Entergy Gulf States, Inc. v. John Summers

IN THE SUPREME COURT OF TEXAS

IN THE SUPREME COURT OF TEXAS

════════════

No. 05-0272

════════════

Entergy Gulf States, Inc., Petitioner,

v.

John Summers, Respondent

════════════════════════════════════════════════════

On Petition for Review from the

Court of Appeals for the Ninth District of Texas

════════════════════════════════════════════════════

Argued October 16, 2008

Justice Willett, concurring.

In times of political rancor, vengeful motives are swiftly attributed (and swiftly believed). This is unfortunate, but also unaffecting. The judiciary, rightly understood, is not a political institution but a legal one, meaning we must decide cases on the basis of principled legal points, not political talking points.

This appeal constitutes something of a legal Rorschach test: People see in it what they wish, and one person’s commendable restraint is another’s condemnable activism. Here, one side (Entergy) contends that judicial restraint requires a plain-language reading of the statute, that the surest manifestation of legislative will is found in legislative text. The other side (Summers) brands such a reading judicial activism, that gleaning intent demands a more holistic and embellished approach.

My view is uncomplicated: The law begins with language, and it smacks of Lewis Carroll when critics, voices raised high in derision, inveigh against “judicial activism” because judges refrain from rewriting the text lawmakers chose. This side of the looking glass, reading a statute as enacted is the nadir of activism, not its zenith. It must be stressed that even one of Summers’ amicus supporters concedes the statute can be read in Entergy’s favor[1]     — a result that, sound and fury aside, “will probably not have a substantial impact on the workers’ compensation system as a whole.”[2]    

I agree with the Court’s bottom-line result (and a good deal, though not all, of its analysis), but I write separately because this case raises important issues regarding statutory construction, and the judicial role more generally, that deserve fuller, more head-on treatment.

I. Introduction — Whether Entergy Can Qualify As a “General Contractor”

Today’s issue is simply stated but sharply disputed: Can a premises owner qualify as a “general contractor” under the Texas Workers’ Compensation Act? Amid the spirited debate, two preliminary matters are unchallenged: (1) premises owners who provide workers’ compensation insurance coverage to their own employees are statutorily immune from tort suits over work-related injuries; and (2) general contractors who cover their subcontractors’ employees are also immune. Today’s case presents a hybrid — whether a premises owner can serve as its own general contractor and assert the same exclusive-remedy defense as a contract employer that it asserts as a direct employer.

Consider: Two employees working side by side at a company-owned workplace, performing the same work when the same accident inflicts the same injuries. One worker is the company’s direct employee, the other its contract employee, both having voluntarily elected coverage under the same written, owner-provided workers’ compensation policy. If the owner meets the Legislature’s elastic definition of “general contractor” — written solely in terms of what contractors do, not what they own — then its contract employees are bound by the same agreed-to policy that binds its direct employees. Ownership nowhere proscribes what the Act prescribes.

Two things should drive our analysis — the Legislature’s language, which is open-ended, and this Court’s role, which is not. We must respect policy-laden statutes as written and give wide leeway to the innumerable trade-offs reflected therein. The Act’s definition of “general contractor” is sweeping (“a person who undertakes to procure the performance of work”) and carves out only one narrow exclusion (“a motor carrier that provides a transportation service through the use of an owner operator”). The wording is inclusive in general but exclusive in particular. The pre-1989 Act used a similarly broad definition (with no exclusion) but a companion definition suggested a premises owner could not serve as its own general contractor. Significantly, the Legislature deleted that explicit dual-hat reference in 1989 as part of a substantive overhaul. Today’s Act does not deny the exclusive-remedy defense if the person who procures the work and provides the coverage — the two factors that define “statutory employer” — also owns the jobsite.

I agree with the Court. By “undertak[ing] to procure the performance of work,” Entergy meets the Legislature’s brief-but-broad definition. This, coupled with Entergy’s provision of workers’ comp coverage, confers statutory-employer status.

II. The Legislature’s Chosen Words Dictate the Outcome

The Act’s controlling provisions are straightforward:

·    “General contractor”: Any “person who undertakes to procure the performance of work or a service, either separately or through the use of subcontractors. The term includes a ‘principal contractor,’ ‘original contractor,’ ‘prime contractor,’ or other analogous term.”[3]    

·    “Subcontractor”: Any “person who contracts with a general contractor to perform all or part of the work or services that the general contractor has undertaken to perform.”[4]    

·    Statutory employer: A general contractor “may enter into a written agreement [with a subcontractor] under which the general contractor provides workers’ compensation” coverage to the subcontractor and the subcontractor’s employees,[5]     and such an agreement “makes the general contractor the employer of the subcontractor and the subcontractor’s employees . . . for purposes of the workers’ compensation laws.”[6]    

·    Exclusive remedy: Workers’ compensation benefits are a covered employee’s “exclusive remedy” against an employer for work-related injuries.[7]    

A. Legislative intent is revealed by legislative language

There is one building-block principle this Court has declared repeatedly and emphatically: the “surest guide” to what lawmakers intended is what lawmakers enacted.[8]     We are interpreting words, and where those words are not doubtful, even though their wisdom may be, we are bound to honor them. Accordingly, since intent is driven by text, we must not accept the peculiar view that construing the Act’s definition of “general contractor” by its terms would subvert legislative intent.[9]     Indeed, it is displacing the concreteness of what was actually said with the conjecture of what was allegedly meant that invites activism, a mischievous way for courts to put a finger on the scale (or in the wind) and thus substitute judicial intent for legislative intent. Our place in the constitutional architecture requires fidelity to what lawmakers actually passed.

Consequently, we must focus on what a statute says and, just as attentively, on what it does not say, taking care to honor substantive changes, both additions and deletions, made over the years, and always presuming that the Legislature chose its language carefully.[10]     As for what the Act includes, its definition of “general contractor” is notable for two things: (1) a solitary description (“undertakes to procure the performance of work”), including a non-exhaustive list of synonyms (“‘principal contractor,’ ‘original contractor,’ ‘prime contractor,’ or other analogous term”); and (2) a solitary exclusion (“a motor carrier that provides a transportation service through the use of an owner operator”). Any entity that falls inside the former and outside the latter is shielded from tort liability if it provides workers’ compensation coverage to its contractors’ employees. As for what the Act excludes, we must give effect to the Legislature’s deletion in 1989 of a provision (“contracted with another party”) that contemplated a general contractor contracting upstream with a premises owner.

B. The court of appeals disregarded the Act’s key provisions

1. it ignored the specific definition of “general contractor”

The court of appeals held “Entergy did not establish it had undertaken to perform work or services and then subcontracted part of that work to IMC, as a general contractor would have done.”[11]     To reach this conclusion, the court relied on Williams v. Brown & Root, Inc., a 1997 court of appeals decision stating that “[a] general contractor is any person who contracts directly with the owner.”[12]     The Williams court reasoned an entity that “did not contract with the owner, but instead was the owner” arguably fell outside the definition.[13]     The Williams court, as the Court today notes,[14]     committed a fundamental error, disregarding the Labor Code’s specific definition of “general contractor”[15]     in favor of a more generic definition.[16]     The Legislature often supplies its own dictionary, and where it provides a precise definition, courts must honor that substituted meaning.[17]     Importantly, this admonition holds true even if the Legislature’s technical definition departs from the term’s ordinary meaning.[18]     So while a general contractor may ordinarily be thought to contract with the premises owner — even though, as the Court observes, an owner serving as its own general contractor is “by no means uncommon”[19]     — that construction must give way to the Act’s specific definitions.[20]    

Contrary to the suggestion in Williams that an owner cannot double as a general contractor because it cannot contract with itself, the statute does not blanketly exclude premises owners who otherwise meet the Act’s undemanding criteria.[21]     Nothing in the Act dictates that a premises owner who procures the work and provides the coverage, the only two factors that confer statutory-employer status, lack the same comp-bar immunity granted someone who performs the very same actions but lacks title to the worksite.[22]    

The legal test for determining whether Entergy can invoke the exclusive-remedy defense is not whether the statute explicitly includes “owners.” The test is a simple one: Does Entergy meet Chapter 406's eligibility criteria? The record shows clearly that Entergy “[undertook] to procure the performance of work” from IMC.[23]     Deposition testimony established that Entergy hired IMC to “supplement the Entergy employee workforce” and help perform maintenance, including “water and turbine-related, generator-related work,” at its Sabine plant. Summers’ own summary-judgment response concedes that Entergy “entered into a contract with [IMC] for IMC to perform various maintenance work at Entergy’s plant in Bridge City, Texas.” Entergy undeniably “under[took] to procure the performance of work,” thus meeting the Act’s broad definition, and because it also provided the workers’ compensation policy under which Summers recovered, Entergy is his statutory employer.

