LEMIEUX v. EVERS (2025)

Original Action

¶1 In this original action we must determine whether Governor Tony Evers exceeded his partial veto authority under Article V, Section 10(1) of the Wisconsin Constitution. At issue is Wisconsin's 2023–25 biennial budget bill that included an education revenue limit increase for two fiscal years. Using his partial veto authority, the governor expanded the provision from two fiscal years to 402 fiscal years by striking words and digits from the bill. We conclude that those 2023 partial vetoes do not violate the constitution.

¶2 In challenging the 2023 partial vetoes, petitioners do not ask us to overrule our precedent. Petitioners agree that the partial vetoes at issue satisfy the principles we have applied in our previous cases. Instead, petitioners bring two novel challenges. First, they contend that the 2023 partial vetoes violate § 10(1)(b) because the governor did not veto the bill “in part” when he extended a duration of time, as 402 years is not part of two years. Second, petitioners maintain that the 2023 partial vetoes violate § 10(1)(c) because that provision prohibits the governor from striking digits to create new numbers.

¶3 We reject both arguments. The first argument fails because it improperly relies on our holding in Citizens Utility Board v. Klauser, 194 Wis. 2d 484, 534 N.W.2d 608 (1995) (C.U.B.), which was limited to the specific circumstance of write-in vetoes, which is absent here. The second argument also fails because § 10(1)(c) plainly does not prohibit the governor from striking digits to create new numbers. Consequently, we deny petitioners’ requested relief. But in doing so, we set forth multiple options available to the legislature—one of which specifically addresses the 400-year modification at issue here.

I. BACKGROUND

¶4 The Wisconsin Constitution provides that every two years the legislature is to pass a biennial budget. This budget establishes the level of revenue to be derived from taxes and other sources, as well as authorized expenditures. See Wis. Const. art. VIII, §§ 2, 5. The process begins with the governor presenting the legislature with an executive budget bill. See Wis. Stat. §§ 16.45–16.47 (2023–24).¹ The executive budget bill then proceeds through the legislature's multi-step review and report process involving the joint committee on finance and legislative fiscal bureau. The legislature then submits its bill to the governor. See, e.g., Wis. Stat. §§ 13.093(1), 13.95, 13.102. Before signing the bill into law, the governor may partially veto parts of the bill. Wis. Const. art. V, § 10(1). Subsequently, the legislature may vote to override the governor's partial vetoes by a supermajority. Id., § 10(2).

¶5 This process was followed for the 2023–25 biennial budget. First, the governor presented his 2023–25 executive biennial budget bill, which included three educational revenue limit increases: a $350 per pupil revenue limit increase for 2023–24, a $650 per pupil revenue limit increase for 2024–25, and a subsequent per pupil revenue limit adjustment indexed to inflation.

¶6 Next the legislature reviewed the governor's proposed budget bill and made modifications. Senate Bill 70 provided for a $325 per pupil revenue limit increase for both 2023–24 and 2024–25, without a subsequent inflationary index.

¶7 Then the governor exercised his partial veto power, deleting portions of 2023 Senate Bill 70. As related to this matter, the governor deleted entire words and some numbers from Sections 402, 403, 404, and 408 of Senate Bill 70. The result, published as 2023 Wisconsin Act 19, authorized a $325 per pupil revenue limit increase from 2023–2425, extending the provision by 400 additional years. This is the text of the vetoed sections, with the deleted text struck through:

Section 402. 121.905(3)(c) 9. of the statutes is created to read:

121.905(3)(c) 9. For the limit for the 2023–24 school year and the 2024–25 school year, add $325 to the result under par. (b).

Section 403. 121.91(2m)(j)(intro.) of the statutes is amended to read:

121.91(2m)(j)(intro). Notwithstanding par. (i) and except as provided in subs. (3), (4), and (8), a school district cannot increase its revenues for the 2020–21 school year, the 2023–24 school year, and the 2024–25 school year to an amount that exceeds the amount calculated as follows:

Section 404. 121.91(2m)(j) 2m. of the statutes is created to read:

121.91(2m)(j) 2m. In the 2023–24 school year and the 2024–25 school year, add $146.

․

Section 408. 121.91(2m) (t) 1. (intro.) of the statutes is amended to read:

121.91(2m)(t) 1. (intro.) If 2 or more school districts are consolidated under s. 117.08 or 117.09, in the 2019–20 school year, the consolidated school district's revenue limit shall be determined as provided under par. (im), in the 2020–21 school year, 2023–24 school year, or 2024–25 school year, the consolidated school district's revenue limit shall be determined as provided under par. (j), and in each school year thereafter, the consolidated school district's revenue limit shall be determined as provided under par. (i), except as follows:

2023 Wisconsin Act 19, §§ 402–04, 408.

¶8 The senate subsequently voted to override the partial vetoes, but the assembly declined to vote on the override. Consequently, the effort to override the governor's vetoes failed. The law went into effect and this original action followed.

II. ANALYSIS

¶9 We interpret a constitutional provision by “focus[ing] on the constitutional text, reading it reasonably, in context, and with a view of the provision's place within the constitutional structure.” Wisconsin Just. Initiative, Inc. v. WEC, 2023 WI 38, ¶21, 407 Wis. 2d 87, 990 N.W.2d 122; see also SEIU v. Vos, 2020 WI 67, ¶28, 393 Wis. 2d 38, 946 N.W.2d 35 (“The text of the constitution reflects the policy choices of the people, and therefore constitutional interpretation similarly focuses primarily on the language of the constitution.”).

¶10 We begin our analysis with the relevant text of § 10(1)(b) and (c) and an outline of the principles this court has applied when interpreting these constitutional provisions. Then we explain why the 2023 partial vetoes satisfy both provisions. We conclude by highlighting potential avenues available to the legislature, should it decide to alter the governor's partial veto power.

A. Partial Veto Principles

¶11 Article V, Section 10(1) of the Wisconsin Constitution sets forth the governor's partial veto power. It provides in pertinent part:

(1)(a) Every bill which shall have passed the legislature shall, before it becomes a law, be presented to the governor.

(b) If the governor approves and signs the bill, the bill shall become law. Appropriation bills may be approved in whole or in part by the governor, and the part approved shall become law.

(c) In approving an appropriation bill in part, the governor may not create a new word by rejecting individual letters in the words of the enrolled bill, and may not create a new sentence by combining parts of 2 or more sentences of the enrolled bill.2

¶12 Over the past 90 years, our precedent has established four principles that we have applied to “deletion vetoes,” the traditional partial veto in which the governor strikes text:

Deletion veto principles:3

1. The governor's deletion vetoes are constitutional as long as the remaining text of the bill constitutes a “complete, entire, and workable law.” State ex rel. Wis. Tel. Co. v. Henry, 218 Wis. 302, 314–15, 260 N.W. 486 (1935); see also State ex rel. Martin v. Zimmerman, 233 Wis. 442, 450, 289 N.W. 662 (1940).

2. The governor may exercise deletion vetoes only on parts of bills containing appropriations within their four corners. State ex rel. Finnegan v. Dammann, 220 Wis. 143, 147–48, 264 N.W. 622 (1936).

3. The governor's deletion vetoes may not result in a law that is “totally new, unrelated or non-germane” to the original bill. State ex rel. Wis. Senate v. Thompson, 144 Wis. 2d 429, 451–53, 424 N.W.2d 385 (1988).

4. The governor may strike “words, letters, or numbers.” Id. at 434. But “the governor may not create a new word by rejecting individual letters in the words of the enrolled bill.” Wis. Const. art. V, § 10(1)(c). Nor may the governor “create a new sentence by combining parts of 2 or more sentences of the enrolled bill.” Id.

¶13 Separate from deletion vetoes, there is one scenario in which the governor may exercise “write-in” vetoes by striking certain text and then writing in different text:

Write-in veto principle:

The governor may strike an appropriation amount and write in a smaller appropriation amount. See C.U.B., 194 Wis. 2d at 505–06 (holding the governor may reduce an appropriation of $350,000 to $250,000 because the latter was a “part” of the former under § 10(1)(b)); see also Risser v. Klauser, 207 Wis. 2d 176, 181, 558 N.W.2d 108 (1997) (enforcing the narrow application of the write-in holding in C.U.B. to appropriation amounts).

B. 10(1)(b) Analysis

¶14 Again, § 10(1)(b) provides in pertinent part that “[a]ppropriation bills may be approved in whole or in part by the governor.” To address petitioners’ challenge under this provision, we begin by explaining how the partial vetoes here satisfy the four deletion veto principles, none of which the parties ask us to disturb.⁴ Then, we explain why the write-in veto principle adopted in C.U.B. is not relevant to our analysis.

