The Great Detroit Charter Fight: A Case Study in Statute-Specific Purpose Versus Trans-Statutory Substantive Canons

In Sheffield and Harbin v. City Clerk, the Michigan Supreme Court is on the verge of deciding whether Michigan law gives Governor Gretchen Whitmer the power to impose an absolute veto on a new charter for Detroit. The case is naturally irresistible to a long-time Michigander and local government law junkie like myself. But it contains some broader lessons about statutory interpretation that are useful for any lawyer.

After the jump, I will suggest how the Great Detroit Charter Fight ("GDCF") illustrates a dilemma at the heart of statutory interpretation: When text is ambiguous (as it often, even usually, is), then what priority should be given between trans-statutory substantive canons versus statute-specific purpose? In Sheffield and Harbin, briefs on both sides invoke “expressio unius est exclusio alterius” canon with equal plausibility. As I argue after the jump, these lawyerly textual arguments are ultimately indeterminate not only in the GDCF but also in most other cases. Faster than you can say “Karl Llewellyn was right,” the argument over the proposed Detroit charter boils down to a conflict between two extra-textual sources of authority: The specific purpose of a 1913 statute giving the governor powers to veto proposed city charters and the general “liberal construction” canon, codified in the 1963 Michigan Constitution, requiring state laws conferring power on local governments to be “liberally construed.”

Here’s the conflict in a nutshell. The political context of the 1913 statute strongly suggests that the state legislature distrusted Detroit and wanted to give governors a hard brake on Detroit’s plans to expand municipal services like a city-owned streetcar system. If that general purpose governs statutory meaning, then Detroit’s charter commission loses and Governor Whitmer wins. On the other hand, if the statutory text is ambiguous enough to justify invoking such unwritten statutory purposes, then maybe it is ambiguous enough to be “liberally construed” to favor of Detroit’s autonomy. Perhaps the Court should sidestep a divisive, ideologically charged inquiry into how much or little the state legislature distrusted urban progressives in the early 20th century and instead invoke a bright-line trans-statutory canon favoring local power.

I do not have any easy answer to how willing judges should be to favor trans-statutory canons over statute-specific purpose. The question turns on a trade-off between fidelity to the enacting state legislature’s likely purpose versus protection of an apolitical simplicity in statutory interpretation that keeps judges out of fraught ideological controversies. This problem, however, transcends Michigan law: The Chevron canon, for instance, raises precisely the same dilemma. For those who are interested in statutory interpretation, therefore, the fight described after the jump will hold some interest.

I. Why textual canons are ultimately indeterminate in the Great Detroit Charter Fight

The question to be decided in the GDCF is whether Michigan’s Home Rule City Act allows the Detroit Charter Commission to send to Detroit’s voters a new charter over the objections of Governor Gretchen Whitmer. Opposed by Detroit’s mayor, the new charter was also “vetoed” by Governor Whitmer pursuant to MCL 117.22, a 1913 statute requiring new city charters to be presented to the governor so she can either “approve” them for a popular vote of city electors or “return” the charter with her “objections.”

Where does the “expressio unius” come in? MCL 117.22 contains two provisions for overruling gubernatorial vetoes of charter “amendments,” but it is silent about whether and how a gubernatorial veto of a charter “revision” can be overruled. (“Amendments,” in case you asked, are focused on single subjects, while “revisions” are comprehensive overhauls. No one doubts that this proposed charter comprehensively overhauls Detroit’s 2012 charter).

Does the “expressio” of two overruling procedures have any bearing on whether the Commission’s overruling of Governor Whitmer ’s veto is “exclusio”? To the surprise of no one who has ever taught statutory interpretation, both sides cite the “expressio unius” canon in favor of their position. Supporters of Governor Whitmer note that MCL 117.22 specifically provides two mechanisms for overruling a governor's veto with respect to amendments. First, for amendments enacted by the City Council, Council can overrule the Governor's veto with a 2/3 majority vote. Second, for amendments "proposed by initiatory petition" by Detroit electors, the amendment "shall be submitted to the electors notwithstanding such [gubernatorial] objections." Expressio two specific ways to overrule a governor's veto is exclusio any unwritten third way, right? Otherwise, as the brief of Lewis and White (Detroit voters opposed to the charter revision) puts it, "the last two sentences of Section 22 [would be] meaningless, because there would be no need to provide an exception if all revisions and amendments could make it on to the ballot notwithstanding the Governor’s objections" (page 17 of brief).

