Things Fall Apart
By 
I have a piece at the Washington Post about the Supreme Court and the post-1965 norm of strategic retirements. We’ve been through the fact that most voluntary retirements since then have been strategic before, but a couple of other points are relevant in terms of where the nomination process is headed. The first is that prior to the Civil Rights Act-generated realignment, party coalitions were too loose to make strategic retirements a very meaningful concept:
The Supreme Court is of course not an apolitical institution, and justices are hardly naive about the consequences of nominating replacements who would repudiate their own ideological vision. Chief Justice (and former president) William Howard Taft, for example, refused to resign until President Herbert Hoover agreed to nominate Charles Evans Hughes rather than Harlan Fiske Stone (a more liberal justice who was ultimately promoted to chief justice by Franklin Roosevelt).
Still, for the most part, strategic resignations were not an issue for much of the 20th century, for the simple reason that the loose nature of partisan coalitions and the lesser degree of elite polarization made timing a resignation to ensure an acceptable successor difficult, if not impossible.
For example, Woodrow Wilson nominated both the great liberal Louis Brandeis and the grossly bigoted and intellectually undistinguished James McReynolds. FDR nominated William O. Douglas, who has arguably the most liberal voting record of any justice in the history of the court, but also Jimmy Byrnes, a fairly standard-issue Southern segregationist. For his part, Republican president Dwight Eisenhower nominated the heart of the liberal court of the 1960s, Earl Warren and William Brennan. With no clear ideological pattern in nominations, there was little reason for justices to put much weight on who was in the White House when deciding whether and when to leave.
It was Warren himself whose resignation marked the new partisan era that began with the Republican nomination of Barry Goldwater for president and the full Democratic embrace of civil rights in 1964. The chief justice wanted to protect the legacy of his court and despised fellow Californian Richard Nixon, so he resigned in 1968 to allow Lyndon Johnson (the first Democratic president to consistently name liberals to the court) to choose his successor. However, Warren’s strategic retirement was foiled, as a conservative coalition of Republicans and southern Democrats who wanted revenge on the Warren court they hated teamed up to filibuster the nomination of Abe Fortas to be chief justice, keeping him from the position. This led to Nixon getting not one but two immediate nominations when Fortas was forced to resign because of corruption charges.
The second point is that life-tenured appointments are bad:
Whatever one thinks of these decisions, it is clear that justices thinking strategically about retirement is common. (Indeed, even in the case of Ginsburg and Breyer, it’s not entirely absent; does anyone think they would resign with Trump in the White House unless medical reasons forced their hand?) It should also be clear that this is not a good system. Justices should serve fixed, nonrenewable terms, rather than having to make a decision about when the most appropriate time to step down should be; some presidents should not get a disproportionate impact on the constitutional development of the country through the vagaries of chance. In 2013 and 2014, Ginsburg and Breyer were faced with a dilemma they never should have had to deal with in the first place. And now that they’ve made their choice, their fellow liberals will have to wait anxiously, one health scare at a time, until the political ground shifts again.
Indeed both life tenure (a product of an context in which convincing qualified candidates to do a job with little power of status to take the job at all was more of a concern than justices staying on too long) and the advice and consent process (a product of the framers being in denial about parties, and a supermajority veto of presidential nominations would be much better) are bad rules. But they’ve worked out OK historically in large measure because of relatively low levels of polarization and ideologically heterogenous parties. But in the current partisan context the two rules will combine to produce results like, say, “the Senate refusing to consider appellate judicial nominees from an opposition president and then a president not even elected with plurality support entrenching his party with a long-term Supreme Court majority.” Such results are not going to be conducive to the long-term legitimacy of the institution.