I get regular emails from the Barack Obama campaign, and the one that came today contains a piece of rhetoric involving numbers that annoys me. I should preface the post with a disclaimer: the gripe is not specific to the Obama campaign; if I were receiving emails from the Clinton or McCain campaigns, I’m sure I’d find the same kind of misleading rhetoric there as well.

So here’s the offending text:

In February alone, more than 94% of our donors gave in amounts of $200 or less. Meanwhile, campaign finance reports show that donations of $200 or less make up just 13% of Senator McCain’s total campaign funds, and only 26% of Senator Clinton’s.

OK, read that through, and think about what the words mean, and whether you’ve learned anything or not about how the campaigns of the respective candidates are financed.

The implication is clear: the Obama campaign gets more of its funds from small contributions than the McCain and Clinton campaigns do. But this implication is generated by a clever manipulation of words that, with a bit of thought, are seen not to support this claim at all.

Take the first sentence:

In February alone, more than 94% of our donors gave in amounts of $200 or less.

What does this mean? It means there are a certain number of donors, and that 94% of the donors gave $200 or less. We learn nothing about what percentage of the total dollar amount of campaign contributions are from donors who gave $200 or less. The distinction is important; this is not the same thing as saying that 94% of Obama’s total campaign funding comes from people who gave $200 or less.

Let me make this concrete: Say that there were 100 donors, and 94 of them gave $200. This makes the above sentence true. Now say that the remaining 4 gave 100,000 dollars each. So the total dollar amount of campaign contributions is 94*200 + 4*100,000, which comes out to $418,800.

Now we can ask what percentage of total campaign funds were from donors giving less than $200. The answer is given by dividing the amount given by those donors by the total campaign funds, which comes out to about 4.5%. So, in this made up example, the percentage of total campaign funds coming from small donors is only 4.5%, despite the fact that 94% of the donors were small donors.

Now, consider the second sentence from the offending paragraph:

Meanwhile, campaign finance reports show that donations of $200 or less make up just 13% of Senator McCain’s total campaign funds, and only 26% of Senator Clinton’s.

It should be clear at this point how meaningless and misleading this juxtaposition is; we learned nothing about what percentage of Obama’s total campaign funds came from donors giving $200 or less, and we also learned nothing about what percentage of donors to the Clinton and McCain campaigns gave $200 or less. But the text asks us to make a meaningful comparison between the percentages given in the two sentences. So the cooperative reader thinks ‘Oh, I guess this means that the Clinton and McCain campaigns are getting most of their money from big donors, while the Obama campaign is getting most of its money from small donors’. The problem is that this conclusion does not follow from the facts given. The mathematically unsophisticated or lazy reader might even come away from the email thinking that 94% of Obama’s campaign money comes from small donors, compared to the 13% for McCain and 26% for Clinton, which is definitely false.

This sort of numerical trickery makes me mad, and even though it’s completely commonplace in any kind of political rhetoric, that doesn’t make it ok. Given the degree of innumeracy in the population, I suppose it is quite effective though.

I’m posting some pictures of my room, for the benefit of my family and friends who might be curious to see where I live.

pict0106.jpg

pict0107.jpg

pict0108.jpg

I just saw on the BBC that Arthur C Clarke has died. For science and science fiction lovers, this is sad news.

It’s been an unforgivably long time since my last post. I’ve been kept quite busy this semester with my classes and generals papers. And in the back of my mind is the looming prospect of settling on a dissertation topic.

I heard a few days ago that my abstract was accepted to Logic and Engineering of Natural Language Semantics, to be held in Hokkaido, Japan. In my two years in Japan, I never had a chance to go to Hokkaido, so that should be an exciting trip.

I got back into Northampton yesterday. It’s pretty cold.

Thanks to everyone in Texas for a great winter break. See y’all again this summer…

In this podcast Kai von Fintel discusses open access scientific publication in the context of his and David Beaver’s new open access journal Semantics and Pragmatics. There are interesting reflections on dissemination of scientific results in the pre-internet age, and the ongoing revolution in making current scientific results and information more widely and freely available.

