I agree that the sheer power and reach of SCOTUS today, and increased partisanship in practice (even though everyone pretends it's a non-partisan body), does necessitate some reform. But electing judges makes no sense - you might as well then just give the fullness of power to Congress, a la UK's parliamentary sovereignty.
What I think we should do is revisit what exactly SCOTUS does, and why. Right now they basically have the final say in any question of constitutionality, and the outcomes are either "it's constitutional" or "it's unconstitutional". I think that's wrong - the third possible outcome should be "Constitution is ambiguous on this". Currently this gets folded into one of the other options, depending on the majority of the court, but I think it's a poor model - if Constitution really is ambiguous, I don't want a simple majority of a few unelected people, many of whom are quite partisan, to make that decision.
Instead, I think this option (ambiguity) should be explicit. The way it would work is something like this - if the court decision is unanimous (or maybe with at most one dissenter) one way or the other, then it's assumed that the Constitution is really unambiguous on the subject, and that's the ruling - same as now.
But if you get a bigger split, then the ruling is automatically "ambiguous". At that point all the disagreeing parties on the court should have to sit down and write a short opinion on what changes to the Constitution they would require to make the other side's opinion unambiguously correct (if there are more than two sides - which can be the case if different judges rule the same way for different and unrelated reasons - then such opinions should be written for all parties other than the one in question).
Then, those opinions are automatically submitted as proposed constitutional amendments to the states for ratification, per usual procedure, except that each state can only ratify one at a time, and there's a reasonable time limit. If one of the amendments wins, then (since all judges have already stated under oath that this is what is required to remove any ambiguity) the ruling is in favor of the corresponding opinion.
If none of the amendments get the requisite majority of state ratifications, then court decides based on simple majority, just like today - but the resulting decision is not considered binding precedent, and only applies to that one case. If the same ambiguity arises in future cases, the process has to be repeated.
Ideally, this should be combined with a lower bar for constitutional amendments - 3/4 of states is really quite ridiculous, given the sheer number of them, and population differences. Something like 2/3 would be more sensible. Although ideally it should incorporate direct popular vote in a referendum as well, in a series of cascading vetoes to check each other - e.g. 2/3 of popular vote is enough to amend, but a simple majority of states can veto that, but 3/4 of popular vote can override the veto.
This is the part of the system which makes me scratch my head.
The constitution is the source of where the courts derive their power. Being able to change the level of power you have seems to be against the constitutions purpose of defining, limiting, and binding the Govt.
Article III, Section 2, Clause 1 of the Constitution states:
The judicial Power shall extend to all Cases, in Law and Equity, arising under this Constitution,
That kind of parsimonious interpretation has no legal bearing. A dispute over the meaning of the Constitution would naturally arise under [the laws of] the Constitution.
It should also be noted that it was long-standing British common law that courts ruled on the interpretation of law, and that there was ample precedent in the US revolutionary period of state Supreme Courts voiding state laws under state constitutions. Virtually every reference to the notion of questions of constitutionality pre-Marbury v Madison accepts that the judicial courts would play a role in this regard. The only extent to which the decision would have been surprising would have been in arguing whose opinion won it in the case of conflicts. (Note that nullification crises continued up to the Civil War).
What I think we should do is revisit what exactly SCOTUS does, and why. Right now they basically have the final say in any question of constitutionality, and the outcomes are either "it's constitutional" or "it's unconstitutional". I think that's wrong - the third possible outcome should be "Constitution is ambiguous on this". Currently this gets folded into one of the other options, depending on the majority of the court, but I think it's a poor model - if Constitution really is ambiguous, I don't want a simple majority of a few unelected people, many of whom are quite partisan, to make that decision.
Instead, I think this option (ambiguity) should be explicit. The way it would work is something like this - if the court decision is unanimous (or maybe with at most one dissenter) one way or the other, then it's assumed that the Constitution is really unambiguous on the subject, and that's the ruling - same as now.
But if you get a bigger split, then the ruling is automatically "ambiguous". At that point all the disagreeing parties on the court should have to sit down and write a short opinion on what changes to the Constitution they would require to make the other side's opinion unambiguously correct (if there are more than two sides - which can be the case if different judges rule the same way for different and unrelated reasons - then such opinions should be written for all parties other than the one in question).
Then, those opinions are automatically submitted as proposed constitutional amendments to the states for ratification, per usual procedure, except that each state can only ratify one at a time, and there's a reasonable time limit. If one of the amendments wins, then (since all judges have already stated under oath that this is what is required to remove any ambiguity) the ruling is in favor of the corresponding opinion.
If none of the amendments get the requisite majority of state ratifications, then court decides based on simple majority, just like today - but the resulting decision is not considered binding precedent, and only applies to that one case. If the same ambiguity arises in future cases, the process has to be repeated.
Ideally, this should be combined with a lower bar for constitutional amendments - 3/4 of states is really quite ridiculous, given the sheer number of them, and population differences. Something like 2/3 would be more sensible. Although ideally it should incorporate direct popular vote in a referendum as well, in a series of cascading vetoes to check each other - e.g. 2/3 of popular vote is enough to amend, but a simple majority of states can veto that, but 3/4 of popular vote can override the veto.