Peterson, Presiding Justice, concurring. I concur in the Court’s denial of certiorari in this case. The issues here have gravity, and Cazier makes a number of good points: interpreting the law is the role of courts; judicial deference to executive branch legal interpretations poses a risk to the separation of powers; and, in any event, whatever permissible role there might be for such deference, a law is not “ambiguous” simply because interpreting it is hard. See City of Guyton v. Barrow, 305 Ga. 799, 803-04 (2) (828 SE2d 366) (2019). But the law of the case doctrine prevents us from reaching those important issues here, and so the Court correctly declines to do so. Nevertheless, I write separately to express my growing doubt about our recent precedents requiring judicial deference to executive branch agencies’ interpretation of legal text. My doubt has a constitutional origin: judicial deference to executive branch legal interpretations implicates the Separation of Powers Provision of the Georgia Constitution. See Ga. Const. of 1983, Art. I, Sec. II, Par. III (“The legislative, judicial, and executive powers shall forever remain separate and distinct; and no person discharging the duties of one shall at the same time exercise the functions of either of the others except as herein provided.”); see also Ga. Const. of 1983, Art. VI, Sec. I, Par. I (“The judicial power of the state shall be vested exclusively in the following classes of courts: magistrate courts, probate courts, juvenile courts, state courts, superior courts, state-wide business court, Court of Appeals, and Supreme Court.”); City of Guyton, 305 Ga. at 799 (“At the core of the judicial power is the authority and responsibility to interpret legal text.”). Because the Separation of Powers Provision has been carried forward unchanged through every Georgia Constitution since 1877, and every previous Georgia Constitution contained a similar provision, our historical understanding of judicial deference to the legal interpretations of other branches informs our understanding of whatever limitations our current Constitution may place on the subject. Cf. Black Voters Matter Fund, Inc. et al. v. Kemp, et al., 313 Ga. 375, 396 & n.27 (870 SE2d 430) (2022) (Peterson, J., concurring) (detailing the history of the Provision and suggesting that a century of pre-1983 standing precedent may be “baked into the 1983 Constitution['s]” Provision).[1] But our history of deference is messy; our precedent is all over the place, and has been for nearly the entire existence of our Court. Eight years ago, we announced for the first time in our state’s history that our precedent on judicial deference to executive branch legal interpretations is best understood through the federal lens of Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (104 SCt 2778, 81 LE2d 694) (1984). That recency poses a problem. If — as it appears to me — our post-1983 decisions pronounced deference principles without proper grounding in our cases interpreting the earlier versions of the Constitution, then those post-1983 decisions do not shed light on the original public meaning of the current Separation of Powers Provision. And as I explain below, our pre-1983 precedent does not appear to support a Chevron-style regime. So, in an appropriate case, I think we should reconsider the matter. 1. Our recent discovery of Chevron-style deference in our precedent was ill-founded. In Chevron, the United States Supreme Court held that courts should defer to a federal agency’s reasonable interpretation of ambiguous federal statutes, reasoning that such ambiguity represents an implicit delegation from Congress for the agency to decide that question. See 467 U.S. at 843-44, 865-67. Under Chevron, courts must first ask “whether Congress has directly spoken to the precise question at issue.” Id. at 842-43. “[I]f the statute is silent or ambiguous with respect to the specific issue, the question for the court is whether the agency’s answer is based on a permissible construction of the statute.” Id. at 843. Eight years ago, we asserted for the first time that our deference precedent was properly understood as “in accord with that identified by the United States Supreme Court in Chevron.” See Cook et al. v. Glover, 295 Ga. 495, 500 (761 SE2d 267) (2014). We were wrong. Nothing we cited in Cook supported that view. And Georgia deference precedent historically had different rationales from Chevron, was less binding, and principally applied to longstanding interpretations (again, unlike Chevron). a. None of the cases cited in Cook support its claim about our deference precedent. Cook‘s novel assertion about Chevron and our own precedent came in a case involving the application by a state agency of a legal interpretation of federal law by a federal agency (to which Chevron would apply anyway as a matter of federal law), and came over the disagreement of two justices. 295 Ga. at 497-98, 502-03. But on closer examination, none of the cases Cook cited for the idea that Georgia law resembles Chevron actually support it — and certainly not in any way that reveals a consistent and definitive construction of our Separation of Powers Provision.