I was reading this BBC article about the upcoming parliamentary `elections’ in Cuba, and was struck by the following line:

There are 614 candidates contesting 614 seats.

OK, remember that we’re talking about Cuba here, so the situation is this: For each seat, there is a single candidate. There is no campaigning allowed, no choice for a given seat. Or, one could say:

Each seat is uncontested.

This is one usual way to characterize a situation like that of the Cuban parliamentary elections. Limiting ourselves to a single seat, the following is (at least for me) distinctly odd:

There is one candidate contesting one seat.

The badness of the sentence can be accounted for by appealing to the semantics of the verb contest – simplifying things for the case at hand, for a seat to be contested there must be two or more people contesting it. If this condition fails, and there is only a single candidate, we say that the seat is uncontested, but not that `there is only one candidate contesting the seat’ (in fact, the latter sentence seems to mean that there are two candidates, one incumbent and one challenger).

So far so good, but now what about the BBC sentence describing the situation in Cuba? What is being conveyed (as is made clear in the context of the article) is that, for each seat, there is one and only one candidate, so that each seat is uncontested. But as we just saw, when a seat is uncontested, we can’t describe it as being contested by one person. So the following two sentences should sound equally bad; but are they?

(1) There are 614 candidates contesting 614 seats.

(2) There is one candidate contesting one seat.

To my ear, the second sentence sounds much less natural than the first. This is surprising; given my rough account for why the second sentence sounds bad, shouldn’t the first sound just as bad? What could account for the amelioration of the badness of (2) in (1)?

Here’s an idea: While the intended meaning of (1) qualifies it for the same sort of oddness as we feel reading (2), there are other potential (though unlikely) readings for (1) as well. For example, one could imagine a situation in which a given candidate could win any number of seats. In such a situation, it could come out, for example, that the 614 seats are divided among just 600 of the 614 candidates, with 14 of the candidates winning two seats each. In this situation, each seat is being contested by multiple candidates; in fact, each seat is being contested by all 614 candidates.

Though world knowledge (and the information made available in the article) rule out this and related readings of the sentence in (1), they are nevertheless available as possible interpretations of (1). And on these other interpretations, the weirdness of (2) no longer applies, since the weirdness in (2) was argued to be due to the fact that a certain seat is described as being contested when there is only one candidate running for the seat. Since there is only one candidate in the sentence in (2), there are no other potential readings under which the oddness would disappear.

If this story is on the right track, then what it means is that a sentence that should sound odd under the intended reading can be made to sound (more) natural if there are other readings for the sentence which do not trigger the oddness that the intended reading in fact triggers.

This weekend I took a trip down to Austin. On Friday, me and Dad drove down to Austin and hit up the Half Price Books on Guadalupe, and then walked around the UT campus and ate at Kerby Lane.

Dad at Half Price

UT Tower

On Saturday I went to Zilker park with Jun-san, and had a Mexican dinner with Jun-san and Hatanaka-sensei.

chris-zilker-river_01-08.jpg

jun-junko-chris-eating-01-08.jpg

And of course there were souvenirs. Dad bought Coal a UT slicker. Hook ‘em!

(There’s supposed to be a picture of Coal in his UT slicker here, but WordPress keeps fritzing out…)


I had a tasty burrito bol (that’s how they spell it) with my mom at chipotle yesterday.

Chipotle

For those of you living in the pioneer valley, Chipotle is like Bueno y Sano, but better. I’m not sure what makes it better (nostalgia?), but it is.

Confession: the real motivation for this post was to try out my new camera ^_^

I just discovered a widget for my WordPress blog (what you’re reading) that allows you to post videos, either your own or ones from sites like youtube, to your blog!

The widget links to a site called VodPod, where you can create a free account and start uploading and linking to videos. There’s even a firefox add-on that puts a button at the top of the browser so if you see a video you like you can add it to your pod with a single click.

I like my new blogging service. WordPress rocks! (Don’t worry Google, I still use your search engine).

Oh, and Barack Obama is off to a good start – and I’m feeling happy about it. Oh God, is my cynicism failing me?

Currently Reading

Archive

Watch videos at Vodpod and other videos from this collection.