¶15 As for the first principle, this court established the “complete, entire, and workable law” principle in 1935, five years after § 10(1)(b)’s enactment, in Henry, 218 Wis. at 314–15; see also Zimmerman, 233 Wis. at 450. This must be an objective inquiry. Wis. Senate, 144 Wis. 2d at 453. In other words, we look only at the remaining text, not whether the partial vetoes substantively changed the policy set forth in the enrolled bill. See State ex rel. Kleczka v. Conta, 82 Wis. 2d 679, 715, 264 N.W.2d 539 (1978). That is because our constitution vests the governor with “broad and expansive,” “quasi-legislative” partial veto power. Wis. Senate, 144 Wis. 2d at 442, 453–54. Thus, we accept that the constitution “anticipate[s] that the governor's action may alter the policy as written in the bill.” State ex rel. Sundby v. Adamany, 71 Wis. 2d 118, 134, 237 N.W.2d 910 (1976).

¶16 The 2023 partial vetoes comply with the first principle. When looking only at the remaining text, it is clear that a bill that increases the $325 per pupil revenue limit until 2425 is complete and workable.

¶17 These vetoes also satisfy the second principle. All parties agree that the 2023 partial vetoes were part of the biennial budget containing appropriations. See, e.g., Finnegan, 220 Wis. at 147–49 (establishing the governor's veto power extended to all parts of an appropriation bill, not just provisions expressly dealing with appropriations).

¶18 The 2023 partial vetoes meet the third deletion veto principle as well. It is undisputed that the law resulting from these partial vetoes is germane to the enrolled bill because both versions address educational funding. Only a change in the duration of that funding is at issue.

¶19 Last, the 2023 partial vetoes are valid under the fourth principle. As we explain further below, in part II.C., these partial vetoes, which struck only words and numbers, satisfy the requirements of § 10(1)(c). There are no instances of the governor striking letters to make new words, or combining portions of sentences to create new sentences.

¶20 Having addressed all four deletion veto principles, we turn to petitioners’ request to apply the C.U.B. write-in veto principle here. Petitioners ask that we invalidate the 2023 partial vetoes because under C.U.B., the 402-duration created by these partial vetoes is not “less than” and thus not “part” of the legislatively-approved two-year duration. Even though 402 years are clearly more than two, C.U.B. does not apply here.

¶21 In C.U.B. we evaluated the unprecedented scenario in which the governor decreased an appropriation amount from $350,000 to $250,000 by deleting “350,000” and writing in “250,000.” 194 Wis. 2d at 488. We determined that this write-in partial veto was constitutional under the very narrow facts presented in that case. Petitioners correctly note that to reach that holding, we applied the definition of “part” referenced—but not applied—in Henry: “ ‘something less than a whole; a number, quantity, mass, or the like, regarded as going to make up, with others or another, a larger number, quantity, mass, etc.’ ” Id. at 505 (quoting Henry, 218 Wis. at 313 (quoting Webster's New International Dictionary at 1781 (2d ed.))). We also drew on Wisconsin Senate’s express recognition that the governor has the authority to reduce appropriations. Id. at 506. Putting those two principles together, we concluded that because the write-in veto was only to an appropriation amount, and $250,000 is less than $350,000, $250,000 was part of $350,000 for purposes of § 10(1)(b).

¶22 We reject petitioners’ request that we apply that reasoning here because both the facts of C.U.B. and the analytical principles underpinning its narrow holding are absent. Of import, there is no write-in element to the 2023 partial vetoes; they are deletion vetoes. So, on its face, C.U.B. does not apply. Aside from this threshold distinction, any effort to incorporate “part” as applied in C.U.B. would force us to overrule our express holdings in C.U.B. and Risser. Critically, and fatal to petitioners’ contentions, this court expressly limited C.U.B.’s holding to modifications of appropriation amounts. Id. at 510 (the write-in veto power “stems from the right to reduce appropriations recognized in Wisconsin Senate and extends only to monetary figures and is not applicable in the context of any other aspect of an appropriation”). We cemented that limit by specifically rejecting the notion that the “less-than” meaning of “part” could apply to other concepts, including dates and durations. Id. at 511 n.18. And in Risser we reinforced C.U.B.’s limited reach to only appropriation amounts. 207 Wis. 2d at 188 (the C.U.B. ruling “expressly dr[ew] a distinction between appropriation amounts and other parts of appropriation bills”).

¶23 Here, we are tasked with evaluating a change in years, not appropriation amounts, which plainly falls outside C.U.B.’s holding and analytical principles. Petitioners fail to reckon with C.U.B.’s explicit boundary and do not attempt to equate appropriations with durations. As significantly, petitioners do not ask that we overrule or revisit our precedent. Therefore, we do not extend the write-in veto principle to the 2023 partial vetoes.

¶24 In sum, the four deletion-veto principles apply to the 2023 partial vetoes, and the write-in veto principle does not apply. Because these partial vetoes satisfy all four deletion-veto principles, they are valid under § 10(1)(b).

C. 10(1)(c) Analysis

¶25 We next consider petitioners’ contention that the governor impermissibly deleted digits to create new numbers. Central to this challenge is the proper interpretation of the first clause of § 10(1)(c): “[T]he governor may not create a new word by rejecting individual letters in the words of the enrolled bill.” We conclude that this provision relates exclusively to the deletion of letters to create new words, not the deletion of digits to create new numbers.

¶26 The plain meaning of “word” does not include numbers written out using digits, and the plain meaning of “letters” does not include digits. By way of example, all agree with petitioners that the number “ten” is a word written with letters. However, when we write the number “10” using digits, we have used no letters. Simply put, letters and digits are not interchangeable for purposes of § 10(1)(c). This has not demonstrably changed since 1990 when this provision passed.

¶27 Moreover, this court has explicitly treated letter and digit vetoes separately, both before and after § 10(1)(c)’s adoption. In Wisconsin Senate, this court discussed the governor's power to strike “phrases, digits, letters, and word fragments.” 144 Wis. 2d at 433; see also id. at 437, 457, 462. This straightforward language establishes that in 1988, this court viewed words, letters, and digits as distinct types of text that the governor may strike. This reading is consistent with our jurisprudence discussing § 10(1)(c). Risser, 208 Wis. 2d at 183 (the “governor may strike words or digits from an appropriation bill”); C.U.B., 194 Wis. 2d at 501 (noting that the partial veto power includes the power to strike letters and that the “power to veto letters to create new words” was subsequently limited by § 10(1)(c)). Petitioners nonetheless propose an alternative interpretation of § 10(1)(c). They argue that because dictionary definitions of the terms “word” and “letter” may incorporate the concepts of numbers and digits, then § 10(1)(c) incorporates those same concepts as well to prohibit the governor from striking digits to create a new number. We reject that strained interpretation. Section 10(1)(c) did not include the terms “digit” or “number”; it invoked just “word” and “letter.” We must give meaning to those omissions. The only logical interpretation here is that the people of Wisconsin were prohibiting the deletion of letters to create new words. In short, the plain language of the constitutional text permits striking numbers written out with digits. Accordingly, we conclude that the 2023 partial vetoes did not violate § 10(1)(c).

D. Legislative Options

¶28 We uphold the 2023 partial vetoes, and in doing so we are acutely aware that a 400-year modification is both significant and attention-grabbing. However, our constitution does not limit the governor's partial veto power based on how much or how little the partial vetoes change policy, even when that change is considerable. As our precedent recognizes, the governor's constitutionally-vested, quasi-legislative role defeats “any separation of powers-type argument that the governor cannot affirmatively legislate by the use of the partial veto power.” Wis. Senate, 144 Wis. 2d at 453. Indeed, the governor's reliance on his partial veto authority to potentially increase taxes without legislative approval is neither new nor unique in our partial veto jurisprudence. See Kleczka, 82 Wis. 2d at 715 (holding that gubernatorial vetoes that created the possibility of increased expenditures from the state general fund were permissible because the legislature could have passed the same law).

¶29 The bottom line is that the partial vetoes were within the bounds of the constitution. But the legislature is not without recourse. It has multiple options at its disposal, including:

Future budget bills: Unlike an appropriation amount typically spent during the biennium in which the funds were appropriated, the 2023 partial vetoes affect revenue limits 400 years into the future. Accordingly, the legislature may address those partial vetoes during the 2025–27 biennial budget process, or in a subsequent biennial budget.

Constitutional amendment: The legislature has the power to introduce a constitutional amendment. In the past 35 years, the people of Wisconsin have twice amended the constitution to limit the governor's partial veto power. A constitutional amendment to address the 2023 partial vetoes is currently under advisement with the legislature. 2023 Enrolled Joint Resolution 16 would amend the constitution to prohibit the governor from using the partial veto to create or increase any tax or fee.5 If the legislature adopts that joint resolution without change, it will be submitted to the voters. If the voters ratify it, the constitution will be amended.

At present, legislators are circulating a proposed joint resolution for a constitutional amendment that would change the governor's partial veto power to permit him or her to only veto entire sections of the proposed bill or to reduce appropriation amounts.6 If the proposed joint resolution is adopted in 2025, it will go through the same process for submission to the voters. Such a constitutional amendment would substantially supersede this court's partial veto precedent.

Legislative drafting: Legislators may draft bills separate from appropriation bills to avoid the governor's partial veto. And, legislators may anticipate the governor's use of her or his power when crafting appropriation bills.