[2] See id. at 499-501 (citing Handel v. Powell, 284 Ga. 550, 553 (670 SE2d 62) (2008); Schrenko v. DeKalb County School District. See 276 Ga. 786, 791 (2) (582 SE2d 109) (2003); Center for a Sustainable Coast v. Coastal Marshlands Protection Committee, 284 Ga. 736, 741 (2) (761 SE2d 267) (2014); Ga. Real Estate Comm’n. v. Accelerated Courses in Real Estate, 234 Ga. 30, 32 (2) (214 SE2d 495) (1975); Ga. Dept. of Community Health v. Medders, 292 Ga. App. 439, 440 (664 SE2d 832) (2008)). In Handel, the Secretary of State argued that the courts were required to defer to the Secretary’s interpretation of the statute governing the residency requirement for a PSC candidate. 284 Ga. at 553. We rejected that argument, pointing out that “[an] agency’s interpretation is not binding on the courts, which have the ultimate authority to construe statutes.” Id. Although we did say that “judicial deference is [generally] afforded an agency’s interpretation of statutes it is charged with enforcing or administering,” id., we made clear that the role of the judicial branch is to make “an independent determination as to whether the interpretation of the administrative agency correctly reflects the plain language of the statute[.]” Id. (quoting Sawnee Elec. Membership Corp. v. Ga. Pub. Svc. Comm., 273 Ga. 702, 706 (544 SE2d 158) (2001)). Courts independently determining whether an agency’s interpretation is correct is not Chevron-style deference. In the next case, Schrenko, we cited our 1939 decision State v. Camp, 189 Ga. 209, 216-17 (6 SE2d 299) (1939), and a Court of Appeals case following the Camp line, for the proposition that “[w]here statutory provisions are ambiguous, courts should give great weight to the interpretation adopted by the administrative agency charged with enforcing the statute.” Schrenko, 276 Ga. at 791 (citing). But Camp itself did not defer to the State Revenue Commissioner’s interpretation in that case — it applied (among other interpretive principles) the longstanding canon of construction that “ambiguous tax statutes . . . should be construed most strongly in favor of the citizen.” 189 Ga. at 216-17. And in the very next sentence, our opinion in Schrenko cited Sawnee for the proposition that “this Court is ‘not bound to blindly follow’ an agency’s interpretation,” and we defer to an agency’s interpretation only when it reflects the meaning of the statute. Schrenko, 276 Ga. at 791 (2). Indeed, Sawnee went further and explained that “[a]dministrative rulings are not binding on this Court, and will only be adopted when they conform to the meaning which the appellate court deems should properly be given[.]” 273 Ga. at 706 (emphasis added). Again, this does not sound anything like Chevron. The next case Cook relied on, Sustainable Coast, relied on Schrenko and Georgia Department of Revenue v. Owens Corning, 283 Ga. 489, 490 (660 SE2d 719) (2008). See Sustainable Coast, 284 Ga. at 741. Schrenko, as I just explained, cannot support a state version of Chevron. Owens Corning, in turn, relied on Kelly v. Lloyd’s of London, 255 Ga. 291, 293 (336 SE2d 772) (1985), which relied on a Court of Appeals case grounded in the unquestioning application of federal deference cases. See id. This sort of unreasoned precedent tells us little about how properly to think about judicial deference to executive branch legal interpretation, much less anything about the original meaning of the Georgia Constitution’s Separation of Powers Provision and its implications for this question (especially when that precedent is from the Court of Appeals, which lacks jurisdiction to decide unsettled questions of constitutional law). The only pre-1983 case Cook cited on the subject was Georgia Real Estate Commission, which purported to find a generally- applicable two-part test for the validity of agency action: “(1) Is it authorized by statute, and (2) is it reasonable?” 234 Ga. at 32 (2) (citation omitted). But the case it cited for that proposition, Eason v. Morrison, did not announce a general requirement of deference to reasonable agency interpretations; instead, it interpreted a statute that gave the agency “power to adopt all reasonable rules and regulations” required to implement the statute. 181 Ga. 322, 322 (182 SE 163) (1935). We held there that the statute’s textual requirement of reasonableness meant that an otherwise statutorily- authorized rule must also be reasonable in order to pass muster. Id. at 323. That holding — based on the specific statutory text at issue in that case — was necessarily limited to statutory contexts with similar textual requirements. So if later cases like Georgia Real Estate Commission took that to establish a broader rule of deference to any agency rules that were reasonable, then they grossly misread Eason. See 234 Ga. at 30. In any event, Georgia Real Estate Commission itself did not invoke reasonableness as a sufficient threshold for deference — it purported to say that regulations must be substantively reasonable as well as authorized by statute. See id. at 32 (1) (“An agency rule might be reasonable but unauthorized by statute, or authorized by statute but