2. it utilized a long-discarded definition of “subcontractor”

The court of appeals in this case (and the Williams court) also erred in relying on Wilkerson v. Monsanto Co., a federal district court decision holding that a premises owner cannot be a statutory employer (because it cannot be a general contractor).[24]     Here, too, the mistake concerns a misused statutory definition. Wilkerson, unlike today’s case, was governed by the pre-1989 definition of “subcontractor”: “a person who has contracted to perform all or any part of the work or services which a prime contractor has contracted with another party to perform.”[25]     The Wilkerson court interpreted “contracted with another party” to mean the prime contractor and premises owner must be distinct entities.[26]     The court said this phrase, the law from 1983 until 1989,[27]     meant a general contractor was necessarily an intermediary that contracts both upstream with the premises owner and downstream with the subcontractor. As the owner’s contracts in Wilkerson were all downstream, he could not be a statutory employer.[28]    

Assuming Wilkerson was correctly decided, it lacks any interpretive force today, for a simple reason: Wilkerson turned entirely on four words the Legislature removed during its 1989 substantive rewrite.[29]     Here are the pre- and post-overhaul definitions that, construed together, control our decision:

	prime/general contractor	subcontractor
pre-1989	“the person who has undertaken to procure the performance of work or services” and “‘prime contractor’ includes ‘principal contractor,’ ‘original contractor,’ or ‘general contractor’ as those terms are commonly used”[30]	“a person who has contracted to perform all or any part of the work or services which a prime contractor has contracted with another party to perform”[31]
current	“a person who undertakes to procure the performance of work or a service . . . . The term includes a ‘principal contractor,’ ‘original contractor,’ ‘prime contractor,’ or other analogous term. The term does not include a motor carrier that provides a transportation service through the use of an owner operator”[32]	“a person who contracts with a general contractor to perform all or part of the work or services that the general contractor has undertaken to perform”[33]
key change	current definition excludes a single class of otherwise eligible persons: certain motor carriers, nobody else	current definition no longer imposes an “upstream contract” condition on general contractors

As seen above, the 1989 reform bill deleted “contracted with another party,” the critical upstream-contract phrase that anchored Wilkerson and suggested a premises owner could not wear the hat of general contractor. The before-and-after comparison is difficult to brush aside. While the 1983-1989 Act indicated that a contractor undertook action on behalf of someone else (the owner), the Legislature in 1989 removed that upstream inference. Our cases require us to treat such omissions as meaningful and not meaningless,[34]     a principle even more prudent when deletions occur, as here, within a substantive overhaul that constitutes the lone piece of legislation that lawmakers are considering.[35]     Wilkerson remains instructive only to underscore that statutory construction must honor statutory definitions.

Summers urges a construction rooted in now-repealed language. While conceding that “contracted with another party” appears nowhere in the current statute, Summers insists the upstream-contract notion was not deleted but transplanted, subsumed now by the phrase “undertakes to procure” in the definition of “general contractor.” This contention — that the upstream-contract condition was moved but not removed — is facially counterfactual, betrayed by this inconvenient truth: “undertake[ ] to procure” also appeared in the pre-1989 definition. Even though this phrase predated the 1989 overhaul, Summers argues it became implicitly freighted with what was once explicitly stated (in a different definition). This argument is unpersuasive. Updated criteria require updated analysis. It is untenable that the four words so important in Wilkerson were, though deleted, imported into three words that predated Wilkerson. Summers’ argument would reinsert what lawmakers took out and declare this part of a massive modernization bill — the part that anchors the precedent upon which Summers relies most — wholly nonsubstantive and merely aesthetic.[36]    

We cannot treat the upstream-contract language in the 1983-1989 Act as mere surplusage and its 1989 deletion a nullity. Nor does the dissent pivot on Summers’ argument that “undertakes to procure” necessarily implies an upstream obligation and must be read as “undertakes to procure for someone else.” The deletion of something explicit means more than the retention of something implicit. Indeed, several Texas statutes use “undertake” to describe a person acting to benefit himself.[37]     More to the point, when lawmakers have in mind an entity doing something on another’s behalf, they have no difficulty saying so explicitly, often using “undertakes” in tandem with clear third-party language like “for another person.”[38]     In such instances, including elsewhere in the Labor Code, the Legislature has done more than imply a third-party obligation; it has stated one outright, something lawmakers in 1989 did not do, instead choosing to scrap preexisting third-party language.[39]    

The Act as written bars Summers’ claim, and it merits mention that even certain counsel supporting Summers concede the statutory text can be read in Entergy’s favor: “Based on statutory language alone, reasonable persons may differ on whether a premises owner may also act as a general contractor in the procurement of work and provision of workers’ compensation coverage, thus receiving the exclusive remedy protection from third party actions.”[40]     Thus, we are directed to arguments that look beyond the statute itself.

III. Settled Precedent Bars Summers’ Extratextual Arguments

Summers and his aligned amici contend that several factors outside the Act’s actual language support a more flexible statutory reading. The Court correctly rejects these arguments, and notably the dissent implicitly does likewise.

A. Failed bills predating and postdating the Act’s 1989

overhaul carry no interpretive force

Summers and various amici exhort us to construe language that passed in light of language that failed to pass. As the Court makes clear, we cannot. Precedent from both the United States Supreme Court and from this Court counsel against supplanting unequivocal enacted text with equivocal unenacted inferences drawn from failed legislation.

First, counsel supporting Summers direct us to the 1989 overhaul effort itself. It is undisputed that the 71st Legislature was consumed with the task of restructuring the State’s then-76-year-old workers’ compensation system.[41]     The regular 140-day session failed to produce a reform bill, and Governor Clements immediately called a special session. Summers places great weight on the fact that during this first of two special sessions, House members once considered an omnibus bill that used “owner or general contractor” in section 406.123’s predecessor. House-Senate negotiations collapsed, reportedly over two unrelated issues,[42]     and lawmakers adjourned and went home for several months. Later that year, Governor Clements called a second special session, and in its final hours the Legislature passed Senate Bill 1, which did not expressly include the word “owner,” a fact Summers views as dispositive.

This argument is unavailing, as the United States Supreme Court recently explained: “It is always perilous to derive the meaning of an adopted provision from another provision deleted in the drafting process.”[43]     The word’s momentary presence during Special Session No. 1 and absence several months later during Special Session No. 2 suggests nothing that can override the express terms of the enacted statute.[44]     Under Summers’ position, we must assign great meaning to never-enacted language (“owner”) that appeared in a prior session’s bill draft but no meaning to once-enacted language (“contracted with another party”) that the Legislature affirmatively removed from an on-the-books statute. We cannot bestow all significance on proposed alterations in failed bills while ignoring enacted alterations to the statute itself. Settled law requires the opposite approach, respecting changes to actual statutes and discounting changes to would-be statutes.

Second, counsel supporting Summers ask us to examine post-1989 legislative efforts and conclude that intent to bar premises owners from invoking statutory-employer immunity is implicit in the Legislature’s consideration, but not adoption, of various bills since 1989 related to premises-owner liability.[45]     Summers sees the failure of these measures as tantamount to a legislative command to exclude premises owners from asserting the exclusive-remedy defense.

We cannot draw such an inference for two reasons. First, the Act itself controls, and its definitions include no such exclusion. Far more probative than proposed legislation is passed legislation, what the people’s elected representatives actually enacted as a collective body. The Legislature’s “broad definition, narrow exception” approach to “general contractor” and deletion of the upstream-contract language constitute dual reasons for not barring dual roles for those meeting the Act’s liberal definitional criteria.

