I am just back from two weeks in Greece. A visit to the cradle of democracy and contemplation of events Before the Common Era provides perspective. Besides, Greece is beautiful and retired people get to travel in October. But so much back here is messier now than when I left.
Despite all the polls, analytics, and forecasts, I think it is unwise to be too confident that any of us know what will happen in 11 days. A lot is close; in the last few election cycles, close polling has presaged a wave in one direction or the other, and the trend the last couple weeks has not been good for the Democrats. But the past is an imperfect predictor of the future, or even of the present. I am concerned, also, that such prophecies become self-fulfilling, creating rather than measuring momentum. Past performance is a useful predictor in targeting as well, but it does seem to me a bit overdone. Upsets do happen as a result of candidates or chemistry. The first U.S. Senate race for which I polled was Paul Wellstone’s in 1990, back when I was too new and naive to understand he couldn’t win. (For those who do not remember, he did indeed win.)
So, having learned from the oracle at Delphi how to be properly ambiguous my prediction is JUSTICE WILL PREVAIL NOT LOSE GROUND NOW. The meaning of that depends on how you see justice and whether you place the comma before or after the word “not.” Thus it’s correct – if interpreted properly.
If this is a wave election, there will be blood. It seems rather likely that many will call for a blood sacrifice of the pollsters. I do not think that will work any better than the blood sacrifices of pre-classical, pre-democratic Greece. True, if you sacrifice animals or even people after an earthquake, you are unlikely to have another earthquake right away but that may well have nothing to do with the sacrifice.
Now, I have been very clear in this blog and to anyone who asks that I believe people need to change and expand their research protocols. Polling is hardly the only form of research available, it does not work the way it used to – or the way people think it does. It also looks at the aggregate, which is less useful in the internet age, encouraging aggregate media like TV and lessening the emphasis on organizing on the ground, or by internet networks. There is utility in knowing aggregate attitudes but as an early step in a strategic process which now in my view over-relies on polling.
But the problems with polling should not swamp an examination of the problems with campaigning, which seems far less connected to people than in times past. And the media’s coverage of politics seems highly problematic and often destructive of the democratic process. It emphasizes polarization for the drama, forecasting and predictions for their ease, and in the process makes change, creativity, and conversation with the middle more difficult for everyone. The middle, which is bigger than some think and includes soft partisans, is increasingly non-participatory in polls and in reality, which also makes the polarization worse.
So, yes, we need better ways of doing research. But also a different attitude about listening to people and their views, more individual contact, less nationalization. And far less forecasting which does not, as far as I can see, contribute much to the dialog at all and risks creating a conversation from the top-down which alters results from the bottom-up. Besides, even if you sacrifice the pollsters, it wont affect the timing of the next earthquake.
JUSTICE WILL PREVAIL NOT LOSE GROUND NOW