unreasonable. In either event, it could not stand.”). Even if that conclusion had been correct as applied to statutory schemes without a textual reasonableness requirement, it still was wholly inconsistent with a Chevron-style deference regime, in which reasonableness replaces a de novo judicial determination of whether a rule is authorized by statute. Finally, Cook cited a Court of Appeals case, Medders, which (through a series of earlier Court of Appeals decisions) finds its roots in three of our cases. See 292 Ga. App. at 440. The first is Camp, which, as discussed above, did not defer at all. The second is Solomon v. Commissioners of Cartersville, 41 Ga. 157, 161 (1870), which appeared to defer to the longstanding practice of the executive branch as a prudential reason not to rule otherwise. See id. (“If this was an original question, independent of any construction heretofore given by the Executive Department . . . we should be inclined to hold, that the Governor could not approve and sign any bill after the adjournment of the General Assembly; but on looking into the past history of our legislation, we find that it has been the practice for many years . . . and that a large number of the most important Acts now upon the statute books of the State have been so approved and signed, which usage and practice of the Executive Department of the State Government, should not now, in our judgment, be disturbed or set aside.”).[3] And the third is Wilson v. Pollard, which essentially confirmed a prior interpretation of a statute based on a theory of legislative acquiescence. 190 Ga. 74, 79 (4) (8 SE2d 380) (1940) (“the legislature has met many times during the half century that has passed since the [prior] decision was rendered . . . . During this long period the legislature has at no time indicated dissatisfaction or disagreement with the construction placed upon that act . . . . By such silence the conclusion is inevitable that the legislature is in agreement with this court’s interpretation of that act.”). Simply put, Cook did not cite anything that actually justified its statement that our deference precedents align with Chevron — and certainly nothing of an age or stature to show a consistent and definitive construction of our Separation of Powers Provision. b. Our decision in Tibbles similarly failed to show that we have applied this Chevron-style approach for a long time. And yet, the year after Cook, we doubled down and said that this Chevron-style approach “is not a new one” in Georgia. Tibbles v. Teachers Retirement System of Ga., 297 Ga. 557, 559 (1) & n.1 (775 SE2d 527) (2015) (citing David E. Shipley, “The Chevron Two-Step in Georgia’s Administrative Law,” 46 Ga. L. Rev. 871, 888-916 (III) (2012)). We relied on largely the same cases for that notion. See id. (citing, among other things, Cook, 295 Ga. at 500 and Sustainable Coast, 284 Ga. at 741).[4] We also elaborated that “the General Assembly properly may leave some matters to the discretion of the Executive Branch,” Tibbles, 297 at 559 (1) (citing Dept. of Transp. v. City of Atlanta, 260 Ga. 699, 703 (1) (398 SE2d 567) (1990) (“DOT’)), and suggested “that some ambiguities may be better resolved by officers and agencies of the Executive Branch,” id. (citing Bentley v. Chastain, 242 Ga. 348, 350-51 (1) (249 SE2d 38) (1978)). But those are just potential rationales for deference — they are not indications of a consistent and definitive Chevron-style approach to deference. DOT decided that the exercise of the eminent domain power by an executive agency was not an unconstitutional delegation of legislative power. 260 Ga. at 703-04 (1). It said nothing about whether deference to executive branch legal interpretations posed a separation of powers problem. See id.[5] And while Bentley discussed rationalizations for deference to agency decisions, it was not itself a deference case at all. 242 Ga. at 350-51 (1). The issue there was whether the General Assembly could expand judicial review of agency action beyond “[asking] whether the agency acted beyond the discretionary powers conferred upon it” — i.e., whether the agency acted within the bounds of its statutory authority — into “a de novo jury trial” on the appropriateness of the agency’s decisions. Id. at 352 (1). We held (rightly or wrongly) that it could not.[6] Id. In any event, nothing in Bentley established or revealed any general rule of deferring to an agency’s reasonable interpretation of its statute.[7] In short, our recent discovery of longstanding Chevron-type deference within our precedents appears wrong. The few pre-1983 cases cited in support of deference don’t offer sufficient support for Cook or Tibbles, and only four of our post-1983 cases in this line even held that deference was appropriate. This is little basis for a state version of Chevron, let alone one consistent with the original public meaning of our Separation of Powers Provision. And as we’ll see in the next division, our pre-1983 case law — the cases most relevant to the original public meaning of the 1983 Constitution — simply cannot be read as consistent with Chevron. 