The Washington Times

Last week’s election, the midterm of a first presidential term, was “supposed” to produce gains for the party opposing the White House. That was also “supposed” to happen in 1998, but didn’t. In 2000, Al Gore was “supposed” to win big thanks to the strength of the economy, according to every political science model on the market. Maybe the time has come to throw out all of the popular generalizations about elections.

In fact, while we are rethinking, maybe we should rethink George Santayana’s famous axiom that “those who cannot remember the past are condemned to repeat it.” I propose: Those who spend too much time remembering the past are condemned to think the present is repeating it.

I’m not going to push this proposition to the limit, which would be to see the world as through a nihilistic fog in which nothing has anything to do with anything else – to conclude that no guidance based on an understanding of the past can be presumed to have any validity in terms of assessing the present. [Santayana also said, “History is a pack of lies about events that never happened told by people who weren’t there.”]

But really, the bold conclusions drawn about patterns in elections are wildly out of proportion to what is, in fact, the very small body of empirical evidence we have to work with. When does the “modern” political period begin, meaning, how far back can we look before we reach the conclusion that circumstances then were so different as to make the inclusion of data from that time more misleading than useful? Not so long, I think. The postwar period? Perhaps FDR’s time? But even playing it fairly safe, the particularities sometimes overwhelm the generalities. If you approach the question of California gubernatorial politics with a lens that looks back to Ronald Reagan’s victories, you are spanning vastly different state demographic characteristics.

The number of first-term midterm elections since FDR is equal to the number of presidents since FDR [although you get to the interesting question of how to deal with those who assumed office before the midterm of their predecessor, as in the case of Harry Truman and Gerald Ford; and do you count LBJ’s first term as his first elected term, or was that a second-term midterm?]. That’s 11 or 12, depending on how you reckon.

That’s not much to work with. If you’re in charge of the party, do you really want to base your political strategy on some supposed inevitability based on a universe this small? Second-term midterms? Can you count Mr. Nixon, who resigned before the election? Or is that Gerald Ford’s first [and last] midterm? In any case, the answer here is five to seven. Yet Republicans, on the basis of this underwhelming data set, thought that victory would be theirs in the House in 1998.

And about Mr. Nixon and the Republicans in the 1974 election: Did that election really have anything to do with “historical trends”? A pretty plausible case could be made that it had more to do with the particular history that had just unfolded – a president resigning in disgrace under the cloud of impeachment. I’m prepared to say that the next time a president resigns in disgrace before his second midterm, I will expect his party to do badly. But I’m not prepared to see the outcome that year as contributing hugely to our ability to generalize about what happens in second-term midterms absent resignation in disgrace.

Then there are the supposed trends that don’t even hold up to minimal scrutiny, my favorite being the one about how difficult it is for sitting vice presidents to get elected. The prevalence of this bit of folklore probably owes something to George H.W. Bush, who was fond of saying that he felt a special affection for Martin van Buren, the last sitting vice president to be elected. OK, but how many sitting vice presidents have run? The answer, going back to van Buren in 1836, is five. [Mr. Nixon, Mr. Humphrey, Mr. Bush and Al Gore were the others.] That’s not many.

And by the way, the sitting vice presidents have done very well: Mr. Humphrey lost very narrowly. Many Republicans believe Mr. Nixon would have won but for vote fraud in Illinois; many Democrats think Mr. Gore would have won but for the Supreme Court halting the recount in Florida.

As for presidential elections and the economy, you certainly would want a good economy if your party occupied the White House. But there is also a pretty good competing generalization about presidential contests, for which I am in debt to my friend Douglas Cox: Simply put, the winner is the one who is the better campaigner. If you think back, you will find that this is remarkably powerful.

So here’s a generalization for you: Elections are what you make of them. That seems to have been the central insight of George W. Bush and his political adviser, Karl Rove, in 2002.