Towards A Good Samaritan World

Wednesday, January 19, 2005


I visited CBS News's main page today for the first time, and immediately ran across a great example of why it's difficult to read the MSM after you get used to blogs. The third paragraph of the lead story, amidst paragraphs consisting of clear, indisputable facts, is as follows:

Bush won re-election by besting John Kerry on the issue of who could best handle the terrorist threat and by casting his arguments in moral and religious terms.

How does CBS know? Indeed, what would the statement even mean?

The first problem with a statement like this one is that it is a generalization where generalization is highly dubious. Bush won because he got 3.5 million more votes than John Kerry. Why he got them varies with each vote: some were voting against abortion, some because they liked Bush personally, some were supporting the troops, some admired Bush for liberating Iraq, some wanted to keep taxes low, some didn't trust Kerry.

In view of all this, could any one-sentence generalization beginning "Bush won re-election by..." be valid?

Unless every single Bush voter voted for the reasons CBS states, the meaningfulness of any such generalization depends on some kind of statistics. For example, if 35% of voters supported Bush because they like his hairdo, 10% of voters supported him because he comes first in alphabetical order, and 6% of voters supported him for a variety of policy-related reasons including Iraq, tax cuts, and loose environmental policies, we could justify a statement such as

"Bush was elected because of his appealing hairdo and his last initial."

because most of Bush's supporters voted for him for one of those reasons. We would ignore the policy-related voters because they were relatively few.

Yet on the other hand, without the 6% of policy-related voters, Bush would still have lost. Those voters were necessary to give him a majority, so perhaps it is unfair to exclude them. Let's now suppose that 37% of voters supported Kerry because they preferred his hairdo, and 12% because his name came later in the alphabet. In this case, Kerry leads among hairdo voters and among alphabet voters, but Bush gets all the policy voters. Since the policy voters tipped towards Bush and gave him a majority despite his trailing in the other categories, we might justify the statement:

"Bush was elected because his policies were more appealing to swing voters than Kerry's were."

So, even if all voters have well-defined reasons for their choices and we have perfect information about those reasons, any brief generalization we make about the reasons for the electoral outcome will be in large part arbitrary.

In real life, voters have many and complex reasons for their choices. A peacenik free-trader might say: "I would have voted for Kerry if he promised to withdraw from Iraq, but since Kerry wasn't offering a clear alternative, I voted against Kerry because of his protectionism." Is this voter voting based on trade, or based on Kerry's failure to define his positions? A Christian fundamentalist might say, "I stayed home in past elections because the Republicans didn't convince me they were really pro-abortion; Bush did, so I voted for him." Should we consider this guy a swing voter? And what if someone said "I opposed Bush on the deficit and the Patriot Act, but I supported him on the war in Afghanistan, tax cuts, and trade; I decided the positives outweighed the negatives." Which issues is this voter voting "on?"

More importantly, we don't know the reasons for voters' choices. We have patchy evidence in the form of polls, but polls have to resort to crude simplifications of people's real opinions in order to standardize them and compile them into statistics. In the process much is lost, and much is read back in.

Now what is really remarkable, even weird or shocking, about CBS News's statement is that it is placed in an article alongside all sorts of straightforward facts like:

The schedule begins with a Military Gala on Tuesday, January 18th and ends with a National Prayer Service on Friday, January 21st.

This mixing-up of cold facts with conjectures is highly deceitful, because it implies that all the statements in the article have the same epistemic quality.

Now if a blogger made CBS's statement about why Bush won, he or she might feel the need to offer justification for the statement, including, for example, links to and analysis of, polls. If the blog had comments it might touch off a debate. Even if the blogger did not justify the statement, at least we would understand that the statement was offered as an opinion, and in good faith.

But when CBS puts the statement in a news story, my reaction is different. Because they mask opinion as fact, I don't even trust that they hold the opinions in good faith. They have an agenda, and they manipulate the facts to match the agenda, rather than letting their views be guided by the facts. They feel entitled to do so.

I've gotten spoiled. After the epistemic honesty of the blogosphere, reading the MSM makes me feel violated. And that's probably not a good thing. Because the blogosphere will probably not adequately substitute for the MSM's functions for a long time.


Post a Comment

<< Home