2. Viewed as a whole, our historic deference precedents diverge from each other and the federal regime many times over. Looking beyond the few cases we’ve cited in support of a state version of Chevron, we see that our case law is far more diverse than we have acknowledged. Our cases diverge from each other — and from the federal regime — in numerous and important ways. a. The justifications for deference vary. First among the many points on which our precedents diverge is the justification for deference to executive branch legal interpretation. Most recently, as the United States Supreme Court has done in Chevron and its progeny, we have focused on the notion that the General Assembly purports to delegate to the executive branch some authority to resolve ambiguity in the statutes it administers. See, e.g., Tibbles, 297 Ga. at 563-64 (2) (b) (relying on United States v. Mead Corp, 533 U.S. 218, 226-27 (121 SCt 2164, 150 LE2d 292) (2001)); Cook, 295 Ga. at 500. But I see little evidence of this justification in our earlier cases. At least one of our earliest deference cases, for example, justified deference to a long-standing legal interpretation by county officials as a particularly strong form of legislative acquiescence. In Rice v. Johnson, we noted that county tax collectors had for decades acted under a particular understanding of tax law. 20 Ga. 639, 644 (1856). We held that this long-standing practice was not merely evidence of the county officials’ understanding, but also of the General Assembly’s approval of it: “for if the Legislature had not thought so, they would have passed a law to authorize a change of the practice.” Id. We thus concluded that the General Assembly intended the law to be as the county officials understood it, and we applied that construction. See id. We then cited Rice two decades later as authority for the proposition that “the usage of the executive department may be invoked properly and legally to throw light upon words used in the statutes of the state.” Miller v. Wilson, 60 Ga. 505, 507-08 (1879). And that principle re-appeared in later deference cases. See Camp, 189 Ga. at 217 (adding, in support of the rule that tax statutes be construed in favor of the taxpayer, that the trial court’s ruling was correct because it was “in accord with the [prior] interpretation which ha[d] been given by the State administrative authorities for a number of years, during which time there ha[d] been several sessions of the General Assembly without any disturbance of such administrative interpretation”); Schrenko, 276 Ga. at 791 (2) (relying on Camp for the proposition that courts should give great weight to the interpretation adopted by the administrative agency charged with enforcing the statute). Since that time, though, we have become critical of applying the principle of legislative acquiescence to our own decisions, let alone those of executive branch agencies: Legislative silence is a poor beacon to follow in discerning the proper statutory route. Legislative inaction frequently betokens unawareness, preoccupation, or paralysis. It is at best treacherous to find in legislative silence alone the adoption of a controlling rule of law. Where, as in the case before us, there is no indication that a subsequent General Assembly has addressed itself to the particular problem, we are unpersuaded that silence is tantamount to acquiescence, let alone approval[.] State v. Jackson, 287 Ga. 646, 659 (5) n.8 (697 SE2d 757) (2010) (cleaned up) (quoting Zuber v. Allen, 396 U.S. 168, 185 & n.21 (90 SCt 314, 24 LE2d 345) (1969) (citations and punctuation omitted)). b. The force or strength of deference is inconsistent. A second point on which our precedents have been inconsistent is the force with which deference applies. Recently, mimicking federal law, we have held that courts are bound to follow an executive branch legal interpretation when the principles of Chevron would so require. See Cook, 295 Ga. at 501 (reversing Court of Appeals for substituting its own plausible construction for the reasonable construction of the relevant agency). But although some language in our older opinions might be read as supporting such a result, our actual historical practice does not. A substantial portion of our deference precedent has held that executive branch legal interpretations are persuasive authority, entitled only to “consideration” (and “weight,” if weight and consideration are different things). See, e.g., Howell v. State, 71 Ga. 224, 229 (1883) (“The practice of the various departments of the government, as a means of collateral interpretation, is not to be rejected by the courts, in passing upon the constitutionality of a law. It is entitled to consideration and weight, especially in view of another settled rule, that a law is not to be set aside unless its conflict with the provisions of the constitution is plain and obvious.”); Miller v. Wilson, 60 Ga. 505, 507-08 (1878) (“[T]he usage of the executive department may be invoked properly and legally to throw light upon words used in the statutes of the state.”); Blount v Monroe, 60 Ga. 61, 66 (1878) (“[T]he meaning of a tax act may be gathered from the custom of the executive department.”); McClendon