¶30 The court takes no position regarding these measures. We merely outline them to illustrate legislative alternatives to the action before us.

III. CONCLUSION

¶31 We conclude that Sections 402, 403, 404, and 408 of 2023 Wisconsin Act 19 were vetoed consistent with Article V Sections 10(1)(b) and (c) of the Wisconsin Constitution.

By the Court.—Relief denied.

¶32 I agree with the majority/lead opinion's conclusion that the partial vetoes at issue in this case do not violate Article V, §§ 10(1)(b) and (c) of the Wisconsin Constitution. I write separately, however, because I have a different understanding of Petitioners’ argument that those partial vetoes are unconstitutional under § 10(1)(b) and why that argument should be rejected. Accordingly, I join only ¶¶1–19 and 25–31 of the majority opinion.

¶33 Section 10(1)(b) authorizes the governor to approve appropriation bills “in whole or in part ․” Wis. Const. Art. V., § 10(1)(b). Petitioners argue that the partial vetoes at issue here exceeded the governor's authority under § 10(1)(b) because he did not approve “part” of the original bill. They cite to State ex rel. Wisconsin Telephone Co. v. Henry and Citizens Utility Board. v. Klauser (C.U.B.) for the assertion that the ordinary meaning of “part,” at least when applied to numbers, is “something less than a whole.” See C.U.B., 194 Wis. 2d 484, 505, 534 N.W.2d 608 (1995); Henry, 218 Wis. 302, 313, 260 N.W. 486 (1935). They claim that applying that definition in this case requires us to determine whether, as a matter of “substance rather than form,” the governor's partial vetoes approved “something less than [the] whole” of what the legislature passed. See C.U.B., 194 Wis. 2d at 497. And because the substantive effect of those vetoes was to increase the two-year duration the legislature passed to a 402-year duration it never contemplated, the governor's partial vetoes did not approve something less than the whole of what the legislature passed.

¶34 Petitioners’ argument has some support in the reasoning of C.U.B. Indeed, one reason we cited for upholding the veto at issue in that case—crossing out a $350,000 appropriation and writing in $250,000—was that the result of the veto was substantively “part” of what the legislature originally passed. See C.U.B., 194 Wis. 2d at 489, 505–06. As we said in C.U.B., $250,000 is “part” of $350,000 because it is “something less than” $350,000. Id. at 505–06. C.U.B’s use of this reasoning, Petitioners contend, demonstrates that there is a threshold requirement, imposed on all partial vetoes by § 10(1)(b), that the result of the veto must be substantively “part” of the original bill.

¶35 Petitioners’ substantive-part analysis should be rejected, however, because it cannot be squared with the rest of our cases interpreting § 10(1)(b), none of which Petitioners ask us to overturn. We have long held that the only test under § 10(1)(b) for whether a veto approved “part” of a bill is simply whether the veto results in a complete and workable law. See State ex rel. Kleczka v. Conta, 82 Wis. 2d 679, 704–08, 264 N.W.2d 539 (1978); State ex rel. Wis. Senate v. Thompson, 144 Wis. 2d 429, 451, 424 N.W.2d 385 (1988). Thus while Petitioners argue that their substantive-part analysis is separate from, and in addition to, the “complete and workable law” requirement, our case law in fact holds that if the veto results in a “complete and workable law,” then the veto approved the original bill “in part.” See Wis. Senate, 144 Wis. 2d at 457 (clarifying that “the test in the veto of parts is simply whether what remains after the governor's veto is a complete and workable law.”).

¶36 But even more importantly, our cases have repeatedly emphasized that a partial veto may affirmatively change the policy of the original bill. As we said in State ex rel. Sundby v. Adamany, 71 Wis. 2d 118, 134, 237 N.W.2d 910 (1976), “the constitutional requisites of art. V, sec. 10, fully anticipate that the governor's action may alter the policy as written in the bill sent to the governor by the legislature.”¹ In other words, the governor may, through a partial veto, change the bill's substance. To date, the only limitation we have placed on the governor's ability change the substance of a bill via partial veto is that a partial veto may not “result in the creation of totally new, unrelated or non-germane provisions.” Wis. Senate, 144 Wis. 2d at 451. While this limitation does require the substance of the post-veto text to be related in some way to the substance of the pre-veto text, it does not require the post-veto substance to be “part” of the pre-veto substance.

¶37 Our decisions in Sundby and Kleczka illustrate the tension between our case law and Petitioners’ position. In Sundby, the governor's partial vetoes converted a provision for optional, voter-initiated referenda on proposed local tax increases into mandatory referenda. 71 Wis. 2d at 124. And in Kleczka, the governor's partial vetoes transformed a taxpayer's option to contribute $1 of her own money to a public campaign fund into a taxpayer's power to obligate the state to contribute $1 to the fund. 82 Wis. 2d at 685. We upheld the vetoes as constitutional in both cases without considering—as Petitioners argue we must—whether what remained after the vetoes was substantively “part” of the original bills. See Sundby, 71 Wis. 2d at 135; Kleczka, 82 Wis. 2d at 707–08. If we had, we would have rejected the vetoes as unconstitutional. After all, a mandatory referendum is not substantively “part” of an optional one, and a $1 obligation by the state is not substantively “part” of a $1 contribution by a taxpayer.

¶38 A final problem with Petitioners’ position is that we expressly stated that C.U.B. should not be read as conflicting with any of our prior decisions. Under Petitioners’ reading, C.U.B. represents a sea-change in our approach to assessing the constitutionality of attempted partial vetoes under § 10(1)(b) by imposing a requirement, never before articulated or applied, that the result of the veto must be substantively “part” of the original bill. But in C.U.B., we described our decision as merely the “logical extension” of our prior decision in Wisconsin Senate ² and as “not infring[ing] on the prior case law regarding the governor's partial veto authority.” C.U.B., 194 Wis. 2d at 503. Petitioners’ interpretation should be rejected because it would put C.U.B. in tension with our previous decisions, contrary to C.U.B.’s express directive.

¶39 In sum, I reject Petitioners’ argument that § 10(1)(b) requires the result of a partial veto to be substantively “part” of what the legislature originally passed because it is incompatible with our longstanding approach to the constitutionality of partial vetoes under § 10(1)(b). And perhaps for the same reasons, even the dissent does not adopt Petitioners’ position. Instead, the dissent argues that we should revisit all of our case law under § 10(1)(b), at least since Henry. See, e.g., Dissent, ¶92.

¶40 Although I am open to revisiting our § 10(1)(b) jurisprudence, this case is not a “clear opportunity” to do so. Id., ¶3. Petitioners do not ask us to overturn any of our prior decisions, let alone reimagine completely our approach under § 10(1)(b). And Bartlett v. Evers, 2020 WI 68, 393 Wis. 2d 172, 945 N.W.2d 685, wasn't such a clear opportunity either as the petitioners there did not offer a workable alternative to our existing approach. See Bartlett, 2020 WI 68, ¶111, 393 Wis. 2d 172, 945 N.W.2d 685 (Ann Walsh Bradley, J., concurring in part and dissenting in part). Accordingly, because upholding the partial vetoes in this case is consistent with our precedent, I respectfully concur.

¶41 How does a bill become a law? According to the majority, one option looks like this: The legislature passes a bill in both houses and sends it to the governor. The governor then takes the collection of letters, numbers, and punctuation marks he receives from the legislature, crosses out whatever he pleases, and—presto!—out comes a new law never considered or passed by the legislature at all. And there you have it—a governor who can propose and enact law all on his own.

¶42 This fantastical state of affairs did not appear all at once. The people of Wisconsin gave the governor the power to partially veto appropriation bills 95 years ago. But as governors pushed the boundaries over the last half-century, this court largely responded by throwing up its hands. And now, what the constitution calls the power to “approve[ ] in whole or in part” has transformed into the monarchical authority of one person to create brand new laws from scratch. Instead of reading what the bills actually say, and construing the partial veto power accordingly, this court treats bills presented to the governor as simply a set of alphanumeric ingredients from which the governor can cook up whatever he pleases.

¶43 One might scoff at the silliness of it all, but this is no laughing matter. The decision today cannot be justified under any reasonable reading of the Wisconsin Constitution; the majority does not suggest otherwise. Yet when presented with a clear opportunity in this case to reboot our mangled jurisprudence, the majority responds by blessing this constitutional monstrosity, all the while pretending its hands are tied. The cases the majority relies on make a mockery of our constitutional order. This is a mess of this court's making, and it is long past time for us to fix it.

¶44 Our constitution grants the legislature the power to make the law, and the governor the power to veto—that is, to reject—proposed legislation. Here, the legislature passed a proposal that permitted school districts to increase taxes during the two years of the 2023–25 biennial budget. The governor then used his “veto” pen to rewrite this proposal to permit a tax increase every year until 2425—a nifty 400-year tax increase. This new law was never voted on, passed, or proposed by the legislature. Our constitution does not countenance the creation of new laws that never go through the legislative process. The governor has no power to unilaterally enact laws that were never passed by the legislature, and we should say so. It is not groundbreaking to recognize that the legislature is vested with lawmaking authority, and the executive branch is not, even if this court has ignored that for some time. I respectfully dissent.

