install a guarding device that cannot be easily circumvented by his employees.") See also, e.g., Long Mfg. Co., N. C., Inc. v. OSHRC, 554 F.2d 903 (8th Cir. 1977) (Commission finding of machine guarding violation upheld where employer had installed recognized control circuit safety device (dual hand controls) but it could be and was being circumvented). Thus, I would find a violation in Docket No. 91-3116, even under the majority's interpretation of the standard.

Accordingly, I respectfully dissent.

Dated: April 26, 1995