Michael Greve on Preemption
At Liberty Law Blog, Michael Greve has an interesting post objecting that I (and Justice Thomas) have too narrow a view of preemption. In particular, he says, a textualist approach to preemption, reflected in Justice Thomas’ concurring opinion in Wyeth v. Levine, conflicts with Chief Justice Marshall’s approach in such great cases as Gibbons v. Ogden and McCulloch v. Maryland.
To restate, the view I expressed is that the conflict necessary for a federal law to displace a state law under Article VI's supremacy clause should arise from the federal statute's text, not from a judicial guess as to the statute’s purpose. That arises partly from Article VI itself, which gives supreme status to what was enacted, not what may have been intended. And as a practical matter, statutes often don’t have a single readily-determinable purpose beyond their text; statutes often reflect compromises between competing goals -- or, put differently, an accommodation of various interests. If a court picks one of those interests as the overriding “purpose” and treats it as if it had been enacted, the court creates a statute very different from the one actually passed. On the other hand, if a conflicting federal purpose does follow unmistakably from the text, the state law should be preempted – this isn’t a call for a clear statement rule or particular magic words. (Crosby v. National Foreign Trade Council is a good example of the conflicting purpose being manifest in the text).
Is this view contrary to Gibbons and McCulloch, and indeed all of Marshall’s jurisprudence, as Professor Greve extravagantly claims? I don’t think so. Let’s start with Gibbons. The question there was whether New York’s steamboat monopoly prevented a ship with a federal coasting license from operating between New York and New Jersey. Marshall found the state law preempted. Here is the core of the preemption discussion (pp. 213-15) – to me it seems well anchored in the statute’s text:
The first section [of the federal statute] declares that vessels enrolled by virtue of a previous law, and certain other vessels enrolled as described in that act, and having a license in force, as is by the act required, "and no others, shall be deemed ships or vessels of the United States, entitled to the privileges of ships or vessels employed in the coasting trade."
This section seems to the Court to contain a positive enactment that the vessels it describes shall be entitled to the privileges of ships or vessels employed in the coasting trade. These privileges cannot be separated from the trade and cannot be enjoyed unless the trade may be prosecuted. The grant of the privilege is an idle, empty form, conveying nothing, unless it convey the right to which the privilege is attached and in the exercise of which its whole value consists. To construe these words otherwise than as entitling the ships or vessels described to carry on the coasting trade would be, we think, to disregard the apparent intent of the act.
The fourth section directs the proper officer to grant to a vessel qualified to receive it, "a license for carrying on the coasting trade," and prescribes its form. After reciting the compliance of the applicant with the previous requisites of the law, the operative words of the instrument are,
"license is hereby granted for the said steamboat Bellona to be employed in carrying on the coasting trade for one year from the date hereof, and no longer."
The word "license" means permission or authority, and a license to do any particular thing is a permission or authority to do that thing, and if granted by a person having power to grant it, transfers to the grantee the right to do whatever it purports to authorize. It certainly transfers to him all the right which the grantor can transfer, to do what is within the terms of the license. …
Notwithstanding the decided language of the license, it has also been maintained that it gives no right to trade, and that its sole purpose is to confer the American character.
The answer given to this argument that the American character is conferred by the enrollment, and not by the license, is, we think, founded too clearly in the words of the law to require the support of any additional observations. The enrollment of vessels designed for the coasting trade corresponds precisely with the registration of vessels designed for the foreign trade, and requires every circumstance which can constitute the American character. The license can be granted only to vessels already enrolled, if they be of the burthen of twenty tons and upwards, and requires no circumstance essential to the American character. The object of the license, then, cannot be to ascertain the character of the vessel, but to do what it professes to do -- that is, to give permission to a vessel already proved by her enrollment to be American, to carry on the coasting trade.