I. GUBERNATORIAL LAWMAKING

¶45 This case arose when Governor Tony Evers engaged in unilateral gubernatorial lawmaking. The 2023–25 budget bill initially proceeded through the normal course of lawmaking. It was debated and passed both the senate and assembly. Among the budget bill's manifold provisions was a policy permitting school districts to increase their tax revenues for both the 2023–24 and 2024–25 school years.

¶46 The bill was presented to the Governor, which he signed along with making various partial vetoes. With regard to the “veto” challenged here, the Governor selectively deleted numbers, words, and punctuation marks, rewriting the bill to provide that the increase was approved not for the 2023–24 and 2024–25 school years, but for the 2023–2425 school years. His creative editing is shown below:

Section 402. 121.905 (3)(c)9. of the statutes is created to read: 121.905(3)(c)9. For the limit for the 2023–24 school year and the 2024–25 school year, add $325 to the result under par. (b).

Section 403. 121.91 (2m)(j)(intro.) of the statutes is amended to read: 121.91(2m)(j)(intro.) Notwithstanding par. (i) and except as provided in subs. (3), (4), and (8), a school district cannot increase its revenues for the 2020–21 school year, the 2023–24 school year, and the 2024–25 school yearto an amount that exceeds the amount calculated as follows: ․

Section 404. 121.91 (2m)(j)2m. of the statutes is created to read: 121.91(2m)(j)2m. In the 2023–24 school year and the 2024–25 school year, add $146.

Section 408. 121.91(2m)(t)1.(intro.) of the statutes is amended to read: 121.91(2m)(t)1.(intro.) If 2 or more school districts are consolidated under s. 117.08 or 117.09, in the 2019–20 school year, the consolidated school district's revenue limit shall be determined as provided under par. (im), in the 2020–21 school year, 2023–24 school year, or 2024–25 school year, the consolidated school district's revenue limit shall be determined as provided under par. (j), and in each school year thereafter, the consolidated school district's revenue limit shall be determined as provided under par. (i), except as follows: ․

2023 Wis. Act 19, §§ 402–04, 408.

¶47 Two Wisconsin taxpayers, Jeffrey LeMieux and David DeValk, brought an original action against Governor Evers and others arguing that Evers’ actions violated the governor's partial veto authority under both Wis. Const. art. V, § 10(1)(b) and Wis. Const. art. V, § 10(1)(c). We granted the petition.

II. CONSTITUTION 101

A. The Legislature Makes the Law

¶48 The issue before us is whether the Governor's purported “veto” violates the Wisconsin Constitution. This requires us to interpret the constitution—something the majority gives lip service to, but never actually does. “[T]he purpose of constitutional interpretation is to determine what the constitutional text meant when it was written, commonly called the original public meaning or original understanding.” Wis. Just. Initiative, Inc. v. WEC, 2023 WI 38, ¶21, 407 Wis. 2d 87, 990 N.W.2d 122. And we do this by directing our attention to “the constitutional text, reading it reasonably, in context, and with a view of the provision's place within the constitutional structure.” Id. We may also examine other helpful aids, “such as the debates and practices at the time of adoption.” Id.

¶49 The crucial question in this case is how law is made in Wisconsin and what role the partial veto plays in the creation of new laws. While the majority quotes the constitutional text describing the governor's veto powers, it never endeavors to interpret it. And it entirely ignores the constitutional context in which that language appears. So we begin with that broader context and answer a foundational question: Under our constitution, how does a policy proposal become law?

¶50 Article IV of the Wisconsin Constitution specifies that the “legislative power shall be vested in a senate and assembly.” Wis. Const. art. IV, § 1. This language parallels the two other kinds of power in the constitution—the executive and judicial powers—which are also “vested” in a governor and unified court system, respectively. Id. art. V, § 1; id. art. VII, § 2. To “vest,” we have explained, means to “clothe” with, or “put in possession of,” a particular power. Serv. Emps. Int'l Union, Loc. 1 v. Vos, 2020 WI 67, ¶31, 393 Wis. 2d 38, 946 N.W.2d 35 (internal citations omitted).

¶51 The words “legislature” and “legislative” come from the Latin word “legis,” which means “law,” and the suffix “-latus,” which means carrying or bringing or proposing. Thus, the legislative power is, quite literally, the power to bring forth or propose the law. We have described it as the power “to declare whether or not there shall be a law; to determine the general purpose or policy to be achieved by the law; [and] to fix the limits within which the law shall operate.” State ex rel. Wis. Inspection Bureau v. Whitman, 196 Wis. 472, 505, 220 N.W. 929 (1928). Laws, as understood at the time of the ratification of the Wisconsin Constitution, were understood to be “rules of civil conduct, or statutes, which the legislative will has prescribed.” Thomas M. Cooley, Treatise on the Constitutional Limitations Which Rest Upon the Legislative Power of the States of the American Union 90 (1868).

¶52 And indeed, that is exactly how our constitution speaks. Article IV, § 17 of the constitution is entitled, “Enactment of laws.” Subsection 1 specifies the style of all laws, which “shall be ‘The people of the state of Wisconsin, represented in senate and assembly, do enact as follows.’ ” Lest the significance be missed, this means that if a law has been enacted, it has occurred as an act of the people through their elected representatives in the assembly and senate. Whatever becomes law must be the result of the will and action of the legislature.

¶53 In every sense and throughout the constitution, the legislature is described as the primary actor in the enactment of laws.¹ This is true on general matters as well as finance-specific matters and appropriations. It is the legislature that must appropriate money, levy taxes, and borrow money. Wis. Const. art. VIII, §§ 1, 2, 5, 7. This is important enough that our constitution requires an on-the-record vote for all fiscal bills in the legislature with yeas and nays recorded and a 3/5 quorum present. Id. art. VIII, § 8.

¶54 So subsection one of Article IV, § 17 makes clear that the legislature is the branch that makes all laws. Subsection two tells us how: “No law shall be enacted except by bill.” Id. art. IV, § 2. At their most fundamental, bills are comprised of policies that have been formed into a set of legal commands and instructions. See Bartlett v. Evers, 2020 WI 68, ¶191, 393 Wis. 2d 172, 945 N.W.2d 685 (Kelly, J., concurring in part, dissenting in part); id., ¶233 (Hagedorn, J., concurring). All bills “originate in either house of the legislature.” Wis. Const. art. IV, § 19. After a majority in one house votes to pass it, it must go to the other for further deliberation, potential amendments, and another vote. Id.

¶55 This requirement that both houses must pass the bill, called bicameralism, is not a useless procedural hoop to jump through. In our constitutional framework, as in the federal one, bicameralism ensures that before a law “can impose new legal limits or obligations on the people, it must secure the concurrence of many different actors, answering to many different electorates, in many different elections.”² Neil Gorsuch, A Republic, If You Can Keep It 40 (2019). Subjecting bills to scrutiny by representatives with diverse interests protects the voice of those whose ideas might otherwise be ignored by the majority. Id. It also provides a “salutary check upon rash and inconsiderate legislation” by giving more opportunities for legislators and interested constituents to weigh in. See Movement for Statehood, 1845-1846 343 (Milo M. Quaife, ed. 1918).

¶56 Our founders were intentional about placing the legislative power in the hands of the legislature. Joseph A. Ranney, Trusting Nothing to Providence: A History of Wisconsin's Legal System 51 (1999). They designed the Wisconsin legislature to be the institution most animated by, representative of, and responsive to the people. See State ex rel. Van Alstine v. Frear, 142 Wis. 320, 348, 125 N.W. 961 (1910); Gabler v. Crime Victims Rts. Bd., 2017 WI 67, ¶60, 376 Wis. 2d 147, 897 N.W.2d 384. This representative design was meant to “unit[e] a disparate group of people into one society” by providing a collective “mechanism for bringing together, negotiating, and resolving the different interests.” Neomi Rao, Why Congress Matters: The Collective Congress in the Structural Constitution, 70 Fla. L. Rev. 1, 10, 12 (2018). Following the pattern of the federal constitution, the Wisconsin legislature is the institution “by which the body of the people can act; the only way in which their opinions can be known and collected; the only means by which their wills can be united, and their strength exerted.” John Adams, Defence of the Constitutions of the Government of the United States, reprinted in 1 The Founders’ Constitution 119, 120 (Philip B. Kurland & Ralph Lerner eds., 1987). In other words, by vesting the power to legislate in the legislature, the Wisconsin Constitution situates the power to make laws in “the collective wisdom of the people and their representatives.” Gorsuch, supra at 40.