Second, we eschew guesswork, and a bill’s failure to pass sheds no light because, as even casual Capitol observers know, bills fall short for countless reasons, many of them “wholly unrelated” to the bill’s substantive merits or “to the Legislature’s view of what the original statute does or does not mean.”[46]     Bills rise and fall for reasons both incalculable and inscrutable, and courts’ reluctance to draw inferences from subsequent legislative inaction is deeply rooted, as explained by the United States Supreme Court a half-century ago: “Such non-action by Congress affords the most dubious foundation for drawing positive inferences. . . . Whether Congress thought the proposal unwise . . . or unnecessary, we cannot tell; accordingly, no inference can properly be drawn from the failure of the Congress to act.”[47]     We, too, reject searching for confirmation or contradiction in later sessions’ unsuccessful bill drafts.[48]     As non-adoption infers nothing authoritative about an earlier statute’s meaning, we do not consult failed bills to divine what a previous Legislature intended.

Even if our precedent allowed us to conflate inaction with intention, the bills, as the Court notes, were not only unsuccessful but immaterial. The bill that comes closest, Senate Bill 1404 from the 76th Legislature in 1999, would have amended “general contractor” to include “an owner or lessor of real property.”[49]     Any relevance ends there. Senate Bill 1404 would have let general contractors (whether owners or not) invoke the exclusive-remedy defense either by providing coverage directly or “by entering into a written agreement with another person under which the other person provides the coverage.”[50]     Today’s question differs significantly: whether a premises owner who meets every current statutory-employer criteria is nonetheless excluded. So while some bills over time would have extended comp-bar immunity to owners (1) who merely require coverage (Senate Bill 1404),[51]     (2) who directly provide coverage but do not also act as their own general contractors (House Bills 2279 and 3024),[52]     (3) who are a parent or subsidiary corporation of an entity that provides coverage (House Bill 3120, House Bill 3459, and Senate Bill 675),[53]     or (4) who are engaged in construction or building with a general contractor and a subcontractor where one of them provides coverage (House Bills 2982 and 1626),[54]     those expansions of immunity are committed to the Legislature’s broad policymaking discretion. They are not today’s case, which examines whether Entergy is disqualified under existing law despite meeting every applicable criteria.

As for Senate Bill 1404, the legislative record is completely bare as to the individual sponsor’s (or anyone else’s) objective. The bill was referred to committee and then left pending; no hearing, no testimony, no bill analysis, no action whatsoever. Even if the bill were on all fours, a single bill — filed the day before the filing deadline[55]     and never heard from again — hardly constitutes the Legislature “repeatedly reject[ing]” the notion of a premises owner acting dually as a general contractor under the Workers’ Compensation Act. In fact, even if there were a failed bill that added “owner” to the existing definition and nothing else, it would be immaterial. Lawmakers may have thought such a bill unwise, or maybe unnecessary. Who knows? Either way, it is imprudent for courts to draw forensic truths from legislative machinations, ascribing intent and motivations based on nothing more than a judge’s hunch as to what 181 autonomous lawmakers collectively had in mind. As Judge Easterbrook observes, “Intent is elusive for a natural person, fictive for a collective body.”[56]    

B. Neither purposive analysis nor off-the-mark representations regarding legislative history can trump the Legislature’s enacted text

On a related front, amici supporting Summers exhort us to throw off our interpretive “shackles” and embrace a “thorough” and “expansive methodology” that relies on various interpretive tools that look beyond the Legislature’s chosen language. Given the lack of textual ambiguity, I reject this eclectic approach.[57]     The text that lawmakers passed is the truest index of legislative will, and the Legislature defines “general contractor” in terms of what a contractor does, not in terms of what a contractor owns. The definition uses the word “owner” exactly one time, to make clear that motor carriers that use owner operators to provide transportation services are excluded. There is indeed an owner-related exclusion in the Act, but it is specific, not general.

Notably, the dissent, while siding with Summers, also declines this nontextual approach. True, we periodically consult external materials when text is nebulous and susceptible to varying interpretations, but even then, we proceed “cautiously,”[58]     mindful that such materials conflict as often as they converge and that our goal is “to solve, but not to create, an ambiguity.”[59]     Even in rare cases where we mine secondary sources to help clarify ambiguity, judges, while not limited to the text, should always be limited by the text.[60]    

Indeed, this case demonstrates vividly the perils of uncritical reliance on legislative history. It is distressing that those citing the legislative record in this case sometimes do so:

$ inaccurately: misstating when key legislative changes to the draft Act occurred;[61]    

$ selectively:        playing up friendly snippets that they believe reinforce a wished-for interpretation and ignoring snippets that subvert it;[62]    

$ misleadingly: mischaracterizing the import of legislative actions;[63]    

And then, when confronted with a tidbit from the record that can be spun Entergy’s way, Summers dismisses it as something uttered mistakenly.[64]    

Laws exist to guide behavior, and by resting on statutory language rather than embarking on a scavenger hunt for extratextual clues prone to contrivance,[65]     we ensure that everyday Texans struggling to decode the law and manage their affairs consistent with it can rely on a statute “to mean what it says,”[66]     without having to hire lawyers to scour the legislative record for unexpressed (and often contradictory) indicia of intent. As we recently held, if text is not hazy, we must resist morphing statutory construction into statutory excavation and instead “take the Legislature at its word and not rummage around in legislative minutiae.”[67]    

C. The Act is not “absurd” if injured deemed employees receive the same

relief as injured direct employees

Summers insists we must adopt a relaxed interpretation more consonant with fairness because reading “general contractor” to limit contract workers to the same recovery that direct workers receive would render the term “meaningless and absurd.”[68]     While a looser reading is warranted when a straight-up reading produces a patently nonsensical result (not merely an unpleasant one), this is not such a case.

Under Summers’ reading, a separate contractor would escape tort liability, but a premises owner who performs every contracting-related chore the separate contractor would perform would not. More to the point, a general contractor that oversees work on its own property could not qualify as a general contractor under the Act. That was perhaps true in the 1983-1989 Act, as Wilkerson held, but the Legislature’s top-to-bottom rewrite amended the law.

One can complain that current comp benefits are inadequate, but it is unpersuasive to equate equality — direct and contract employees receiving the same benefits when the employer owns the jobsite — with absurdity.[69]     There is nothing nonsensical (or even uncommon[70]    ) about a premises owner serving as its own general contractor or a reading of the Act that results in expanded jobsite coverage by urging premises owners to secure coverage for their subcontractors’ workers.[71]     The comp system quid pro quo — exchanging uncertain tort recovery for no-fault medical and income benefits — has been the embedded public policy of Texas since Woodrow Wilson became President, and wider coverage — that is, more injured workers receiving such compensation — only advances that policy.

D. Judges have no authority to second-guess the myriad policy judgments codified

in the Workers’ Compensation Act

The 1989 restructuring of the Texas workers’ compensation scheme — labeled “the most divisive legislative endeavor in contemporary Texas politics”[72]     — consumed the 71st Legislature for one regular and “two special sessions fraught with obstinacy and emotion.”[73]     What emerged embodied innumerable and quintessential legislative judgments. The recovery of workers’ comp benefits is dictated by the Legislature’s definitions, not by this Court’s declarations. We must refrain from rewriting the text lawmakers chose, here by reinserting third-party language the Legislature deleted.[74]    

Laid bare, Summers’ core complaint is that benefits under the Act are too stingy. We are ill-equipped to assess this charge. The Act, whatever its alleged shortcomings, embodies century-old public policy, and courts must read the Legislature’s words as enacted, not revise them as desired. “The wisdom or expediency of the law is the Legislature’s prerogative, not ours”[75]     — a fundamental point we recently reaffirmed: “arguments that the statute is unwise or unfair must be addressed to the Texas Legislature.”[76]    

It may be correct that lawmakers in 1989 did not intend to permit a dual-hat role for premises owners. Workers’ comp reform was a Herculean, multiple-session undertaking, one made tougher with short deadlines for drafting and short fuses for drafters. Heaven knows laws sometimes pass quickly amid urgent circumstances with scant discussion, yielding untoward ramifications over time. Recent examples of voting-without-reading abound, including the newly passed $789,000,000,000 (and 1,073-page) American Recovery and Reinvestment Act of 2009, which provided “a rare window into the mad cookery of complex legislation” — the final draft “filled with hand-written copy-editing marks, insertions scrawled in the margins, deletions of whole paragraphs boxed with X’s slashing through them, and a variety of curious hash marks and other annotations.”[77]     Even Evelyn Wood would struggle mightily to read the bill, much less cast an informed vote.

