Rick (Pildes) has a nice post thinking through whether a First Amendment theory of partisan gerrymandering renders the practice more like racial gerrymandering or racial vote dilution. I just want to add a few thoughts to Rick’s discussion:

1. There’s no reason why a First Amendment theory of partisan gerrymandering should be limited to a single district, even analogizing the theory to racial gerrymandering. If, as Rick says, the crux of the claim is “tak[ing] political affiliation into account in sorting voters into districts” and “mov[ing] groups into or out of districts based on their political viewpoints,” this is plainly a governmental action that can be—and generally is—taken with respect to a district map as a whole. Take the Maryland plan at issue in Benisek. It’s only the Sixth District whose winner flipped because of voters being sorted into districts based on their political affiliations. But voters were equally moved based on their political views into and out of every other district. The mapmakers just didn’t intend to flip any of those districts, because they were satisfied with how the districts were already performing (namely electing six Democrats and one Republican).

2. Whether labeled a First Amendment theory or something else, any approach to partisan gerrymandering that imposes liability based only on the politically motivated sorting of voters faces significant doctrinal obstacles. In Bandemer, the Court held that discriminatory effect is a necessary element of the cause of action. In Vieth, a majority of the Court (including Justice Kennedy) rejected Justice Stevens’s predominant-intent standard, which was borrowed from the Court’s racial gerrymandering cases and, like them, would have applied to individual districts. In LULAC, a majority of the Court (with Justice Kennedy writing) rejected the appellants’ sole-intent standard for district plans in their entirety. And since LULAC, the Court has summarily affirmed a number of lower court decisions declining to invalidate maps on First Amendment grounds. Obviously, these obstacles aren’t insurmountable for the Court, which is free to treat its precedents however it wishes. But they still represent major stumbling blocks for litigants and lower courts.

3. While there’s a clear distinction between intent-only racial gerrymandering claims and effect-only racial vote dilution claims, this divide doesn’t translate neatly into the partisan gerrymandering context. Even if the First Amendment theory is intent-only, the alternative approach, advocated in the North Carolina and Wisconsin cases, isn’t effect-only. Rather, it requires discriminatory intent and discriminatory effect and no legitimate justification for this effect. The choice is therefore between intent alone and intent plus other elements—in other words, between a theory that’s easier to satisfy (intent-only) and one that’s more difficult (intent-plus).

4. There’s reason to think that an intent-only First Amendment standard could be satisfied in quite a large number of places. In the current cycle, 14 out of 24 congressional plans with at least seven seats, and 21 out of 40 state house plans for which data is available, were enacted by a single party in full control of the redistricting process. The vast majority of these maps would presumably be in jeopardy if partisan intent alone were enough to establish liability. (If First Amendment claims have to proceed district by district, the number of plausible lawsuits would be even larger.) This isn’t necessarily a feature or a bug; it’s just a fact that partisan intent is usually present when a single party has the power to design districts.

5. As Rick notes, the logical remedy if a map (or district) is struck down on an intent-only theory is a map (or district) redrawn without taking into account electoral data. I worry that this may be quite a toothless remedy. It’s obviously very hard to stop mapmakers from considering election results. There’s also no way to prevent them from taking advantage of their own, informal, but often extensive, knowledge of how different areas vote. In Maryland, for example, I have little doubt that line-drawers could come up with a new Sixth District, without looking at any electoral data, that’s still highly likely to elect a Democrat. In that case, I wonder if the game is worth the candle. If all the time and effort of litigation yield a plan that performs exactly the same as its predecessor, only this time (supposedly) cleansed of the taint of partisan intent, is it even worth suing in the first place?