B. The Governor May Veto Legislation, Not Create It

¶57 This brings us to the governor's role in the legislative process. While the legislature is the prime actor in turning policy proposals into law, it is not the only constitutional actor. Article V, Section 10 of the Wisconsin Constitution provides that after passing both houses of the legislature, bills must be presented to the governor. Wis. Const. art. V, § 10(1)(a). For non-appropriation bills, the governor is given two and only two options. He may “approve[ ] and sign[ ] the bill,” in which case “the bill shall become law.” Id. art. V, § 10(1)(b). Or he may “reject[ ]” the bill, which the title of the section calls a “veto.” Id. art. V, § 10(2)(a). If the bill is rejected, he is to “return the bill, together with the objections in writing, to the house in which the bill originated.” Id. art. V, § 10(2)(a). After the legislature “reconsider[s]” the bill, it may “agree to pass the bill notwithstanding the objections of the governor” by a two-thirds vote in each legislative house. Id. art. V, § 10(2)(a). Save for appropriation bills, the governor's veto power is all-or-nothing. Id. art. V, §§ 10(1)(b), 10(2)(a). Either the entire bill becomes law or none of it does.

¶58 While the general veto power has been in the constitution since its adoption in 1848, the partial veto authority that is the subject of this case did not come until later in Wisconsin's history. But before describing that development, it is worth considering what the drafters of our constitution meant by authorizing the governor to “veto” proposed bills.

¶59 To “veto” comes from the Latin for “I forbid.” It was a term well known in the law at the time it was put into the Wisconsin Constitution. As a legal dictionary at the time described it, to “veto” meant “the power enjoyed by the executive department of a government, of negativing bills which have been passed by the legislature.” Veto, A New Law Dictionary 1035 (1851). Indeed, our constitution uses the term “veto” in the title, and then uses the synonym “reject” in the text of art. V, § 10(2)(a). To reject means “[t]o refuse (something offered)” or “to decline.” Reject, The Shorter Oxford English Dictionary on Historical Principles 1694 (1933). And thus, to “veto” is to reject, refuse, or decline what has been proposed; it is the power to say “no.” Since a veto is the power to reject a proposal, it logically cannot be the power to create. 82 C.J.S. Authority for and Nature of Veto of Bill by Executive § 66 (2024) (“The veto power is not the power to enact new laws, recall or modify old laws, repeal laws, broadly affect public policy, alter legislative intent, or declare a bill constitutionally invalid.”). To put it simply, the veto power is one of negation, not creation.

¶60 The veto power is an aberration from and exception to the default constitutional structure. A veto gives the governor a powerful voice in whether a legislative proposal becomes law. This function is not naturally within the power to execute the law. Rather, it serves as a limit on the legislature's vested power to make law. Alexander Hamilton similarly described the president's veto in Federalist No. 73 as a “qualified negative” meant to check legislative power. The Federalist No. 73, at 494 (Alexander Hamilton) (J. Cooke ed. 1961). The delegates to the Wisconsin Constitutional Convention of 1846 were clear that “the governor's powers should be limited and the primary lawmaking power should reside with the legislature,” while acknowledging a narrow exception for the gubernatorial veto. Ranney, supra at 51. In that way, the “veto is simply one of the instances in which our framers broke off a small piece of power that naturally belongs in one branch and put it in another.” Bartlett, 393 Wis. 2d 172, ¶186 (Kelly, J., concurring in part, dissenting in part). But this quasi-legislative power to veto is still at root the power to influence legislation by subtraction, not addition. All laws—policy proposals that have been formed into legal commands and instructions in a bill—must come from the legislature.³ And while the governor may reject bills presented to him, he may not affirmatively design them on his own. This is Wisconsin's constitutional design.

¶61 The amendment granting the governor the power to partially veto appropriation bills did not upend this constitutional structure. Before the turn of the 20th century, appropriation bills—bills that authorize the spending of public money—contained a single appropriation. See Richard A. Champagne, The Wisconsin Governor's Partial Veto after Bartlett v. Evers, Legislative Reference Bureau 3 (July 2020). If the governor thought the appropriation ill-advised, he could veto it, as he could any other bill. But then the legislature began to place multiple, unrelated appropriations into a single bill. State ex rel. Martin v. Zimmerman, 233 Wis. 442, 447–48, 289 N.W. 662 (1940). This presented the governor with a dilemma: veto large, omnibus appropriation bills in their entirety, or approve them in full. See Id. at 448.

¶62 It was this dilemma that led to the adoption of a constitutional amendment in 1930. State ex rel. Wis. Tel. Co. v. Henry, 218 Wis. 302, 315, 260 N.W. 486 (1935). The amendment read, as it does today, “[a]ppropriation bills may be approved in whole or in part by the governor, and the part approved shall become law.” Wis. Const. art. V, § 10(1)(b). Since the legislature began to add multiple unrelated policy proposals into one piece of legislation, the idea was to grant the governor a concomitant power to approve some of the policy proposals and reject others. See Henry, 218 Wis. at 315; Bartlett, 393 Wis. 2d 172, ¶233 (Hagedorn, J., concurring).

¶63 While this amendment certainly conferred significant power on the governor, nothing from the debates at the time, early cases, or language suggests this kind of veto was a grant of magical, unilateral power to make law; it was still a license to reject. No textual, historical, or structural evidence suggests the ability to partially veto appropriation bills was aimed at giving the governor power to singlehandedly fashion new legislation.⁴ There is no indication this new authority was intended to blow a hole through the vesting clauses, giving the governor grand new powers to affirmatively legislate. No one suggested this was anything but the simple power to reject some legislative proposals, and accept others, in an appropriation bill.

III. THIS COURT IGNORES THE CONSTITUTION

A. How This Court Strayed

¶64 Given this background, one wonders how this court transformed the power to reject some legislative proposals into a new kingly power—one in which a single person can rewrite a bill to say something totally different and make that the law instead. It's enough to make even King John blush.⁵ It is no stretch to say that everyone who was involved in the passing of this amendment, every early governor, every legislator, and every voter who voted for or against this amendment would be appalled at how this court has distorted it. No one could have seen this coming, because what the majority sanctions today has no relationship to the amendment adopted in 1930. So how did we get here?

¶65 We first considered the meaning of the partial veto amendment in 1935, just five years after its ratification. Henry, 218 Wis. 302. We engaged in a plain-meaning analysis of the text to determine whether the governor could veto portions of a policy proposal or only entire legislative policy proposals called “items.” Id. at 310–11. We noted that our constitution allowed bills to be approved “in part,” rather than copying other states’ provisions that permitted “line item” vetoes, in which governors could only veto entire policy proposals. Id. at 311; see also Bartlett, 393 Wis. 2d 172, ¶247 (Hagedorn, J., concurring). So in using the word “part” instead of “item,” the Wisconsin Constitution gave the governor broader ability to veto something less than an entire item or proposal. Henry, 218 Wis. 2d at 313. Because of that, the court had to determine what constituted a “part” of an appropriation bill. Turning to contemporary dictionaries, the court defined a part as:

One of the portions, equal or unequal, into which anything is divided, or regarded as divided; something less than a whole; a number, quantity, mass, or the like regarded as going to make up, with others or another, a larger number, quantity, mass, etc., whether actually separate or not; a piece, fragment, fraction, member, or constituent.

Id. This definition of “part” did not give the governor boundless discretion to veto any jot or tittle contained in the bill, as the court would later corrupt it to mean. Instead, the Henry court recognized the power to veto “in part” was bounded by “both procedural and substantive limitations” on the governor's partial veto power. Bartlett, 393 Wis. 2d 172, ¶248 (Hagedorn, J., concurring).

¶66 Procedurally, the governor could not strike parts of the bill such that what remained did not make sense as a matter of form. That is, a governor could strike parts of an appropriation bill so long as “the parts approved, as they were in the bill, as it was when originally introduced ․ constitute, in and by themselves, a complete, entire, and workable law ․” Henry, 218 Wis. at 314. Said another way, the governor could strike a part of a bill so long as what remained could actually become a coherent law. Bartlett, 393 Wis. 2d 172, ¶248 (Hagedorn, J., concurring).

¶67 Substantively, the court suggested that the governor could only veto parts that “were not essential, integral, and interdependent parts of those which were approved.” Henry, 218 Wis. at 317. The court invoked and cited principles of severability, explaining that even if the governor's veto would leave a complete and workable law, it is impermissible if it is evident “from the [bill] itself that the legislature intended [the bill] to be effective only as an entirety and would not have enacted the [remaining] part alone.” Id. at 316. We summarized our holding in Henry in a case the following year as follows: “the 1930 amendment permits the veto by the governor of any separable part of an appropriation bill.” State ex rel. Finnegan v. Dammann, 220 Wis. 143, 146, 264 N.W. 622 (1936) (emphasis added).

¶68 Thus, from the very beginning, this court recognized that there were limitations on the governor's ability to approve in part well beyond the procedural requirement that the remaining law be a complete and workable bill. In no sense did Henry suggest or imply that governors could pick and choose among letters and numbers to rewrite a proposed law. Logically, Henry's separability discussion only makes sense if the separate policy items were there to begin with and approved by the legislature. Rewriting language to mean something different is worlds away from questions of severability and interdependence. And not a single word of Henry supports the idea that the power to partially veto a bill allows the governor to enact a new policy never passed by the legislature.