v. Frost & Crenshaw, 59 Ga. 350, 351 (1877) (“The uniform course of practice is evidence of the law.”). In fact, we’ve even framed the persuasiveness of such authority in these terms. See Elder v. Home Bldg. & Loan Assn., 188 Ga. 113, 116 (2) (3 SE2d 75) (1939) (“We recognize the rule entitling such collateral interpretation to consideration by the court in passing on the constitutionality of a law; and where the invalidity of a statute is doubtful, it has much weight with the court in determining its validity; but, after all, the responsibility for determining the constitutionality of an act is . . . imposed upon the courts.”). At least on its own terms, this strikes me as much more like the “power to persuade” deference the federal system affords under Skidmore than the controlling force Chevron requires. See Skidmore v. Swift & Co., 323 U.S. 134 (65 SCt 161, 89 LE 124) (1944). c. Some of our cases suggest we defer only to longstanding interpretations. There’s at least one more point on which the inconsistency of our precedent undermines the conclusion that we have long applied a version of Chevron. Namely, Chevron defers to a particular legal interpretation contained in a particular agency rule or policy. See 467 U.S. at 842-45. And that’s (sort of) how we’ve recently applied the Georgia version. See Tibbles, 297 Ga. at 563-65 (2) (b) (deferring to interpretation in agency rules); Cook, 295 Ga. at 500-01 (deferring to interpretation in policy manual).[8] But much of our historic precedent focused instead on whether there was a long-standing practice, not merely a single, one-time interpretation. See, e.g., Standard Oil Co. of Ky. v. State Revenue Commission, 179 Ga. 371, 376 (176 SE 1) (1934) (“Long-continued practice and the approval of administrative authorities may be persuasive in the interpretation of doubtful provisions of a statute, but [they] cannot alter provisions that are clear and explicit when related to the facts disclosed.”); Southern Ry. Co. v. Melton, 133 Ga. 277, 283 (3) (65 SE 665) (1909) (“It is, however, legitimate, as new problems arise, to draw light from contemporaneous construction, or long-continued practice of the departments of government, in reference to matters somewhat analogous to the creation of such commissions and the conferring of powers upon them.”); Temple Baptist Church v. Ga. Terminal Co., 128 Ga. 669, 680 (58 SE 157) (1907) (“The long-continued practice of the executive or the legislative department will be treated as persuasive authority by the courts, and has, in numerous cases, been followed, although the individuals composing the court at the time would have doubt as to the true construction if the question were left unaffected by the construction placed upon it by another department of government.”); Epping v. City of Columbus et al., 117 Ga. 263, 273 (43 SE 803) (1903) (“the long-continued practice of co[-]ordinate departments of the government may well be resorted to by the judicial department as an aid in determining what is the true meaning of words contained in the instrument”), overruled on other grounds by Harrell et al. v. Town of Whigham, 141 Ga. 322 (80 SE 1010) (1914). * * So even within our deference precedent — regardless of the strength of its foundation — there are numerous, important rifts between our own cases and between our cases and Chevron deference. 3. The law of the case prevents us from revisiting our deference precedents here. In the end, though, this is not the case to explore these issues. Here, relying on our binding precedent in Tibbles (and other cases approving judicial deference to administrative interpretations), the Court of Appeals held that the trial court did not abuse its discretion in deferring to the Public Service Commission’s view that the terms “usage revenue” and “total revenue” are synonymous. Cazier v. Ga. Power Co., 362 Ga. App. 112, 118 (1) (a) (866 SE2d 827) (2021). For the reasons just discussed, I have grown increasingly skeptical that our precedent — which the Court of Appeals relied upon below[9] — was rightly decided. But in the previous appeal in this case, we held (citing only federal precedent) that if the words requiring interpretation were “used in a . . . technical sense,” necessitating extrinsic evidence “to determine their meaning,” then the “inquiry is essentially one of fact and discretion in technical matters[.]“ Ga. Power Co. v. Cazier, 303 Ga. 820, 826 (3) (815 SE2d 922) (2018) (quoting United States v. Western Pacific R. Co., 352 U.S. 59, 65-66 (77 SCt 161, 1 LE2d 126) (1956)). Although I joined that holding four years ago, I am not nearly as confident today that it was correct. But the law of the case doctrine precludes our reconsideration of our holding. OCGA § 9-1160 (h) (“any ruling by the Supreme Court or the Court of Appeals in a case shall be binding in all subsequent proceedings in that case in the lower court and in the Supreme Court or the Court of Appeals as the case may be”). And absent that reconsideration, we have little ability to reach what strikes me as the legitimate points in Cazier’s cert petition. Accordingly, I concur in the denial of that petition.