"Not so fast," say the supporters of the charter commission. Perhaps the enumeration of only one constraint on overruling governors' vetoes implies the exclusion of all other constraints! Only the City Council is obliged to come up with a super-majority to overrule the governor: "Initiatory plebiscites" can simply disregard the veto entirely. The state legislature plainly knew how to write in an obstacle to overruling a veto if they wanted to impose such an obstacle. The silence about charter revisions, therefore, should be construed as a grant of an implicit power of the commission to disregard the governor’s veto. (An amicus brief by "Legal Scholars" makes this clever riposte).

I might add another argument against inferring that there are only two ways to overrule a gubernatorial veto of a proposed revision from the enumeration of two ways to overrule a proposed amendment. The statutory enumeration of two mechanisms for overruling vetoes serves a useful function even if the Detroit commission can disregard Governor Whitmer's veto without such statutory authorization: Perhaps statutory silence preserves a local option for overruling vetoes of charter revisions that the explicit rules for overruling vetoes of amendments takes away. Those specific clauses on overruling vetoes of amendments, after all, preempt alternative local mechanisms contained in city charters. "Initiatory plebiscites", for instance, have a right guaranteed by state statute to be considered by voters, regardless of what Detroit's city charter might say to the contrary. The silence of the state statute regarding proposed charter revisions, by contrast, might reasonably be understood to leave the question of how gubernatorial vetoes should be overruled entirely up to the local charter governing the charter commission. On this reading, if the city's existing charter gives the charter commission the power to disregard the veto, then the commission can send a vetoed revision to the people for a vote. If the charter bars the commission from overruling the governor, then the governor’s veto stands. Note that such a reading does not render redundant the express means for overriding a veto enumerated in the state statute: Those enumerations limit local power to control vetoes, while the statutory silence constitutes a delegation to the city to decide how its charter will be changed.

There is a larger point underlying all of these technical arguments about whether or not lists in statutes are exhaustive: Expressio unius is actually a pretty weak canon. Federal courts, for instance, ignore it in favor of deferring to agency interpretations that treat statutory lists as merely illustrative. The reason for giving less weight to expressio unius is easy to see: In ordinary English, lists are often not intended to be exhaustive! As my co-bloger Ethan Leib has argued, sometimes legislatures redundantly list lots of stuff "belt and suspenders" style, just to be extra-sure they cover bases. Moreover, where there is a strong background norm in place, the enumeration of specific items is a pretty weak, indirect way to waive that norm. If Michigan law pervasively favors local power as argued by Justice Thomas Cooley in People v Hurlbut, then statutory silence should be filled by a presumption that the state legislature wanted to preserve local options. (No one believes, for instance, that my instructions to my children to stop kicking each other means that they are thereby authorized to punch each other: The background norm against intra-familial violence remains despite the specific reiteration of one example of violence, because the norm is so basic that no sensible person would think that the enumeration was intended to be exclusive).

All of those "expressio unius" arguments in the GDCF, therefore, essentially cancel each other out. The most one can say is that the state legislature had definite ideas about how to treat gubernatorial vetoes of charter amendments and left the text incomplete with respect to vetoes of charter revisions. And that gives rise to the central question of this blog post: Viz....

2. What priority should courts give to trans-statutory substantive canons as opposed to statute-specific purpose in resolving textual ambiguities?

Suppose you agree with me that MCL 117-22 is ambiguous as to the power of the Detroit charter commission to overrule Governor Whitmer. How should that textual ambiguity be resolved?

One simple solution would be to invoke Article 7, section 34 of the Michigan Constitution, which provides that "[t]he provisions of this constitution and law concerning counties, townships, cities and villages shall be liberally construed in their favor." On such a reading, the gap in MCL 117.22 regarding vetoes of proposed revisions should be filled with a presumption of local power: If the existing (2012) Detroit charter gives the Detroit charter commission the power to send a proposed revision to the people of Detroit for a vote despite a gubernatorial veto, then the commission should be able to disregard Governor Whitmer's veto and hold a referendum this August on its proposed revisions.

But the "liberal construction" required by Article 7, section 34 of the Michigan Constitution implies only that ambiguities in statutes should be construed favorably ("liberally") to local power. If the statute is not really ambiguous, then there is nothing to construe. Moreover, just because the bare semantics of statutory text are ambiguous does not mean that the statute lacks a determinate meaning. As even diehard textualists like John Manning note, textual ambiguities are traditionally resolved by looking to extra-textual purpose -- what Manning calls the statute's "policy context."