¶69 Rather, for decades, Henry continued to be cited for the idea that when some parts of a law are invalid, the remainder stands so long as it is “consistent with the intention of the Legislature which enacted it.” Zimmerman v. Zeimet, 259 Wis. 619, 624, 49 N.W.2d 924 (1951). It would not be permissible under Henry if “it clearly appears that the provisions [struck] are so intimately and inherently related to, and connected with, the general provisions to which it relates that the legislature would not have enacted the latter without the former.” Id. Indeed, it is fair to say that Henry was understood and cited as black letter law for its statements on severability. See, e.g., State ex rel. McStroul v. Lucas, 251 Wis. 285, 291, 29 N.W.2d 73 (1947) (citing Henry for a severability analysis focused on whether the provisions “are separable ․ and were probably intended to stand even if said final clause is invalid”); State ex rel. Milwaukee Cnty. v. Boos, 8 Wis. 2d 215, 224, 99 N.W.2d 139 (1959) (same); Town of Burke v. City of Madison, 17 Wis. 2d 623, 636, 117 N.W.2d 580 (1962) (same). When what remains is an entirely new proposal, it logically follows that the remainder was not intended by the legislature.

¶70 This was the prevailing understanding until a 1976 case in which the governor vetoed parts of sentences within a bill. State ex rel. Sundby v. Adamany, 71 Wis. 2d 118, 121–23, 237 N.W.2d 910 (1976). By removing parts of these sentences, the governor mandated a town referendum to increase tax levies that the legislature originally made optional. Id. at 124. This was a policy that the legislature “had neither proposed nor approved” and, as such, was not a mere negative to the proposed bill, but a creation. Bartlett, 393 Wis. 2d 172, ¶185 (Kelly, J., concurring in part, dissenting in part). The court acknowledged that the governor's use of the partial veto to do this was not an act of negation, but an “affirmative change in the result intended by the legislature.” Sundby, 71 Wis. 2d at 134. Rather than follow Henry (even while saying it was), the court reasoned that since every partial veto creates “a change of policy,” there was no distinction between blocking a part of the bill versus selectively editing it to say something it never was intended to say. Id.

¶71 This was a remarkable conclusion. A policy proposal never passed by the legislature at all—no bicameralism and presentment—could now become law through a complete gubernatorial rewrite. Rather than determine whether the legislature's proposals could be separated, with some approved and some rejected, we now endorsed the governor creating proposals by himself. In so doing, this court reallocated authority belonging to the legislature and gave it the governor—something we have no authority to do. And so the wildest, most bizarre partial “veto” jurisprudence in the country was born.

¶72 We solidified this constitutional inversion two years later in State ex rel. Kleczka v. Conta, 82 Wis. 2d 679, 264 N.W.2d 539 (1978). In Kleczka, we examined a partial veto where the governor struck out words, changing the law so that money taxpayers could have chosen to pay into a campaign fund would now automatically come from the state's general fund. Id. at 703. We allowed this exercise of power by formally adopting only the portion of Henry that said a bill was severable so long as what remained after the veto was a “complete, entire, and workable law.” Id. at 706. In other words, the governor could now veto in “part” by striking words in a sentence that altered the meaning of the bill entirely.

¶73 What of the rest of Henry? We discarded the additional substantive limitations of Henry as mere dicta which did “not correctly state the Wisconsin law.” Id. at 715. We openly declared that the governor can, through vetoing appropriation bills in part, adopt and create new policy that never goes through the constitutional requisites for legislation. Id. The only thing that matters, we declared, is that what is left be “a complete, entire, and workable bill.” Id.

¶74 Justice Hansen wrote a powerful and prophetic partial dissent. He pointed to the separate powers vested in each branch—a principle heretofore “jealously guarded”—recognizing “that an invasion of the province of one branch by another is an attack upon the constitutional foundation of the government itself, and in a sense, upon the liberty of our citizens.” Id. at 718 (Hansen, J., concurring in part, dissenting in part). It is the legislature that has been exclusively vested with the power to make the laws. Id. at 719. “Unless we are prepared to abandon [the concept of the separation of powers] then there must be some palpable limit to the power of the governor to rewrite, by the device of the partial veto, bills which have passed the legislature.” Id. at 719. He presciently warned, “[o]nly the limitations on one's imagination fix the outer limits of the exercise of the partial veto power by incision or deletion by a creative person.” Id. at 720.

¶75 The simple constitutional boundary Justice Hansen identified, which calls to us again today, is this: “At some point this creative negative constitutes the enacting of legislation by one person, and at precisely that point the governor invades the exclusive power of the legislature to make laws.” Id. Our constitution provides that “the governor is to review the laws and not to write them.” Id. He may not “ ‘write with his eraser’ ” “to devise new bills which will become law unless disapproved by two-thirds of the legislators who are elected by the people of the state.” Id. The discarding of substantive limitations on the partial veto power, and the adoption of the “complete, entire, and workable” test as the exclusive limitation on the partial veto means that the governor has “for all practical purposes, unlimited authority to exercise power reserved by the constitution to the legislature.” Id. at 723.

¶76 Since Kleczka, Justice Hansen's worst nightmare has come true. Governors have become ever more creative, and the court has continued down the absurd path of allowing the governor to scratch out anything on the face of the bill and construct entirely new statutory commands. In Thompson, the court sanctioned the veto of “individual words, letters and digits” by a 4-3 vote so long as what remained was a “complete, entire, and workable law.” State ex rel. Wis. Senate v. Thompson, 144 Wis. 2d 429, 437, 424 N.W.2d 385 (1988). A “part” of a bill was no longer a policy proposal reduced to legal commands; it was no longer even words. Now a bill was simply a collection of individual letters and digits. The court also said that the governor could strike out digits to reduce an appropriation amount. Id. This is because letters and digits were “parts” that could be vetoed, even if what was left appropriated an amount never voted on by the legislature. Recognizing the tension in granting the governor seemingly unbounded legislative authority, we also identified a “germaneness limitation” on the veto power. Id. at 452–53. Yet we nonetheless rejected the idea that “the governor cannot affirmatively legislate by the use of the partial veto power.” Id. at 453. In response, the people passed an amendment in 1990 that prohibited the governor from “rejecting individual letters in the words of the enrolled bill.” Wis. Const. art. V, § 10(1)(c). In 2008, the people added a further prohibition forbidding the governor from “creat[ing] a new sentence by combining parts of 2 or more sentences of the enrolled bill.” Id.

¶77 In this new game of cat-and-mouse, the governor then tried something even more novel. Instead of merely striking out certain words or numbers, he struck out a number and wrote in an entirely new one. Citizens Util. Bd. v. Klauser, 194 Wis. 2d 484, 488, 534 N.W.2d 608 (1995). We concluded that too was fine. If the governor could strike out digits to make a smaller appropriation amount, it followed that he could also strike out a number and write in a smaller one. Id. at 506. We further reasoned that the smaller number was necessarily a “part” of the larger one, so this was constitutional. Id. at 505. We limited our holding to appropriation amounts only, however, without offering any textual or logical reason for that limitation. Id. at 510. In dissent, Justice Abrahamson recalled the ghost of partial veto cases past: “Justice Connor T. Hansen, dissenting in the Kleczka case, objected to a governor writing laws with the eraser end of the pencil. Today the majority allows a governor to write laws with the pointed end of the pencil.” Id. at 511 (Abrahamson, J., dissenting). In short, the one-man legislature was alive and well.

¶78 In the last two cases where this court substantively addressed the partial veto power, we struck down the governor's actions. In Risser v. Klauser, we concluded the write-in veto approved in Citizen's Utility Board “may be exercised only on a monetary figure which is an appropriation amount,” striking down a written-in reduction to a revenue bonding limit. Risser, 207 Wis. 2d 176, 181, 558 N.W.2d 108 (1997).

¶79 And five years ago, we began to right the ship, striking down three of the Governor's attempted “vetoes.” Bartlett, 393 Wis. 2d 172, ¶9 (per curiam). In the first, the governor changed a school bus modernization fund into an alternative fuel fund. Id., ¶270 (Hagedorn, J., concurring). Second, the governor transformed a local road improvement fund into a more general local grant fund. Id., ¶272. And finally, the governor rewrote a vapor products tax into a broader tax that includes liquid heated by a vaping device. Id., ¶274. Although the court did not issue an opinion with a controlling rationale, a majority of justices recognized that none of these proposals had been voted on by the legislature. See id., ¶¶223, 225, 228 (Kelly, J., concurring in part, dissenting in part, joined by Rebecca Grassl Bradley, J.); Id., ¶¶271, 273, 275 (Hagedorn, J., concurring, joined by Ziegler, J.). Since the effect of each “veto” was a new gubernatorial proposal, and not merely gubernatorial approval or disapproval of legislative proposals, a majority of the court concluded they were not consistent with the constitution. See id., ¶¶223, 225, 228 (Kelly, J., concurring in part, dissenting in part, joined by Rebecca Grassl Bradley, J.); Id., ¶¶271, 273, 275 (Hagedorn, J., concurring, joined by Ziegler, J.).

