On the Alleged Correlation Between Bad News for Bush and the Terror Alerts
Sadly, I'm going to have to call BS on our own side. Via the indispensible Agonist comes this list of alleged correlations between bad news for the Bush administration and the issuance of terror alerts.
I have admitted and still admit that there is prima facie reason to suspect that politics has entered into the decisions about when to issue terrorism (or "terror") alerts--in large part because politics seems to enter into so much of what this administration does. However, the point at issue is this: how much support does Julius Civitatus's evidence provide for the hypothesis that the administration is manipulating the terrorism alerts for political ends?
I'm no statistician. Far from it. My knowledge of statistics is pretty much limited to (a) what I learned in the gutter, (b) one class in statistics for political science, and (c) listening to many drunken lectures on the Central Limit Theorem by Statisticasaurus Rex. That having been said, I'm still fairly certain that this evidence is prohibitively weak.
I'd remind us all that correlation does not entail causation, but we haven't even gotten that far yet--we don't even have a real correlation here. In order to make a strong argument for the hypothesis on the basis of evidence like this, we'd need to know not only (i) how closely terrorism alerts followed bad news for Bush, but also (ii) how closely they followed good news for Bush, (iii) how closely they followed bad news for Bush'spolitical opponents, and (iv) how closely they followed good news for Bush's political opponents. We'd also have to predesignate (A) how closely such an alert has to be to good or bad news in order to count as following it and, I think, (B) (roughly) what kinds of things will count as good or bad news. (What counts as a terrorism alert is clear enough that we don't have to worry about that.)
The case is complicated by the fact that there has been so much bad news of late for the Bush administration.
It goes without saying that I could be wrong about this, especially about the details. I'm writing this on the fly. But I am fairly certain that caution is called for. Remember: as terrible as this administration is, and as important as it is to get them out of office, there are bigger things at stake here. The issue of what methods of inquiry and persuasion we allow into our political discourse is bigger even than this election, for decisions about this will have repercussions far into the future, affecting not only this election, but unknown numbers of future elections.
Oh, and the truth. That's at stake too. Let's not forget about that.
Sadly, I'm going to have to call BS on our own side. Via the indispensible Agonist comes this list of alleged correlations between bad news for the Bush administration and the issuance of terror alerts.
I have admitted and still admit that there is prima facie reason to suspect that politics has entered into the decisions about when to issue terrorism (or "terror") alerts--in large part because politics seems to enter into so much of what this administration does. However, the point at issue is this: how much support does Julius Civitatus's evidence provide for the hypothesis that the administration is manipulating the terrorism alerts for political ends?
I'm no statistician. Far from it. My knowledge of statistics is pretty much limited to (a) what I learned in the gutter, (b) one class in statistics for political science, and (c) listening to many drunken lectures on the Central Limit Theorem by Statisticasaurus Rex. That having been said, I'm still fairly certain that this evidence is prohibitively weak.
I'd remind us all that correlation does not entail causation, but we haven't even gotten that far yet--we don't even have a real correlation here. In order to make a strong argument for the hypothesis on the basis of evidence like this, we'd need to know not only (i) how closely terrorism alerts followed bad news for Bush, but also (ii) how closely they followed good news for Bush, (iii) how closely they followed bad news for Bush'spolitical opponents, and (iv) how closely they followed good news for Bush's political opponents. We'd also have to predesignate (A) how closely such an alert has to be to good or bad news in order to count as following it and, I think, (B) (roughly) what kinds of things will count as good or bad news. (What counts as a terrorism alert is clear enough that we don't have to worry about that.)
The case is complicated by the fact that there has been so much bad news of late for the Bush administration.
It goes without saying that I could be wrong about this, especially about the details. I'm writing this on the fly. But I am fairly certain that caution is called for. Remember: as terrible as this administration is, and as important as it is to get them out of office, there are bigger things at stake here. The issue of what methods of inquiry and persuasion we allow into our political discourse is bigger even than this election, for decisions about this will have repercussions far into the future, affecting not only this election, but unknown numbers of future elections.
Oh, and the truth. That's at stake too. Let's not forget about that.
posted by Winston Smith at
7:50 PM
0 Comments:
Post a Comment
Subscribe to Post Comments [Atom]
<< Home