Even when laws are meticulously drafted and thoughtfully debated, legislative handiwork must often bend to a still more powerful force: the law of unintended consequences.[78]     To be sure, people are inventive at finding ways to confound lawmakers’ wishes rather than conform to them. But even if we suspected lawmakers intended to retain a third-party requirement despite deleting third-party language, we could not judicially reinsert the requirement, however desirable as a policy matter.

Legislative text is often elastic, like the “general contractor” definition in this case, but the judicial role is not. When divining what lawmakers intended to do, we must focus on what they in fact did do and presume they meant what their words mean. Where language is not unclear, a judge’s doctrinal toolbox is limited. I do not share the view that reliance on text is pretext, that reading laws as written is mere figleafing to disguise judicial willfulness aimed at imposing ideologically congenial results. Purposive decisionmaking is achieved more readily (and easily) by straying from text than by sticking to it, and hewing to the Legislature’s as-written language has repeatedly led me to results I strongly dislike.

Obviously, if lawmakers in 2009 (or later) dislike the Court’s interpretation of the words their 1989 predecessors chose — or believe their predecessors drafted with imprecision — the remedy, and it is a simple one, rests wholly with them. This is precisely how the separation of powers works among co-equal branches of government. The presumption that lawmakers intended what they enacted is not just required and well-settled but desired and well-founded. It is an accommodation rooted in carefulness, not certitude. The Legislature can easily reinsert an upstream-contract provision if it believes our interpretation is wooden legalism that honors the letter of the law but not its spirit, thus letting premises owners slip through an unintended loophole.

IV. A Brief Take on the Dissent

The Court briefly addresses the dissent’s arguments, but more can be said. The dissent’s chief contention is that lawmakers “expressly tethered” general contractor to other terms that are “commonly understood to mean a person who has contracted with an owner”[79]     — like “‘principal contractor,’ ‘original contractor,’ and ‘prime contractor’ . . . all terms that envision a tripartite relationship” among an owner, a general contractor, and subcontractors.[80]     The dissent acknowledges the listed terms “are not exhaustive” but concludes, rather conclusorily, that the notion of an owner-contractor “is simply not analogous.”[81]    

Like the Court, I find the dissent’s argument unpersuasive, for several reasons. First, the dissent cites the “common usage” provision of the Code Construction Act[82]     in urging a “commonly understood” reading of general contractor. However, the Act’s very next provision stresses that “common usage” must yield to specific legislative definitions.[83]     Thus, “ordinary meanings should be applied only to undefined terms.”[84]     The Legislature enacted a specialized definition of general contractor, and in 1989 deleted not only the upstream-contract condition, but also the injunction to interpret the synonyms for general contractor “as those terms are commonly used.”[85]     If a statute defines a term, “a court is bound to construe that term by its statutory definition only,”[86]     deference that seems especially warranted where, as here, the statute omits an earlier directive to apply common usage. In any case, given the ordinariness of premises owners acting as their own general contractors,[87]     I fail to understand the dissent’s outright rejection of “owner contractor” as dissimilar.

Second, the dissent looks for support in statutory definitions of “contractor” outside the Workers’ Compensation Act that explicitly mention a third-party requirement.[88]     However, none of these cited provisions define general contractor. There exists in Texas statutory law only one definition of this term, Labor Code section 406.121, the provision at issue today. The Act nowhere defines “contractor,” though “independent contractor,” the term most analogous to the non-Act “contractor” provisions cited by the dissent, is defined (immediately below the definition of “general contractor”) as someone “who contracts to perform work or provide a service for the benefit of another.”[89]     The definition by its terms requires an upstream relationship, something the “general contractor” definition does not. If anything, the provisions cited by the dissent, and the “independent contractor” definition in the Act itself, only strengthen the Court’s position, showing that the Legislature is adept at including explicit third-party language when it chooses. The fact that the Legislature did not add third-party language here — even more, it subtracted such language — only fortifies the Court’s interpretation.

Third, the dissent relies on two of our prior cases to assert we have “recognized for almost a century that a contractor” has a third-party requirement.[90]     A careful examination of these cases, however, shows that both cases concern whether an injured worker is an employee or an independent contractor and not whether a premises owner can qualify as a general contractor.[91]     Together, the two cases use the phrase “independent contractor” nineteen times and the phrase “general contractor” none at all. The cases are simply inapposite, though again, by focusing on “independent contractor,” they draw attention to the Act’s current definition of that term, one that on its face requires a third-party relationship, unlike the “general contractor” definition that immediately precedes it.

Finally, the absence of “owner contractor” from a list the dissent concedes is nonexhaustive[92]     (something we must construe liberally) is less notable than the absence of “premises owner” from the Act’s exclusion (something we must construe strictly). The analogous terms seem of a kind and interchangeable, which makes the motor-carriers exclusion seem markedly out of place, suggesting that the definition was otherwise broad enough to capture them. Stated differently, there seemed an awareness that entities beyond the listed terms could fall within the broad definition, but only this narrow class was carved out.

V. A Brief Take on Justice Hecht’s Concurrence

If my understanding is correct, the dissenters reject their original view of the case and now insist a premises owner has never been entitled to statutory-employer status by providing comp coverage to subcontractors and their employees. Conversely, the Court and Justice Hecht apparently believe that access to the exclusive-remedy defense by providing such coverage has been available perhaps since 1917, when a provision was added that is now section 406.124 of the Labor Code, and is certainly available today. My position, detailed above, is that the defense was made available in 1989, when lawmakers removed third-party language from the Act.

The Court’s and Justice Hecht’s attention to section 406.124 and its earlier enacted versions is unhelpful (and unnecessary in my view given the deletion in 1989 of the upstream-contract language). Section 406.124 currently provides that if a “person” hires a “subcontractor,” not for any legitimate reason but instead “with the intent to avoid liability” under the workers’ compensation laws, the scheme will fail because the worker will be deemed the person’s employee. This provision earlier applied to “subscribers” rather than “persons,” but in any case it has always applied to all statutory employers subject to the workers’ compensation laws.

Section 406.124 is a rarely employed subterfuge provision intended to thwart sham attempts by an employer to mischaracterize an employee as a subcontractor and thereby avoid comp liability. It says as much — applying to all statutory employers and targeting efforts to evade liability. The general-contractor provision at issue in today’s case, currently section 406.123, addresses the separate matter of extending statutory-employer status to a general contractor who retains a legitimate, independent subcontractor and wishes to cover the subcontractor’s employees. The general-contractor provision was added in 1983, and substantively rewritten in 1989, as I discuss above.

I essentially agree with Justice O’Neill on this point. None of the parties rely on section 406.124, and it is irrelevant to the key inquiry: whether a premises owner can be a general contractor under section 406.123. The latter provision has never applied to all subscribers, but is limited to general contractors. Today’s issue is whether a premises owner can fall within the definition of “general contractor,” a subset of all subscribers.

So while I agree with the Court’s result and most of its reasoning, I part company with the Court and Justice Hecht on the relevance of the subterfuge provision and its history, even though the changes to this provision and the eventual enactment of the general-contractor provision share a common legislative ancestry to some extent. In short, I see less ambiguity than Justice Hecht does in the general-contractor and subcontractor definitions. (Interestingly, he attaches no significance to the Legislature’s 1989 deletion of “contracted with another party.”) If anything, his meticulous effort to lay out the history of sections 406.123 and 406.124 and their interplay convinces me, more than ever, that we should focus on the text as enacted (and amended) and resist entreaties to meditate on the varying motives and atmospherics that may have spurred the thousand-plus Texas legislators who have dealt with workers’ compensation over the past ninety-six years. I simply do not share Justice Hecht’s “ambiguity” diagnosis,[93]     though I certainly share his aversion to divorcing text, plain or not, from context.[94]    

For the reasons discussed in Parts I-IV above, I disagree that the Act can be read either way and thus requires a gestalt examination of a near-century of legislative machinations for whatever authoritative lessons can be gleaned from that odyssey, interesting though it may be.[95]     Again, I would hold that the general-contractor provision, unlike its pre-1989 version, does not forbid a premises owner from acting as its own general contractor.