B. The Majority Continues the Constitutional Detour

¶80 Today, the court is offered a chance to extend Bartlett’s progress and end the anti-constitutional, jurisprudential mess this court has made. The majority instead suggests it is duty-bound to double down on our pre-Bartlett madness. It is not.

¶81 The majority does not even feign interest in the original meaning of the constitution. Instead, it reasons that our precedent and the 1990 amendment establish four—and only four—limitations on the partial veto power. First, the remaining parts of the bill must constitute a complete, entire, and workable law. Second, deletion vetoes may only be exercised on bills containing appropriations within their four corners. Third, the deletion vetoes may not result in a law that is not germane to the original bill. And fourth, while the governor can strike individual words, letters, or numbers, he cannot create a new word by rejecting individual letters, nor may he create a new sentence by combining parts of two or more sentences. Majority op., ¶12. In addition to these limitations on a deletion veto, the governor may strike an appropriation amount and write in a smaller appropriation amount. Id., ¶13.

¶82 The majority then reasons that because the vetoes here do not violate any of these principles, they must be constitutionally permissible. Id., ¶24. What the majority does not explore is whether any of this has anything to do with our actual constitution. The end result is that the majority gives the governor a green light to do what he constitutionally cannot—create new law all by himself.

¶83 Thus, the majority says that a governor's deletion vetoes are constitutional “as long as the remaining text of the bill constitutes a complete, entire and workable law.” Id., ¶12 (internal quotation omitted). The majority cites Henry, implying the court has consistently used this framework since 1935. Id., ¶15. But as previously explained, the idea that this is all that is required is from decades later in Kleczka, which dispensed with language in prior cases and contradicts later holdings that the veto must be germane. Thus, the majority takes disparate holdings and merely assumes this is the final and only word, rather than engaging with the constitution's text, history, and structure to prove it. The same is true with the majority's contention that the governor “may strike words, letters, or numbers.” Id., ¶12 (internal citation omitted). This conclusion from Thompson is totally divorced from the plain meaning of Wis. Const. art. V, § 10(1)(b) and its context. See Thompson, 144 Wis. 2d 429, 434.

¶84 The majority's logic is basically this: Our prior cases suggest the governor can be a unilateral lawmaker by using the potpourri of letters he receives from the legislature and fashioning them to his liking. A few other boundaries have been delineated—germaneness, and the constitutional amendments—limiting the creation of new words or sentences. Changing durations, with the effect of creating new policy that was never proposed nor passed by the legislature, doesn't neatly fall into any of these restrictions, so it must be constitutional. This logic flouts the constitution's text and structure, and the wisdom that underlies both.

¶85 Contrast, for instance, how our constitution prescribes lawmaking with the gubernatorial lawmaking the majority permits. First, a law must start out as a bill that originates in either the senate or assembly. Wis. Const. art. IV, §§ 17(2), 19. Under the majority's reading, however, a law can originate from a single person not entrusted with the constitutional power to make law. Bartlett, 393 Wis. 2d 172, ¶195 (Kelly, J., concurring in part, dissenting in part).

¶86 Furthermore, the governor does not have to abide by the constitutional strictures placed on the legislature when it makes law. For instance, a bill must go to both houses of the legislature for deliberation, amendment, and a vote. Wis. Const. art. IV, § 19. Under the majority's reading, one man, needing no advice, approval, or input, can create brand new policies the legislature never considered.

¶87 The constitution further specifies that bills must be presented to the governor as one final check on the legislature. Wis. Const. art. V, § 10(1)(a). Under the majority's almost-anything-goes doctrine, one person can create policy that only a supermajority of the legislature can overturn. Bartlett, 393 Wis. 2d 172, ¶173 (Kelly, J., concurring in part, dissenting in part). The effect is to flip the constitutional roles of the legislature and the governor. See id., ¶266 (Hagedorn, J., concurring). The governor now makes the law, and the legislature must try to “veto” it. Id., ¶173 (Kelly, J., concurring in part, dissenting in part). This is not how our constitution says the system is supposed to work.

IV. THE CORRECT RESULT

¶88 So how should the court handle the 402-year “veto”? By doing what the majority suggests, but never does: “ ‘focus on the constitutional text, reading it reasonably, in context, and with a view of the provision's place within the constitutional structure.’ ” Majority op., ¶9 (citing Wis. Justice Initiative, Inc., 407 Wis. 2d 87, ¶21). Once we do so, it is clear that the Governor's “veto” in this case is not a veto at all, but merely gubernatorial lawmaking that is repugnant to our constitutional structure. Bartlett, 393 Wis. 2d 172, ¶244 (Hagedorn, J., concurring).

¶89 As we have explained, a “veto” is a power of negation. It allows the governor to do nothing more than to reject laws that the legislature has proposed. See 82 C.J.S. Authority for and Nature of Veto of Bill by Executive § 66 (2024). The fundamental nature of a veto does not change just because the governor can veto “part” rather than all of an appropriation bill. The partial veto simply means that the governor can now reject policy proposals contained within an appropriation bill instead of being forced to reject it in its entirety. As a power to “reject,” it may assuredly change aspects of the legislature's collection of policy prescriptions; the legislature may get most of its proposals, but not all of them, enacted into law. But what the partial veto clause does not do is establish a second lawmaking branch of government. The governor has no constitutional power to create new proposals that did not originate with the legislature or go through the constitution's lawmaking process.

¶90 An appropriations bill is not merely “a potpourri of individual letters, an alphabet soup if you will,” as the majority assumes. Thompson, 144 Wis. 2d at 473 (Bablitch, J., concurring in part, dissenting in part). It contains draft statutes reflecting specific policies that have been considered and voted on by the legislature. This is what the constitution commands with all laws. So when the governor rejects part of an appropriations bill, the policy proposals that remain after the governor exercises his partial veto must still have been created and approved by the legislature in the first instance. See Bartlett, 393 Wis. 2d 172, ¶217 (Kelly, J., concurring in part, dissenting in part). Once again, we cannot lose sight of the constitution's structure. The legislative power is vested in the legislature. And the constitutional amendment giving the governor power to partially veto appropriation bills did not change this.

¶91 Here, when the bill left the legislature's hands, it permitted school districts to exceed their base tax revenue for two years, the 2023–24 and 2024–25 school years. See 2023 Wis. Act 19, §§ 402–04, 408. By striking out numbers, words, commas, and some hyphens, the governor rewrote the bill to say that districts could increase their revenue by those amounts from 2023 through 2425. Id. The legislature never proposed extending the increase through 2425. This simply was not a policy proposal considered and voted on by both houses of the legislature. This is not a policy that was presented to the governor for approval. And contra the majority, we are permitted to read the words in the bill and make sense of them, not just consider the bill an alphabet soup of options. Thus, after the governor exercised his “veto,” there was something in the bill that did not originate from the legislature, was never subject to lawmaking procedures, and was not presented to the governor. This is plainly unconstitutional.

¶92 It is true that the petitioners here do not explicitly ask us to continue the progress we made in Bartlett and formally roll back the missteps of our prior cases. But where the governor's actions are so out of step with the constitutional order, and where we are asked to apply the constitution, “the principle of stare decisis should yield to a result consistent with the plain meaning of the words within the amendment.” Thompson, 144 Wis. 2d at 467–68 (Bablitch, J., concurring in part, dissenting in part). Instead of treating the fractured legal framework with another quick fix of judicial epoxy, it is time to raze it to the ground.⁶

¶93 Even if the majority is correct that these vetoes do not transgress one of the principles identified in some of our recent cases, that does not mean they do not transgress the constitution itself. Cf. Bartlett, 393 Wis. 2d 172, ¶255 (Hagedorn, J., concurring). The only way to determine if these vetoes are impermissible is through analyzing the constitution. This duty is particularly strong given our recent decision in Bartlett, where the court rejected the approach taken by the majority here.⁷ Moreover, there are virtually no reliance interests implicated. Our decision will simply guide future behavior, and not implicate the past. So we have every reason to get the constitution right, and not perpetuate our prior errors.

V. CONCLUSION

¶94 Perhaps hoping to temper the blow to our constitution, the majority closes its opinion by offering some options. The legislature can fix the gubernatorial rewrite by changing the law with a new bill, it suggests. And to protect against future abuses, the legislature can submit a new constitutional amendment to the people. Or it can engage in more creative and defensive bill drafting to mitigate gubernatorial lawmaking. Majority op., ¶29. This will surely be cold comfort coming from a court that simultaneously strips the legislature of its constitutional powers.

¶95 The far better option would be to get the constitution right. As Justice Bablitch said in Thompson, “[i]t is not an answer to say that any gubernatorial excesses may be rectified through the ballot box or constitutional amendment, particularly when, as here, any ‘excesses’ in regard to the governor's partial veto power derive primarily from our own pen.” Thompson, 144 Wis. 2d at 475 (Bablitch, J., concurring in part, dissenting in part). He concludes: “It is far better for this court to adhere to the plain meaning of the words within the amendment and longstanding constitutional principles.” Id. Unfortunately, that work will have to wait for another day.⁸

¶96 The bottom line is this. The constitution grants the legislature the exclusive power to make the law. The governor can say no and refuse legislative proposals in appropriation bills in whole or in part, but he cannot unilaterally make his own proposals the law. This is what our constitution says and plainly means. Because the majority holds that the governor can make the law all on his own, inverting our constitutional order, I respectfully dissent.

