Alternatives to Eidelberg's election scheme

Eidelberg's proposal, in combination with some form of élite "senate" or upper legislative chamber, would be a method of choosing a chief executive who was both independent from the legislative branch and likely to be competent. It would accomplish this by relying on the senate to identify qualified presidential candidates, but making the successful candidate dependent on popular support to remain in office. The familiar directly-elected president, on the other hand, though he is independent from the legislature, is potentially abysmally unqualified, being chosen by voters who have no direct knowledge of the candidates, and in assessing them are forced to depend on media reports. The whole tenor of the political system then seems to hinge on the news media's political competence and sense of public responsibility.

Notwithstanding the indispensable role of the legislative branch, the choice of the chief executive is, as I think Eidelberg somewhere observes, the most crucial decision the political system must arrive at. Theoretically speaking, his proposal thus seems extremely important.

There are two or three other possible ways that I can see of achieving the same combination of competence and independence from the legislative branch. (Again, one assumes that there exists a senate capable of identifying qualified presidential candidates.) These would put the election of the president squarely in the hands of the senate, or of another élite body, removing the general public from the picture. The possible approaches are as follows:

• (a) Make the president independent from all other agents by giving him a single term of fixed length, or, in the extreme case, the life tenure of an elective monarch. The senate then elects the president, but has no further control over him once he has taken office.
• (b) Keep the president formally dependent on the senate for his continuation in office, but require some supermajority (e.g., two thirds or three quarters) of senate votes, rather than the usual simple majority, to remove him, even upon the periodic conclusion of his term of office. The larger the supermajority threshold, the more difficult it will be to depose the president; so at some ill-defined point, he should be adequately independent from the senate. (Recall that Eidelberg agrees that the executive should not be entirely free to ignore the views of the senate.)
• (c) Return to the general approach attempted by the framers of the American constitution. Set up twin élite institutions, an electoral college and a senate, one of which chooses the president while the other constitutes the upper house of the legislature. The president is then entirely independent of the senate. (On the other hand he will tend to be dependent on the electoral college instead; but if this resembles the American one, which comes into temporary existence only at four-yearly intervals, it is unlikely to seek more powers for itself at the expense of the president's.) As already pointed out, the American-style electoral college failed in its purpose, never achieving even minimal independence from the popular electorate; but perhaps its design could somehow be corrected.

Are any of these possibilities likely to prove preferable to Eidelberg's proposal?

In both (a) and (b), one would rely on the upper legislative chamber to choose the president, probably at intervals of several years. A major problem associated with this is that it would concentrate ordinary voters' attention on the likely presidential choices of each candidate for a senate position. It seems quite possible that as a result the senators chosen by those voters would lose their independence from the president, or from the parties led by the main presidential candidates. This would be similar to the fate of the American electoral college, and is also essentially what has happened to the House of Commons in Britain, Canada, and other places with similar constitutions. In each case the ordinary voters are so interested in the choice of chief executive that they are willing to vote for a legislative representative based principally on which candidate for chief executive he is pledged to support, or in other words what tight-knit party he belongs to. The senate would then probably still retain more of real existence than the American Electoral College, because of its important legislative functions; but it would probably be less independent than, say, the current American Senate. By contrast, if the senate merely had a nominating role in the presidential election, as in Eidelberg's proposal, ordinary voters would almost certainly not be sufficiently interested in the nomination decision to base their choice of senator on any pledge on his part to support some particular nominee. Under that proposal, the senators would remain basically legislators in the eyes of the ordinary voters.

Supposing that the senate could somehow be preserved from the control of tight-knit political parties, there is also the problem that it becomes very difficult, under proposals (a) or (b), to remove an unsatisfactory president. An élite body working more or less closely with the president might acquire both personal and policy loyalties that prevented it from coldly removing him from power when necessary—loyalties which the general population, being more detached from the political scene, would not develop. In any event, under proposal (b), it is necessary for the senate's supermajoritarian power to remove the president to be exercised only on rare occasions, if his independence from the senate is to be preserved. Such removal will resemble an impeachment more than an ordinary election.

Consequently, in most cases, the president would have to serve out his maximum term—a result which would not differ much from proposal (a), with its single fixed term of office. As a result, one would be faced with the alternative of setting a relatively short maximum term of office, which would be wasteful of the scarce supply of statesmanship, or of accepting a presidency of long duration, regardless of popular or even senatorial preferences. This latter kind of president would tend to resemble an elective monarch more than a republican chief executive. In order to avoid this prospect, one would need some way of preserving an independent electoral college, as in (c) above, which would be able to depose an unsatisfactory president in relatively short order.

While I think popular opinion might accept the existence of an élite senate with limited powers, under modern conditions it would be unlikely to tolerate a symbolically and actually powerful president who was effectively irremovable for a period of more than eight years or so. And of course, realistically speaking, for only the senate to have the right to choose a new president, without any participation in the decision by the ordinary voters, would also be highly illegitimate in most people's eyes. Eidelberg's proposal, on the other hand, both provides a reasonable way of removing presidents once their incompetence becomes apparent and gives the president the legitimacy of being approved (in some sense) by the popular electorate.