As we approach Independence Day, I tend to listen to my version of patriotic songs.  One of those songs is a Crosby, Stills, Nash, and Young track that they labeled as “49 Bye-Byes/America’s Children” (basically a modified live version of the Buffalo Springfield song “For What It Is Worth” a/k/a “There’s Something Happening Here” with an intro.)  Thursday’s Supreme Court Opinions brought to mind that intro — “Forty-nine reasons all in a line.  All of them good ones.  All of them lies.”

The most prominent example of this concept is the decision in the Commerce Department case concerning the census question.  The ultimate decision in this case was a 4-1-4 split.  On the one side, you had the four most conservative members of the Supreme Court.  These four did not care to look at the truthfulness of the reasons given for asking a citizenship question on the 2020 census.  It was enough that they thought that it was possible to make an argument with a sort of straight face that those reasons would justify adding the question to the census.  The four most liberal members thought that it was clear that the Administration had a solution designed to achieve a political end and went searching for a problem that would justify their proposal.  The deciding vote was Chief Justice Roberts who continued a history of occasionally throwing liberals a bone while pushing a conservative agenda.  Because he had the four conservative justices joining him on the conservative part of the opinion and the liberal justices joining him on the liberal part of the opinion, the Chief Justice’s opinion is an opinion of the “Court” (i.e. it had five votes, so it is precedent for lower courts).

The Chief Justice’s opinion first gives this Administration broad authority to act.  It finds few if any limits on the Census Bureau’s ability to place questions on the census in the statutes governing the census.  It also recognizes that the Census Bureau is entitled to use the census to obtain whatever information the Census Bureau (or the Commerce Secretary) deems is useful.  It also gives broad authority to the Commerce Secretary to determine what weight to put on the information developed by the experts in evaluating the potential questions.  So even though the best characterization of the record is that the question is likely to interfere with the primary goal of the census (an accurate enumeration of all people in the United States), the Commerce Secretary has the power to disregard that evidence as “speculative.” 

However, here is where the Chief Justice decides that this fight is not worth it for the long-term conservative agenda.  It is clear from the record that the desire to place this question on the census has nothing to do with enforcing the Voting Rights Act.  Instead, it is about the desire to use citizenship data to alter the mandates of the Voting Rights Act to serve a political end.  While the opinion implies that this goal might be a valid reason for putting the question on the 2020 census, the Chief Justice believes that administrative law requires that the executive branch be transparent about the reason for an administrative act.   Put in simpler and less polite language, the Commerce Secretary is a liar.  If the Commerce Secretary wants to add this question for political reasons and thinks that the benefits of having this information for states that want to consider redistricting using the “citizen only” numbers outweighs the risks of an inaccurate census, he is free to do so.  He just needs to bite the bullet and accept the consequence of such a decision; so that courts can review whether the record supports his actual decision.    

It is unclear exactly when the Census Bureau must have its final questionnaire ready to go to the printer.  This decision — as soon as implemented by the lower courts — does kick the matter back to the Commerce Department which could readopt the question with a more honest statement of the reasons for wanting the question.  Given the issues about whether the real purpose is to discourage minority participation, one would hope that somebody would tell Secretary Ross and President Trump that this issue is not worth the fight at this time.  But we are likely looking at several more months of litigation.  (And whether Chief Justice Roberts is going to want the Supreme Court to get involved again is debateable.)

The other example is the dissent in the partisan gerrymandering cases.  The majority, again by Chief Justice Roberts, finds that there is simply no measure of when a redistricting plan is a partisan gerrymander and, thus, the courts should treat whether a plan is impermissibly the product of a desire to maintain an unfair partisan advantage is a “political question” — in other words, something for the elected branches rather than the federal courts to decide.  The dissent written by Justice Kagan dismantles the majority opinion at length, demonstrating politely and in appropriate legal terms that the majority is either intentionally misstating things or ignoring important facts.  Put simply, there are a substantial number of “neutral” ways to measure the validity of a map.

While not mentioned in the dissent, every state uses “first-past-the-post” to determine the winner of each district (and presumably no court would allow a state to declare that party X only needed 45% of the vote to win a district but that party Y needed 55% of the vote to win a district).  It follows that the party that gets the most votes state-wide should win the most seats.  It is possible to use past votes to predict the “average” vote in a district (and redistricting is based on this prediction).  By requiring the “average vote” in the median district to be close to the average state-wide vote, the worst versions of partisan gerrymandering would be impermissible.  And courts already do something similar in requiring equal-sized districts — setting a simple numerical criteria and stating at what point of deviation a state will need to provide some justification for the deviation beyond the desires of the legislature.  The ultimate test will probably need to be more complex, but taking the votes for state-wide offices (including President and U.S. Senate) in the last three election cycles before redistricting, it is possible to get a partisan lean of each precinct, each newly drawn district, and the state as a whole.  The median district should be the “bellweather” district with a partisan lean within 0.5% of the state as a whole (if not closer).  Each state would be free to choose whether it wanted a lot of districts near the median district (to flip a large number of seats as the majority changes) or few (so that the legislature is mostly stable and a tiny change does not result in a large number of seats going the opposite way), but 50.1% of the state-wide vote would give each party a fighting chance at being in the majority (subject, of course, to possibility that strong candidates in the swing districts could make a difference).

Justice Kagan emphasizes other criteria — mostly based on state law.  Almost every state has some traditional districting criteria that are considered in analyzing districts under the Voting Rights Act.  For example, many states seek compact districts (more square-like and less amoeba-like), contiguous districts, following existing (county, city, precinct) boundaries, and respecting major “natural” boundaries (rivers, highways, etc.).  With modern technology, it is possible to input those criteria into a mapping program (and leave out any “partisan” preferences),  Those mapping programs could then generate thousands of maps.  Those maps should cluster (i.e. the mode, median, and mean should be in close proximity) around a natural partisan breakdown given the partisan composition and the geography of where each party’s voters live.  While the map-adopters are free to choose which particular map they want, the partisan breakdown of the map should be in close proximity (choose a measure) of that natural breakdown.  The bottom line, however, is that it is possible to determine what a “neutral” plan would look like and measure how far the actual plans deviate from that neutral plan.  Then taking into account other evidence showing that the deviation was intentional, courts can apply traditional tests to show that partisan motives predominated.  We already have five states (four in federal court, one in state courts) where courts were able to apply traditional legal principles to determine that a map was distorted for partisan advantage. 

The majority, however, wanted to pretend that these facts were not well-established and that there were no neutral legal principles out there that would not result in courts deciding the optimal distribution of seats based on the court’s own preferences.  The majority also grossly exaggerated the ease by which voters could seize control back and fix the problem. 

While it will not be easy, the Chief Justice has, unintentionally, given us our marching orders for 2020.  If the Bush-Trump majority on the Supreme Court will not protect voters, then voters must protect themselves.  In states that allow referendums, we need to vote for well-designed plans to place non-partisan actors in control of redistricting with appropriate criteria to assure fair plans.  In states that do not allow referendums, we need to work to elect Democratic governors, state senators, and state representatives to prevent Republicans from drawing unfair plans.  And we need a Democratic President and a Democratic Congress to pass national laws regulating the redistricting process so that, next time, there are clearly established laws for the federal courts to enforce.