FOOTNOTES

1.   All references to the Wisconsin Statutes are to the 2023–24 version.

2.   The provisions at issue are italicized for emphasis.

3.   We note that in our most recent review of the governor's partial veto power, Bartlett v. Evers, 2020 WI 68, 393 Wis. 2d 172, 945 N.W.2d 685, we invalidated three of the four challenged deletion vetoes in a per curiam decision. Because there was no majority opinion, it did not establish any precedent.

4.   In Bartlett, the petitioners asked this court to overrule Henry and its progeny. Bartlett, 393 Wis. 2d 172, ¶118, ¶118 n.2 (Ann Walsh Bradley, J., concurring in part and dissenting in part). The court did not do so because there was no consensus on whether to overrule our precedent or which guiding principle or principles to adopt instead. See id., ¶¶256–66 (Hagedorn, J., concurring) (summarizing the competing doctrine proposed by the litigants and other justices).

5.   See https://legis.wisconsin.gov/lrb/media/kxappfdr/760-2023-enrolled-joint-resolution-1684.pdf.

6.   See https://docs.legis.wisconsin.gov/2025/related/proposals/ajr8.

1.   See also State ex rel. Martin v. Zimmerman, 233 Wis. 442, 449–50, 289 N.W. 662 (1940) (upholding a partial veto as valid because the approved parts provided a complete and workable law, even though the veto caused a change in policy); State ex rel. Kleczka v. Conta, 82 Wis. 2d 679, 708, 264 N.W.2d 539 (1978) (“a governor's partial veto may, and usually will, change the policy of the law”); State ex rel. Wis. Senate v. Thompson, 144 Wis. 2d 429, 451, 424 N.W.2d 385 (1988) (summarizing principles from prior partial-veto cases, including that partial vetoes “may be affirmative as well as negative in effect,” and that “the governor has quasi-legislative power with respect to the exercise of his partial veto authority, and that he can be creative in the exercise of such authority”); Citizens Util. Bd. v. Klauser, 194 Wis. 2d 484, 496, 534 N.W.2d 608 (1995) (summarizing holdings from prior cases, including that partial vetoes “may significantly alter[ ] the legislative intent of the appropriation bill”).

2.   In Wisconsin Senate, we held that the governor has “broad powers to reduce or eliminate numbers and amounts of appropriations in the budget bill” and that “a partial veto resulting in a reduction in an appropriation is precisely the sort of partial veto measure the governor of this state is authorized to take pursuant to art. V, sec. 10 Wis. Const.” 144 Wis. 2d at 457, 461.

1.   Our constitution leaves no doubt into whose hands it placed the legislative power. In almost every article of our constitution, it is the legislature who is consistently tasked with defining certain terms, authorizing certain actions, prescribing certain remedies, providing for certain occasions and establishing certain policies by law. See, e.g., Wis. Const. art. I, §§ 5, 8 (2), (3), 9m, 21; id. art. IV, §§ 11, 22, 23, 24(3), (4), (6), 25, 27, 29, 32, 33; id. art. VII, §§ 2, 5, 6, 8, 10(2), 11, 12, 14, 24(2); id. art. VIII, §§ 1, 5, 7 (2)(e), (2)(f), 2(g), id. art. X, §§ 3, 8; id. art. XI, §§ 1, 3, 3a, 4; id. art. XII, §§ 1, 2; id. art. XIII, §§ 4, 9, 10(1).

2.   In Wisconsin, assembly members are elected every two years by constituents in assembly districts. Wis. Const. art. IV, § 4. Senators are chosen to serve four-year terms by constituents in senate districts. Id. art. IV, § 5.

3.   The text of the partial veto amendment seems to recognize that policy proposals are to originate from the legislature, as the legislature is to “reconsider” any of the parts of the bill the governor rejected. Wis. Const. art. V, § 10(2)(a). Re-consideration, of course, implies that the legislature previously considered the proposals when drafting the bill.

4.   At the time of the amendment's enactment, a “veto” was defined in one dictionary as “[t]he refusal of assent by the executive officer whose assent is necessary to perfect a law which has been passed by the legislative body.” Veto, Black's Law Dictionary 1813 (3d ed. 1933). In another, “[t]he act on the part of a competent person or body of preventing or checking legislative or other political action by the exercise of a prohibitory power.” Veto, The Shorter Oxford English Dictionary on Historical Principles 2352 (1933).

5.   The Magna Carta, sealed by King John in 1215, was a foundational document to our system of English common law. It limited royal authority, and established the principle that even the King was subject to the law. Magna Carta, ch. 61 (1215), reprinted in Boyd C. Barrington, The Magna Carta and other Great Charters of England 246–49 (Philadelphia, J.R. Rodgers Printing Co. 1900). Many principles from our constitutional system are inspired directly by the Magna Carta. Indeed, the idea of a single executive unilaterally rewriting laws to his own pleasure would make that despotic King blush.

6.   Many scholars and legal commentators have sounded the alarm on our bewildering partial veto jurisprudence. See e.g., Benjamin W. Proctor, Wisconsin's Chief Legislator: The Governor's Partial Veto Authority and the New Tipping Point, 90 Marq. L. Rev. 739, 742 (2007) (“[T]hrough a series of decisions addressing the extent of the partial veto, the Wisconsin Supreme Court has cornered itself into an amusingly broad interpretation of gubernatorial authority.”); Frederick B. Wade, The Origin and Evolution of Partial Veto Power, 81 Wis. Law 12, 12–14 (2008) (“The result of the current understanding [of the partial veto] is a profound contradiction. On the one hand, the Wisconsin Constitution makes clear that legislation must be authorized and enacted by the legislature in order to be a legitimate exercise of governmental power․ On the other hand, the partial veto has evolved into a unilateral executive power to create ‘law[s].’ ”); Richard Briffault, The Item Veto: A Problem in State Separation of Powers, 2 Emerging Issues St. Const. L. 85, 94 (1989) (calling Wisconsin's partial veto as interpreted by the court as “an extreme instance of the ‘executive’ approach to the item veto” and saying that “it is hard to believe [the partial veto power] was intended to go this far or that it should.”); Mary E. Burke, The Wisconsin Partial Veto: Past, Present and Future, 1989 Wis. L. Rev. 1395, 1432 (1989) (criticizing the court's partial veto jurisprudence and calling for action to “restore the balance to what was intended by the constitutional framers, to what is desired by state citizens, and to what is healthy for state government.”); John S. Weitzer, The Wisconsin Partial Veto: Where Are We and How Did We Get Here: The Definition of Part and the Test of Severability, 76 Marq. L. Rev. 625, 626, 649 (1993) (arguing that this court's “broad interpretation” of the partial veto power allows “the governor of Wisconsin [to] create legislation and foil the Wisconsin Legislature's intent.”); Anthony S. Earl, Personal Reflections on the Partial Veto, 77 Marq. L. Rev. 437, 441–42 (1994) (noting that in Wisconsin “the governor, through the use of the partial veto, can actually create laws that were never considered by the legislature” and that this arrangement does not “make[ ] for good public policy.”).

7.   It is also not as if the four members in the majority have been scrupulous defenders of stare decisis. Recently, the court has been downright aggressive in overturning cases it has thought incorrect. See Priorities USA v. WEC, 2024 WI 32, ¶61 n.3, 412 Wis. 2d 594, 8 N.W.3d 429 (Rebecca Grassl Bradley, J., dissenting). In fact, some members of the majority have opined that decisions of the court are binding precedent whether there is a controlling rationale or not. Justice Ann Walsh Bradley, joined by Justice Dallet, opined that the split nature of a decision “is of no import” when the mandate of the court is clear. Koschkee v. Taylor, 2019 WI 76, ¶73, 387 Wis. 2d 552, 929 N.W.2d 600 (Ann Walsh Bradley, J., dissenting). When the only thing that “changed is the membership of the court,” that is no reason to ignore stare decisis, my colleagues reasoned. Id., ¶62. Justice Karofsky similarly expressed on the campaign trail that her opponent ignored the rule of law in the Koschkee case by overturning precedent when “the only thing that changed was the makeup of the court.” Campaign 2020: Jill Karofsky for Wisconsin Supreme Court, WisconsinEye at 3:32 (Jan. 17, 2020). One might think these principles should be transferable, and that the logic of these criticisms means the majority should treat Bartlett as precedential. But the majority says nary a peep. It treats Bartlett as if it never happened, with no explanation why.

8.   As perhaps a silver lining, Justice Dallet expresses an openness to revisiting our partial veto precedent. See concurrence, ¶40. That means a majority of this court agrees something may be amiss in our partial veto jurisprudence. This is an encouraging development to be sure, especially when viewed alongside the majority's rather tepid application of precedent and its express reliance on the fact that the parties did not ask us revisit that precedent here.

Jill J. Karofsky, J.