VI. Conclusion

Courts are charged with exercising judgment, not will,[96]     and judicial judgment — awareness of the line between adjudication and legislation and refusing to cross it — means giving wide berth to legislative judgment. On policy matters, we must aim for utter disinterestedness, meaning we must interpret the Act as it is written, not as we might have written it. The Texas Workers’ Compensation Act is replete with countless policy trade-offs, and our confined role, one defined chiefly by limits and duties, not by powers, is to construe statutes as we find them, not to second-guess or refine them.

The Court has reached the correct result, and for the reasons discussed above, I join all but Parts IV, V, and VIII of its decision.

___________________________________

Don R. Willett

Justice

OPINION DELIVERED: April 3, 2009

---

^[1]     Written Testimony of the Texas Association of Defense Counsel: Hearing on Interim Charge Number Eight Before the Senate State Affairs Comm., 80th Leg., Interim (April 28, 2008).

^[2]     According to legislative testimony from this pro-Summers amicus:

Whatever decision the court ultimately makes in the Entergy case will probably not have a substantial impact on the workers’ compensation system as a whole. . . . [O]nly relatively large owners or contractors can afford to administer the kinds of insurance programs involved in the case, so we do not expect a sudden shift in this direction.

Id.

^[3]     Tex. Lab. Code § 406.121(1).

^[4]     Id. § 406.121(5).

^[5]     Id. § 406.123(a).

^[6]     Id. § 406.123(e).

^[7]     Id. § 408.001(a).

^[8]     Leland v. Brandal, 257 S.W.3d 204, 206 (Tex. 2008) (quoting Fitzgerald v. Advanced Spine Fixation Sys., 996 S.W.2d 864, 866 (Tex. 1999)).

^[9]     Many observers, including lawyers, often conflate legislative intent with legislative history. They are distinct. Under our cases, determining intent is the objective, and where text is clear, text is determinative. Legislative history is a device some judges use to discern intent when text is unclear.

^[10]     See Cameron v. Terrell & Garrett, Inc., 618 S.W.2d 535, 540 (Tex. 1981); Eddins-Walcher Butane Co. v. Calvert, 298 S.W.2d 93, 96 (Tex. 1957).

^[11]     __ S.W.3d __.

^[12]     Id. (quoting Williams v. Brown & Root, Inc., 947 S.W.2d 673, 677 (Tex. App.—Texarkana 1997, no writ)) (internal quotation marks omitted) (alteration in original).

^[13]     Williams, 947 S.W.2d at 677.

^[14]     __ S.W.3d __.

^[15]     The Williams court quoted the then-applicable statutory definition of “general contractor,” 947 S.W.2d at 676, which is virtually identical to the current definition.

^[16]     Id. at 677 (“A general contractor is any person who contracts directly with the owner, the phrase not being limited to one undertaking to complete every part of the work.”) (quoting 17 C.J.S. Contracts § 11 (1963)) (internal quotations omitted).

^[17]     Tex. Gov’t Code § 311.011(b).

^[18]     See Tijerina v. City of Tyler, 846 S.W.2d 825, 827 (Tex. 1992).

^[19]     __ S.W.3d __.

^[20]     The 1983-1989 definition of “prime contractor” made clear it was using the term, and similar terms, “as those terms are commonly used.” See Act of May 28, 1983, 68th Leg., R.S., ch. 950, § 1, sec. 6(c), 1983 Tex. Gen. Laws 5210, 5210 amended by Act of Dec. 12, 1989, 71st Leg., 2d C.S., ch. 1, § 3.05(a)(2), 1989 Tex. Gen. Laws 1, 15. This “commonly used” admonition was deleted during the 1989 rewrite, and the current Act uses “or other analogous term.” Tex. Lab. Code § 406.121(1).

^[21]     See Tex. Lab. Code § 406.121(1).

^[22]     The Court is not alone in holding that a premises owner can act as its own general contractor for purposes of a workers’ compensation statute. The Supreme Court of Tennessee in Brown v. Canterbury Corp., 844 S.W.2d 134 (Tenn. 1992), considered whether an owner may nonetheless qualify as a principal contractor under the Tennessee workers’ compensation statute (which like ours does not mention “owner”). See Tenn. Code Ann. § 50-6-113. The court acknowledged that earlier cases “created a distinction between an owner of property and a general contractor, holding that an entity could be considered a principal contractor within the meaning of the workers’ compensation act only if it performed work ‘for another.’” Brown, 844 S.W.2d at 137. The court rejected these older cases, noting that “more recent decisions” allowed an owner to act as its own principal contractor. Id. Tennessee revisited the issue more recently in Rucker v. Rockwood Electric Utilities. No. 03S01-9511-CH-00127, 1996 WL 626292, at *3 (Tenn. Oct. 30, 1996) (not designated for publication). Pursuant to Tennessee Supreme Court Rule 4(A)(3), an opinion of the Special Workers’ Compensation Appeals Panel is not published unless a majority of the Tennessee Supreme Court votes for it to be published. In that case, the Special Workers’ Compensation Appeals Panel of the Supreme Court found that Rockwood Electric Utilities, an owner, was acting as a statutory employer. Id. The Panel relied on an earlier Tennessee Supreme Court opinion that specifically rejected a third-party requirement. Id. (citing Stratton v. United Inter-Mountain Tel. Co., 695 S.W.2d 947 (Tenn. 1985)). Similarly, the Supreme Court of Florida recognizes that a premises owner is entitled to comp-bar immunity “where an owner assumes the role of contractor and employer and, consequently, the duty to provide workers’ compensation benefits.” Ramos v. Univision Holdings, Inc., 655 So. 2d 89, 90 (Fla. 1995).

^[23]     “Procure” means “to obtain by care or effort, to acquire.” Oxford American Dictionary 533 (1980).

^[24]     782 F. Supp. 1187, 1188 (E.D. Tex. 1991).

^[25]     Id. at 1189 (emphasis added) (relying on Act of May 28, 1983, 68th Leg., R.S., ch. 950, § 1, sec. 6(b), 1983 Tex. Gen. Laws 5210, 5210 (previously codified at Tex. Rev. Civ. Stat. art. 8308-3.05)).

^[26]     Id. at 1188–89.

^[27]     Act of May 28, 1983, 68th Leg., R.S., ch. 950, § 1, sec. 6(b), 1983 Tex. Gen. Laws 5210, 5210 amended by Act of Dec. 12, 1989, 71st Leg., 2d C.S., ch. 1, sec. 3.05(a)(5), 1989 Tex. Gen. Laws 1, 15, codified as Tex. Lab. Code § 406.121(5).

^[28]     Another distinction between Wilkerson and this case: In Wilkerson, the contract between the owner and the plaintiff’s direct employer affirmatively disclaimed any employment relationship between the owner and the contractor’s employees, 782 F. Supp. at 1188, while in this case, the contract language explicitly identifies Entergy as the “principal employer” and reserves to Entergy the right to assert statutory-employer status against contract employees’ personal-injury suits.

^[29]     See Act of May 28, 1983, 68th Leg. R.S., ch. 950, § 1, sec. 6(b), 1983 Tex. Gen. Laws 5210, 5210 (amended 1989).

^[30]     See Act of May 28, 1983, 68th Leg. R.S., ch. 950, § 1, sec. 6(c), 1983 Tex. Gen. Laws 5210, 5210, amended by Act of Dec. 12, 1989, 71st Leg., 2d C.S., ch. 1, sec. 3.05(a)(2), 1989 Tex. Gen. Laws 1, 15.

^[31]     See Act of May 28, 1983, 68th Leg., R.S., ch. 950, § 1, sec. 6(b), 1983 Tex. Gen. Laws 5210, 5210 (amended 1989).

^[32]     Tex. Lab. Code § 406.121(1).

^[33]     Id. § 406.121(5). Today’s definition of “subcontractor” was tweaked slightly (and nonsubstantively) as part of the 1993 codification of the Labor Code. Act of May 12, 1993, 73rd Leg., R.S., ch. 269, § 1, sec. 406.121(5), 1993 Tex. Gen. Laws 987, 1158. It is derived almost verbatim from the 1989 overhaul’s definition. Act of Dec. 12, 1989, 71st Leg., 2d C.S., ch. 1, § 3.05(a)(5), 1989 Tex. Gen. Laws 1, 15.