Viewed in its "policy context," MCL 117-22, enacted in 1909 and amended in 1913, looks like an effort by the state legislature to control Detroit with a hard constraint in the form of a gubernatorial veto that cannot be overruled. Consider two aspects of this "policy context" suggesting such a conclusion.

First, debates over home rule in the early twentieth century were inevitably colored by the Progressive fights over municipal ownership that inspired a backlash from state legislative conservatives who feared that municipal ownership would lead to "socialism" and Tammany-style corruption. From William Randolph Hearst's "Municipal Ownership League" in New York City to Mayor and Governor Hazen Pingree's effort to create a municipally owned streetcar system in Detroit, the cry of "home rule" was inevitably associated with municipal progressive's efforts to rid their cities of the power of utility magnates -- the titans who owned "traction" (streetcars), subways, waterworks, gas works, and other infrastructure essential for a modern city. Pingree, in particular, made his name in Detroit as champion of the three-cent fare for streetcars and as the nemesis of Detroit streetcar magnate Tom Johnson.

Pingree wanted to create a wholly city-owned network of streetcars to compete with Johnson's system, but Pingree was consistently foiled by conservatives who believed that such infrastructure was best left to private enterprise. In Michigan, this skepticism about municipal ownership took constitutional form in a state constitutional prohibition, dating from 1850, on the State's investing in "internal improvements." In 1899, then-Governor Pingree was defeated in his effort to create a streetcar commission for Detroit by the Michigan supreme court, which ruled in Attorney General v. Pingree that the McCleod Bill creating such a commission ran afoul of this 1850 ban on state-financed internal improvements. As Professor Charles A. Kent solemnly intoned in arguing for an injunction on Pingree's streetcar plans to the Michigan supreme court, "This is the first tangible movement in the direction of socialism."

From 1899 until the 1920s, Detroit politicians waged a protracted campaign to obtain "home rule" for the purpose of expanding municipal power over streetcars against private owners who relied on their street franchises and skepticism about municipal ownership to defeat such efforts. (The story of this epic struggle is nicely recounted by Neil Lehto in The Thirty-Year War: A History of Detroit's Streetcars, 1892–1922 (MSU Press 2017). Read in light of this battle over municipal ownership, the governor's veto over charter revisions, conferred in 1913, looks like a significant conservative check on municipal power. The Legal Scholars amicus brief argues that the Michigan Constitution is rooted in popular sovereignty and localism favoring Detroit's control over its own affairs. But the legal battles of the early twentieth century tell a much different story: There was deep skepticism from important political and legal leaders about Detroit's bid not only for municipal ownership but also for the power to take back lucrative street franchises controlled by the streetcar owners.

Such a conclusion from the general political context of the early 20th century is re-enforced by the specifics of how MCL 117.22 was amended in the wake of the 1910 mayoral campaign. That campaign largely revolved around the candidates' attitudes towards municipal home rule over streetcars: Popular opinion in Detroit ran strongly in favor of the 3-cent fare and control of the unpopular private street franchises. The 1909 version of home rule merely gave the Attorney General the power to issue a "recommendation" regarding proposed charter revisions. After the 1910 campaign, however, the state legislature amended this provision to create the current version of the statute, giving the governor a power to "return" revisions with "objections." Read in context of the grant of home-rule powers in the 1908 Michigan Constitution, it is difficult not to conclude that the state legislature was offering a quid pro quo: Strict limits on municipal expenditure imposed by the 1850 Constitution would be waived, but gubernatorial control would be substituted for constitutional control over Detroit's proposals for an expansive city-owned system of public infrastructure.

In short, read in light of its "policy context," MCL 117.22's gubernatorial veto looks like a hard constraint on municipal power, not a piece of gubernatorial advice that the Detroit charter commission is free to waive at will. After all, before 1913, there already was a system for the Attorney General's officials' providing mere "recommendations": If the new language in the 1913 law did nothing more with its references to "return[ing]" proposals with "objections," then what exactly was the point of the amendment?

3. So which should govern? Statute-specific purpose? Or the trans-statutory "liberal construction" canon?

The question, then, remains: Should the textual ambiguities in the 1913 statute be governed by its likely contemporary purpose or the "liberal construction" canon later incorporated into the 1963 Michigan Constitution?