It's true that there may be another way to read the statute – that the license signifies only federal approval, subject to additional state requirements as applicable. I think that’s a strained reading, as the statute says federally licensed ships “shall be … entitled to the privileges” of coastal trading, which sounds like an affirmative right to trade, not a mere conditional approval. But in any event Marshall said the preemption came from the text, not from an extra-textual exploration into purpose: “The word ‘license’ means permission or authority” and a “license … transfers to the grantee the right to do whatever it purports to authorize.”
McCulloch is more difficult. There the question was whether the state could tax a federal entity (the Bank of the United States). Even without an express statement, Congress’ creation of the Bank unmistakably carries with it a direction that the state not prevent the Bank from operating – as, for example, by imposing a discriminatory and confiscatory tax. That was arguably the situation in the case, as Maryland demanded $15,000 – quite a sum at the time – to operate, and state banks were not charged that amount.
But Marshall went further, saying that all state taxes of the Bank (and apparently all state taxes of federal entities) were preempted because “the power to tax is the power to destroy.” (He may have meant this as a constitutional rule – the opinion is unclear – but I’ll treat it as a preemption holding). This result seems dubious as a textual matter – it does not automatically follow from any particular text, and indeed is an unfortunate example of adjudication by aphorism.
We can easily imagine that if Congress had proposed an express statutory tax exemption for the Bank, the states would have resisted with enough political force to block it. (Especially because at that time Senators were appointed by the states). The law as actually enacted might well be a compromise between the national interest in an unconstrained Bank and the state interest in taxing entities doing business in the state. For the Court to then say that even a reasonable non-discriminatory tax would be preempted would award the nationalists a victory they couldn’t win in Congress. And it would preempt the state law due to a conflict with something that wasn’t (and as a practical matter wouldn’t have been) enacted – contrary to Article VI.
Moreover, allowing some state taxes would hardly be a disaster – Congress could pass an exemption if it wanted to, and if the danger to the Bank from state taxes was real Congress would surely do so quickly and easily. So although McCulloch may have reached the right result on the particular facts, I agree that Marshall’s reasoning isn’t textualist enough for my taste.
That doesn’t worry me. We shouldn’t make a fetish of Marshall. He was a great man and a great Chief Justice, to be sure, and much closer (even in 1819) to the thinking of the founding era than we are. But that doesn’t mean he didn’t make mistakes, or allow his own position and impulses (which were moderately nationalist) to govern his reasoning at times. (For some delightful gentle-yet-devastating criticism, see David Currie’s classic The Constitution in the Supreme Court: The First Hundred Years).
At the same time, I think Greve grossly overstates the extent to which Marshall’s reasoning in general was non-textual (even saying that he "can’t think of a single Marshall opinion" likely to pass textualist muster). In addition to Gibbons, how about Hodgson v. Bowerbank (Article III’s grant of alienage jurisdiction means what it says) or Hepburn v. Ellzey (District of Columbia isn’t a “state” for Article III purposes)? Or, for that matter, Marbury, which rests in significant part on the language of Article III and Article VI. More importantly, I don’t think any of the cases Greve mentions have fatal textualist objections – Brown v. Maryland, or Justice Story’s opinions in Martin v. Hunter’s Lessee or Swift v. Tyson (though the latter is quite a complicated question). Although these opinions may discuss how their results are consistent with the enactors’ likely purposes – surely a legitimate point to make – at bottom they all invoke plausible interpretations of specific text.
I'll conclude with a note on Wyeth v. Levine, the modern preemption case Professor Greve criticizes in his post and in his important and insightful book The Upside-down Constitution. The question there was whether the FDA's labeling requirements for a prescription drug preempted additional labelling requirements at the state level. I don't have a strong view on the outcome, although it does seem that (as the majority opinion said) there was some ambiguity about whether the FDA-approved label was a minimum requirement or (as in Gibbons) a license. But I don't think it's a big deal if the Court erred slightly in the state's favor. It would surely not be a problem for the FDA to clarify that it intended a maximum as well as a minimum. And national entities are much more likely to correct errors in that direction than the other, so it makes sense to hold them to their text, and let their broader purposes, if there are any, be added through the political process rather than the courts.