^[34]     See In re Ament, 890 S.W.2d 39, 41 (Tex. 1994) (per curiam); Cameron v. Terrell & Garrett, Inc., 618 S.W.2d 535, 540 (Tex. 1981). In a sense, Entergy hired IMC to help Entergy fulfill an upstream obligation, a statutory one at that: to “furnish service, instrumentalities, and facilities that are safe, adequate, efficient, and reasonable.” See Tex. Util. Code § 38.001.

^[35]     We presume that lawmakers enact statutes “with complete knowledge of the existing law and with reference to it,” Acker v. Tex. Water Comm’n, 790 S.W.2d 299, 301 (Tex. 1990), and that any omissions are intentional, Cameron, 618 S.W.2d at 540.

^[36]     One amicus, citing the Sherlock Holmes story of the dog that did not bark, Sir Arthur Conan Doyle, Silver Blaze, The Memoirs of Sherlock Holmes (1894), argues it is more likely the Legislature intended to change a statute by adding words than by deleting them: “That is because the deletion of words is more likely to have been done by mistake and deletions are more difficult for legislators to recognize during the legislative process than added words.” Unlike Holmes, we are not studying the import of conspicuous silence, but the import of conspicuous deletion. The Legislature in 1989 affirmatively removed words indicating an upstream contract. It may well be true, as the amicus asserts, that “omitted language is less likely to come to the consciousness of legislators in the often chaotic process of legislating,” but chaotic or not, Texas law requires us to accept that lawmakers acted intentionally, not inadvertently. However clamorous the Capitol was in 1989, our cases forbid any presumption that the Legislature was inattentive.

^[37]     See, e.g., Tex. Agric. Code § 12.040(d)(3)(C) (“a long-term plan outlining the steps the community will undertake to maintain its desirability . . . .”); Tex. Ins. Code § 751.002(b) (“. . . the department or commissioner . . . may undertake market analysis or market conduct action . . . .”); Tex. Loc. Gov’t Code § 232.093(e) (“Before a planning commission member undertakes the duties of the office . . . .”); Tex. Transp. Code § 454.001(b)(1) (“A municipality . . . may undertake research, development, and demonstration projects . . . .”); Tex. Util. Code § 52.256(b)(3) (“. . . the telecommunications utility will undertake to achieve each of these initiatives . . . .”).

^[38]     See, e.g., Tex. Occ. Code § 1702.108 (“A person acts as a guard company . . . if the person . . . undertakes to provide . . . for another person . . . .”); Tex. Local Gov’t Code § 176.001(1) (“‘Agent’ means a third party who undertakes . . . for another person . . . .”); Tex. Gov’t Code § 27.006(a)(1) (“A justice commits an offense if the justice: . . . undertakes . . . for another . . . .”); Tex. Lab. Code § 21.002(9)(A) (“‘Employment Agency’ means a person . . . who regularly undertakes . . . to procure: (A) employees for an employer . . . .”).

^[39]     Summers and some of his aligned amici advance a safety-related argument, predicting the “complete destruction of the incentive to make the workplaces safe” if Summers loses. But whether one scheme promotes workplace safety over the other is a legislative call, not ours. In any case, the record shows that Entergy employees work alongside IMC employees, and the Act gives those protected by the exclusive-remedy defense concrete incentives to minimize job-related risks. See Tex. Ins. Code. §§ 2053.001-.054; Lee Lewis Constr., Inc. v. Harrison, 70 S.W.3d 778, 795 (Tex. 2001) (Hecht, J., concurring) (a contractor’s employer “has the same economic incentive the contractor has to minimize job-related risks to workers”). Premiums are tied to accident records, and rates are higher for unsafe employers. Also, regulations from the Occupational Safety and Health Administration cover Entergy’s direct workers just as fully as they cover Entergy’s contract workers. See 29 U.S.C. § 654. Finally, nothing in the record suggests the exclusive-remedy defense has spurred employers to care less about preventing jobsite accidents than those who face common-law liability.

^[40]     Written Testimony of the Texas Association of Defense Counsel: Hearing on Interim Charge Number Eight Before the Senate State Affairs Comm., 80th Leg., Interim (April 28, 2008).

^[41]     See Jill Williford, Comment, Reformers’ Regress: The 1991 Texas Workers’ Compensation Act, 22 St. Mary’s L.J. 1111, 1124–25 (1991).

^[42]     See John T. Montford et al., A Guide to Texas Workers’ Comp Reform 3 (1991) (reporting that conferees agreed on fourteen of sixteen chapters in the bill, “but there were colossal differences in the dispute resolution and benefits proposals”).

^[43]     District of Columbia v. Heller, 128 S.Ct. 2783, 2796 (2008).

^[44]     Nothing in the amendment itself defined “owner,” and despite what one amicus describes as “hundreds of hours of hearings that led to the 1989 overhaul,” not a single word was devoted to this single word. The legislative record lacks any signal as to what even one lawmaker thought about expressly including owners, whether it was thought ill-advised or redundant or even thought about at all.

^[45]     See __ S.W.3d __.

^[46]     Tex. Employment Comm’n v. Holberg, 440 S.W.2d 38, 42 (Tex. 1969) (“we attach no controlling significance to the Legislature’s failure to enact the proposed amendment”); see also El Chico Corp. v. Poole, 732 S.W.2d 306, 314 (Tex. 1987); Dutcher v. Owens, 647 S.W.2d 948, 950 (Tex. 1983) (warning against gleaning legislative intent from failed bills: “Any such inference would involve little more than conjecture.”).

    Nor, as the Court stresses, __ S.W.3d __, can post-hoc statements by legislators shed light on what a statute means. See Consumer Prod. Safety Comm’n v. GTE Sylvania, Inc., 447 U.S. 102, 117-18 (1980) (disregarding such statements about an earlier-passed statute: “the views of a subsequent Congress form a hazardous basis for inferring the intent of an earlier one”). The Legislature is composed of 181 diverse members representing diverse areas with diverse priorities; one lawmaker’s perspective may be radically different than that of his or her 180 colleagues. See AIC Mgmt. v. Crews, 246 S.W.3d 640, 650 (Tex. 2007) (Willett, J., concurring) (“The statute itself is what constitutes the law; it alone represents the Legislature’s singular will, and it is perilous to equate an isolated remark or opinion with an authoritative, watertight index of the collective wishes of 181 individual legislators, who may have 181 different motives and reasons for voting the way they do.”); Gen. Chem. Corp. v. De la Lastra, 852 S.W.2d 916, 923 (Tex. 1993) (“[T]he intent of an individual legislator, even a statute’s principal author, is not legislative history controlling the construction to be given a statute.”). This explains our consistent view — reinforced by the U.S. Supreme Court, see, e.g., Heller, 128 S.Ct. at 2805 — that post-passage actions and comments are immaterial:

[C]ourts construing statutory language should give little weight to post-enactment statements by legislators. Explanations produced, after the fact, by individual legislators are not statutory history, and can provide little guidance as to what the legislature collectively intended.

In re Doe, 19 S.W.3d 346, 352 (Tex. 2000) (quoting C & H Nationwide, Inc. v. Thompson, 903 S.W.2d 315, 328-29 (Tex. 1994) (Hecht, J., concurring and dissenting) (citations omitted)). The very notion of “subsequent legislative history” is oxymoronic. After-the-fact comments may constitute history, and they may concern legislation, but they are not part of the legislative history of the original enactment. See Rogers v. Frito-Lay, Inc., 611 F.2d 1074, 1082 (5th Cir. 1980). Judge Posner offers a grave caution: Judges who credit “subsequent expressions of intent not embodied in any statute may break rather than enforce the legislative contract.” Richard A. Posner, The Federal Courts 270 (1985). Judges also risk getting snookered. See, e.g., Am. Hosp. Ass’n v. NLRB, 899 F.2d 651, 657 (7th Cir. 1990) (“Post-enactment legislative history . . . is sometimes a sneaky device for trying to influence the interpretation of a statute, in derogation of the deal struck in the statute itself among the various interests represented in the legislature. Courts must be careful not to fall for such tricks and thereby upset a legislative compromise.”) (citations omitted). Finally, even proponents of legislative history, even those proponents willing to consider legislators’ post-enactment comments, disregard statements from legislators who did not hold office when the disputed statute was enacted.