I confess that I have no easy answer to this question. To my knowledge, the Michigan supreme court has never squarely confronted it. But here are a few relevant considerations.

In favor of allowing unwritten purpose to trump the trans-statutory "liberal construction" canon is the idea of fidelity to a legislative scheme. State lawmakers, after all, presumably have reasons for the laws that they enact. Those reasons might even form a coherent package of compromises that make sense as a matter of policy. Just because those reasons are unwritten does not mean that they should be ignored, especially if the text has gaps that can be filled only by looking to extra-textual sources. Ignoring those reasons, in fact, might make statutes less sensible and coherent than they otherwise would be. Why, for instance, would the state legislature confer on the highest elected official in the state the power to stop a proposed charter revision in its tracks with "objections" if the charter commission could blithely ignore those objections? The Legal Scholars' brief gamely argues that such a procedure could be "politics-forcing" in the sense that the governor's objections might carry weight with the voters even if they could be ignored by the commission. But this seems a little weak, given that the governor could make her objections known without any statutory mechanism for "return[ing]" the charter revision: What exactly does the formal mechanism of "objection" and "return" add to the normal political process of gubernatorial speechifying?

In favor of using trans-statutory canons to resolve disputes about textual ambiguities, however, is the idea that wading into the weeds of long-forgotten political debates is costly, both in terms of time and political legitimacy. It is significant that none of the briefs bother to delve into the debates over home rule in early 20th century Michigan. Reading historical monographs when one bills by the hour is a costly way to resolve constitutional disputes. (For a law professor, of course, reading monographs is not a bug but a feature of legal wrangling!)

Quite apart from the sheer effort it takes to figure out a law's policy context, there is also a cost in terms of judicial legitimacy. Those unwritten statutory purposes, after all, are often hotly contested values. Taking a position on which values the legislature implicitly endorsed is a difficult thing for a judge to do who legitimately pursues a stance of apolitical neutrality. It is one thing to parse grammar, look up dictionary definitions, and invoke expressio unius. It is another thing altogether to conclude that a state statute is an expression of distrust for profligate urban progressives: Those progressives might reasonably take offense at such a conclusion.

The great advantage of trans-statutory substantive canons is that, because they apply the same way across a broad range of disputes, they theoretically absolve judges of taking stances on hotly contested values. Favoring local power in the GDCF will favor Progressive Democrats in Detroit, because the proposed charter is filled with provisions favored by the "People's Slate" that dominated the Commission. But in another dispute where a local government adopts a conservative stance -- say, creating "Second Amendment sanctuary cities" favoring deregulation of firearms -- the "liberal construction" canon will favor the Right. Like analogous substantive canons such as Chevron or the Santa Fe Elevator v. Rice anti-preemption canon, the ideological valence of the "liberal construction" canon is hard to predict across cases.

In theory, the same might be said for using statute-specific purposes to resolve disputes about statutory meaning. In practice, however, figuring out unwritten statute-specific purposes leaves judges with a lot more latitude to favor the judge's own values than applying a crude, simple canon like one favoring local power. After all, as textualists are fond of reminding us, each statute has lots of unwritten purposes, and favoring one of them above the rest is a great way for a judge to favor his or her pet political cause in each statute, reading narrowly those statutes that the judges disfavors and giving a generous reading to the judge-favored purposes in the judge-favored laws.

I do not hold a brief for either side in this dispute. Here's the simple tradeoff in summary. Delving into statutory purpose when the text is indeterminate probably leads to more coherent readings of statutes at the risk of entangling judges in angry ideological fights. Clear and simple canons like "liberal construction" are like meat cleavers: They do not necessarily make policy sense, but they are easy to wield. In a polarized nation, each side can be satisfied that the simple rule was fairly applied in a particular case, and each side can also rest easy that the rule could favor their own position in a future case even if they lose the present case.

Whichever stance the Michigan supreme court ultimately takes in the GDCF, it would be nice if they admitted that they are obliged to take a position. The temptation to declare that the text of MCL 117.22 is "plain" by involving one or another version of "expressio unius" will be great, but yielding to it is disingenuous. Better to come clean that sometimes the text is up for grabs and that the court needs some way to deal with these "jump balls." In this respect, the GDCF is, in microcosm, what statutory interpretation is all about.