^[47]     United States v. Price, 361 U.S. 304, 310-11 (1960). See also Perez v. United States, 167 F.3d 913, 917 (5th Cir. 1999 ) (“deductions from congressional inaction are notoriously unreliable”).

^[48]     Nat’l Liab. & Fire Ins. Co. v. Allen, 15 S.W.3d 525, 527 (Tex. 2000) (“If possible, we must ascertain the Legislature’s intent from the language it used in the statute and not look to extraneous matters for an intent the statute does not state.”).

^[49]     See Tex. S.B. 1404, 76th Leg., R.S. (1999).

^[50]     Id.

^[51]     Id.

^[52]     See Tex. H.B. 2279, 74th Leg., R.S. (1995); Tex. H.B. 3024, 75th Leg., R.S. (1997).

^[53]     See Tex. H.B. 3120, 77th Leg., R.S. (2001); Tex. H.B. 3459, 77th Leg., R.S. (2001); Tex. S.B. 675, 78th Leg., R.S. (2003).

^[54]     See Tex. H.B. 2982, 78th Leg., R.S. (2003); Tex. H.B. 1626, 79th Leg., R.S. (2005).

^[55]     See Tex. S. Rule 7.07(b), Tex. S. Res. 14, 81st Leg., R.S., 2009 S.J. of Tex. 21, 23.

^[56]     Frank H. Easterbrook, Text, History, and Structure in Statutory Interpretation, 17 Harv. J.L. & Pub. Pol’y 61, 68 (1994).

^[57]     As the Court reaffirms today, if the Legislature’s words are not ambiguous, they are not only the best evidence of legislative intent but the exclusive evidence. __ S.W.3d __. Even if we accepted the invitation to utilize extratextual aids to divine a more embellished meaning, it would make little difference here. The record surrounding the Act’s adoption and the nine failed post-1989 bills lacks any indication that lawmakers meant to disqualify owners from acting as their own general contractors under the Act.

^[58]     Alex Sheshunoff Mgmt. Servs. v. Johnson, 209 S.W.3d 644, 652 n.4 (Tex. 2006).

^[59]     United States v. Shreveport Grain & Elevator Co., 287 U.S. 77, 83 (1932) (quoting Hamilton v. Rathbone, 175 U.S. 414, 421 (1899)).

^[60]     Justice Frankfurter cautioned against what he called “loose judicial reading”: “Loose judicial reading makes for loose legislative writing. It encourages the practice illustrated in a recent cartoon in which a senator tells his colleagues ‘I admit this new bill is too complicated to understand. We’ll just have to pass it to find out what it means.’” Felix Frankfurter, Some Reflections on the Reading of Statutes, 47 Colum. L. Rev. 527, 545 (1947).

^[61]     For example, one amicus curiae mistakenly asserts the Legislature deleted the upstream-contract language during the 1993 nonsubstantive codification. This misstep matters, as the amicus argues from this false premise that since “contracted with another party” — a phrase the amicus bolds and labels “enlightening” — was omitted in 1993, its absence cannot aid Entergy since “the Legislature intended no substantive change to the law by its 1993 codification.” Thus, since lawmakers wanted to maintain the status quo, we must read back into the Act a critical phrase the Legislature deleted. One problem: the upstream-contract provision, which is indeed “enlightening,” was deleted not during the 1993 nonsubstantive codification but during the 1989 substantive overhaul, when lawmakers enacted a raft of major changes. See Act of May 28, 1983, 68th Leg., R.S., ch. 950, § 1, sec. 6(b), 1983 Tex. Gen. Laws 5210, 5210 (amended 1989). Interestingly, this amicus apparently realized its error because its supplemental brief acknowledges that “contracted with another party” was deleted during the 1989 reform when “subcontractor” was rewritten, “though not extensively,” the amicus now insists.

    Another amicus brief contains at least two factual missteps in recounting the Legislature’s 1989 deliberations. First, the brief inaccurately states, “In the draft bill considered during the regular session, immunity was extended to owners, as well as general contractors.” The word “owner” never appeared during the regular session; it appeared in a House revision during the first-called special session. Tex. S.B. 1, 71st Leg., 1st C.S. (1989); see also H.J. of Tex., 71st Leg., 1st C.S. 76 (1989). The brief then errs again, stating “in the subsequent special session [the Legislature] removed premises owners from the list of actors granted immunity.” Nobody in either chamber removed “owner” in the first-called special session; just the opposite, “owner” was momentarily added during this special session (and was absent from the bill adopted during the next session several months later). Id. Unless challenged by opposing counsel, such inaccuracies, however inadvertent, carry real potential to mislead judges and skew judicial decisionmaking.

^[62]     See infra note 64. It is unsurprising that advocates accentuate the positive and eliminate the negative. Such cherry-picking, as Judge Harold Leventhal famously observed, is “the equivalent of entering a crowded cocktail party and looking over the heads of the guests for one’s friends.” Conroy v. Aniskoff, 507 U.S. 511, 519 (1993) (Scalia, J., concurring).

^[63]     As noted above, supra note 61, one brief states the Legislature “removed premises owners from the list of actors granted immunity.” This suggests the Legislature took targeted action to eliminate “owner” from a pending draft. The legislative record belies this suggestion. Nobody “removed” the word “owner” during the second special session, because no bill that session ever included the word “owner.” You cannot remove what does not exist, subtracting what is not there. The locution implies that lawmakers in 1989 took proactive steps to strip “owner” from a pending bill, but “removed” is not the same as “not included.” Nothing in the record shows specific action to remove “owner” from any bill during that final special session. The word was absent altogether, though the record contains nothing to suggest why.

    Another brief follows suit, stating that during the first special session, after the House passed a bill that contained “owner,” the Senate “refused to concur in the House amendments.” While literally true — the Senate did in fact reject House changes — it suggests senators balked specifically at “owner.” Not exactly. After the Senate passed Senate Bill 1, the House struck everything below the enacting clause, substituted an earlier House-passed version (which differed markedly from the Senate version) and sent it back across the rotunda. See H.J. of Tex., 71st Leg., 1st C.S. 76 (1989). Nothing suggests the Senate refused to concur because of “owner.” In fact, a treatise (co-authored by the lead Senate sponsor) states the Senate’s refusal rested on entirely separate issues. See Montford, supra note 42, at 3 (noting consensus stalled due to “colossal differences in the dispute resolution and benefits proposals”). Still another brief asserts that “efforts in prior drafts to include the term ‘owner’ were rejected.” This wording suggests multiple targeted efforts to erase “owner” from pending drafts, but the lone mention of “owner” came in the first special session, in a bill that died over unrelated issues. The brief states a House amendment “specifically included the word ‘owner’” and the Senate “refused to concur in the House amendment, and the bill failed.” The inference is misleading: the House amendment was not a rifle-shot inclusion of “owner” but a wholesale substitution of its earlier bill, with all the “colossal differences” noted by Senator Montford.

^[64]     Just before final passage, Senator Dickson, who opposed tort immunity for general contractors, attached a floor amendment stating that a subcontractor’s workers could bring a “third party action . . . against said general contractor.” Senator Dickson contended that, absent his amendment, a refinery owner would be immune if an accident killed a subcontractor’s employees: “Yes sir it seems to me . . . [that] no matter how negligent, no matter how much at fault that operator was, and no matter how many people he killed, he wouldn’t be liable.” Debate on Tex. S.B. 1 on the Floor of the Senate, 71st Leg., 2d C.S. 17 (Nov. 20, 1989) (transcript available from Senate Staff Services Office).

    Senator Edwards: So in a case like the Phillips refinery explosion, if Phillips had been negligent and your amendment wasn’t law, even though dozens of people were killed, Phillips wouldn’t be liable in any way for their negligence?”

    Senator Dickson:       He would be immune. Would not be liable according to Senator Glasgow’s construction and my reading of this new statute.

Id. Senator Dickson’s pro-plaintiff amendment was later removed. This colloquy did not address whether “general contractor” included owners, but it is notable that Senator Edwards’s hypothetical presumed the owner and the general contractor were one and the same. If that contradicted the Senate’s collective intent, no senator rose to correct it. As Senator Dickson described his own amendment, it was to ensure that an owner’s injured contract worker would not be limited to comp benefits. Summers says Senator Dickson was “simply confused” and later “got sort of educated” that the bill did not contemplate owners doubling as general contractors. The floor transcript reveals no such enlightenment.

    While this discussion is not dispositive (or even relevant) — Senator Dickson’s view is merely his alone, not fairly attributable to his 180 colleagues — it posed the issue presented here: a premises owner acting as a general contractor. If Senator Dickson was off-target because the Legislature never intended to let owners claim general-contractor status, no senator called it to the Senate’s attention. The only reason I mention it is to stress how manipulable legislative history can be (by lawyers and judges alike), and that its indeterminacy is only made worse by the selectivity with which it is utilized.

^[65]     It is not hard to imagine a legislator voting for a bill she actually opposes and then reading into the record a restrictive interpretation that aims to blunt the bill’s real-world impact. Looking at today’s case specifically, a crafty lawmaker who wanted “general contractor” construed narrowly could file a bill that explicitly added the word “owner” to the definition, then quietly urge that the bill stay buried in committee so a future litigant can cajole a judge into believing that the Legislature’s failure to pass the bill evinces legislative rejection of an owner-included definition.

^[66]     Fitzgerald v. Advanced Spine Fixation Sys., 996 S.W.2d 864, 866 (Tex. 1999).

^[67]     Alex Sheshunoff Mgmt. Servs. v. Johnson, 209 S.W.3d 644, 652 n.4 (Tex. 2006).

^[68]     We enforce the Legislature’s words as written unless such a reading would produce absurd results. Fleming Foods of Tex., Inc. v. Rylander, 6 S.W.3d 278, 284 (Tex. 1999).

^[69]     Summers argues the Court’s ruling would work an absurdity by opening up workers’ compensation benefits to a neighborhood child who rakes your leaves. The Act expressly contradicts that argument, speaking directly to “a person employed as a domestic worker or a casual worker engaged in employment incidental to a personal residence.” Tex. Lab. Code § 406.091(a)(1). The Act exempts such employees from coverage, though it allows employer-homeowners to voluntarily accept the rights and responsibilities of providing such coverage. Id. § 406.091(b). It would be curious to brand legally “absurd” something the statute itself permits. Comp-covered lawnboys are possible under the Act because the Legislature, not this Court, explicitly says so.

^[70]     __ S.W.3d __.

^[71]     One side point: While recovery of workers’ compensation benefits is a covered worker’s (or his legal beneficiary’s) exclusive remedy in the event of a work-related injury or death, Tex. Lab. Code § 408.001(a), the very next subsection makes clear that a worker’s surviving spouse or heirs may sue for exemplary damages if the death “was caused by an intentional act or omission of the employer or by the employer’s gross negligence,” id. § 408.001(b).

^[72]     Montford, supra note 42, at 1.

^[73]     Williford, supra note 41, at 11-12.

^[74]     Our precedent reaching back a century demands judicial modesty, and for such modesty to take root, “[c]ourts must take statutes as they find them. More than that, they should be willing to take them as they find them.” Simmons v. Arnim, 220 S.W. 66, 70 (Tex. 1920).

^[75]     Tex. Workers’ Comp. Comm’n v. Garcia, 893 S.W.2d 504, 520 (Tex. 1995) (quoting Smith v. Davis, 426 S.W.2d 827, 831 (Tex. 1968)).

^[76]     In re Jorden, 249 S.W.3d 416, 424 (Tex. 2008). We sometimes weigh the Legislature’s power to enact a statute, but, heeding Justice Holmes’s admonition, we never weigh its prudence: “We fully understand . . . the very powerful argument that can be made against the wisdom of the legislation, but on that point we have nothing to say, as it is not our concern.” Noble State Bank v. Haskell, 219 U.S. 575, 580 (1911).

^[77]     Posting of David M. Herszenhorn to the Caucus: The Politics and Government Blog of the Times, http://thecaucus.blogs.nytimes.com/2009/02/13/ final-draft-on-stimulus-bill-complete-with-last-minute-edits/ (Feb. 13, 2009, 9:52 EST). See also Mark Lander, New Terrain For Arbiters Of a Bailout, N.Y. Times, Nov. 4, 2008, at B1 (describing the no-time-for-reading dynamic surrounding passage of the October 2008 federal financial rescue package). Another classic federal example is the 1989 Budget Reconciliation Act, which consisted of a thousand-plus, unnumbered, and unindexed pages bound together with rope. See Rep. Christopher Cox, Why Congress Doesn’t Work, Part I, Lecture #406 (Sept. 11, 1992) (reprinted at http://www.heritage.org/research/governmentreform/ HL406.cfm):

There was no other copy for any member to look at or read, other than what was in this box. Now I will allow, while I was not able to read it, I was permitted to walk down into the well and gaze upon it from several angles, and even to touch it. When we voted on that bill, at about four o'clock in the morning, not a single member of the House had read it; not a single member of the Senate had read it.

^[78]     One example is the federal luxury tax, imposed in 1990 on everything from furs to yachts. Proponents saw it as a pain-free, palatable and progressive way to raise tax revenue, but the impact was staggering. The boat-building industry was capsized, throwing thousands of blue-collar workers out of work, in turn straining public welfare and unemployment budgets as dislocated workers sought relief. Congress swiftly repealed the tax. Phil Hampton, Boating Casts Off Bad Times, Sales Swell After Luxury Tax Drowns, USA Today, Aug. 19, 1994, at 01B.

^[79]     __ S.W.3d __ (O’Neill, J., dissenting).

^[80]     Id. at __.

^[81]     Id. at __.

^[82]     Id. at __ (citing Tex. Gov’t Code § 311.011(a)).

^[83]     Tex. Gov’t Code § 311.011(b) (“Words and phrases that have acquired a technical or particular meaning, whether by legislative definition or otherwise, shall be construed accordingly.”).

^[84]     Tijerina v. City of Tyler, 846 S.W.2d 825, 827 (Tex. 1992).

^[85]     See supra note 20.

^[86]     Tex. Dep’t of Transp. v. Needham, 82 S.W.3d 314, 318 (Tex. 2002); see also Transp. Ins. Co. v. Faircloth, 898 S.W.2d 269, 274 (Tex. 1995) (“[w]e are bound to construe these terms in accordance with their statutory definitions”).

^[87]     See __ S.W.3d __.

^[88]     __ S.W.3d __ (O’Neill, J., dissenting).

^[89]     Tex. Lab. Code § 406.121(2) (emphasis added).

^[90]     __ S.W.3d __ (O’Neill, J., dissenting) (citing Indus. Indem. Exch. v. Southard, 160 S.W.2d 905, 907 (Tex. 1942); Shannon v. W. Indem. Co., 257 S.W. 522, 524 (Tex. Comm’n App. 1924, judgm’t adopted)).

^[91]     Southard, 160 S.W.2d at 906; Shannon, 257 S.W. at 522-23.

^[92]     Tex. Gov’t Code § 311.005(13) (“‘Includes’ and ‘including’ are terms of enlargement and not of limitation or exclusive enumeration, and use of the terms does not create a presumption that components not expressed are excluded.”).

^[93]     __ S.W.3d __ (Hecht, J., concurring).

^[94]     City of Rockwall v. Hughes, 246 S.W.3d 621, 632 (Tex. 2008) (Willett, J., dissenting) (“The import of language, plain or not, must be drawn from the surrounding context, particularly when construing everyday words and phrases that are inordinately context-sensitive.”).

^[95]     Besides finding little useful in the history of section 406.124, I also disagree with the notion that our decision should turn on a rule that construes statutory ambiguities in favor of workers’ comp coverage and against common-law tort liability. I do not believe we can say the Legislature, actually many different Legislatures, urges a comp-over-tort rule whenever uncertainly arises. The Legislature of course could expressly incorporate such a preference into the Act, but it has not done so. The Workers’ Compensation Act and the common law of negligence are both comprehensive, time-honored systems of compensating injured individuals. While I would reject any suggestion from Summers and his aligned amici that workers’ compensation is a disfavored remedy, likewise I cannot agree that in all cases of doubt the Legislature would have us elevate the statutory system over the common-law system and apply the statutory remedy whenever statutory coverage is unclear.

^[96]     See The Federalist No. 78 (Alexander Hamilton).