If you haven’t figured it out by now, I live in Toronto, Canada. This means I get double the dose of politics, which in turn means I get to stick my nose in twice as much stuff.

Approximately seven weeks ago the Canadian people went to the polls to elect a new government. Because there are three main Canadian parties and one dedicated to the interests of Quebec, the likelihood of any one party winning a majority is small. There are a total of 308 ridings and the Conservatives were hoping to win enough new seats so they could have a majority government. They did win more seats (19), but not enough to have a majority. Here is how the election broke down by seats won and percentage of seats available.

• Conservatives:    143  46.4%
• Liberals:               76    24.7%
• Bloc Québécois:  50   16.2%
  
• New Democrats: 37    12%
  
• Independent:       2      0.7%
  

Since the Conservatives have the most number of seats, they are the ruling party. In the last week, for various reasons (which don’t really matter for my purposes here), the leaders of the Liberals, the Bloc Québécois, and the New Democrats decided that they were unhappy with the Conservative government and wanted to form a coalition that would unseat Harper and have the head of the Liberal government as Prime Minister (currently Stephane Dion). Because the combination of these three governments would give the coalition 165 seats as opposed to the Conservatives 143, the coalition could oust the Conservatives and take over.

With that background in mind we can see how many people are reasoning fallaciously concerning the proposed coalition government. The leaders of the coalition parties (and many of their supporters) seem to be reasoning this way:

1. Canadians voted that the Liberals should have 76 seats.
   
2. Canadians voted that the Bloc should have 50 seats.
3. Canadians voted that the NDP should have 37 seats.
4. So, Canadians voted that the Liberals, the Bloc, and the NDP should have 165 seats.
5. Canadians voted that the Conservatives should have 143 seats.
6. Therefore, if the Liberals, the Bloc, and the NDP form together, more Canadians voted for that coalition than for the Conservatives.

So what is wrong with this reasoning? Why is it fallacious? I mean it’s true that more Canadians voted for these parties individually than for the Conservatives, so it must mean that more Canadians would want a ruling party to be formed from these individual parties than for the Conservative party. As we’ll soon see, this is the Fallacy of Composition.

People commit the fallacy of composition when they assume that what is true of the parts must be true of the whole. We can easily see why this is fallacious with a common example.

1. It is true that parts of my body are invisible to the naked eye.
   
2. Therefore, my body as a whole is invisible to the naked eye.

Here is another example that plenty of sports writers are guilty of committing.

1. Team A is better at every postition than Team B.
2. Therefore, Team A as a whole is better than Team B.

Both of these are examples of fallacious reasoning. It is obvious that there are parts of my body that can’t be seen with the naked eye, but it’s also obvious that my body as a whole can be seen with the naked eye. What about the second example? Well, it might be true that Team A has bettter players at every position than Team B, but Team A might not practice very much together or might have more selfish players at every position. Either case would make it reasonable to think that as a whole Team B is better (e.g. the 1980 U.S. hockey team).

Okay, so how does the relate to the recently formed coalition government in Canada. If people assume that because more Canadians voted for parts of the coalition that more people voted for the whole coaltion, then they are reasoning fallaciously. Many people that voted for one of the coalition parties individually might be quite unhappy with what results by combining them altogether. If, for example, I care about issue X and voted for the Liberal party because of their committment to issue X, but in order to form the coalition the Liberals had to give up issue X, then that might be enough to lead me to not vote for the coalition at all. After all, the issue I really care about was just given up by the Liberal leader. If these types of scenarious weren’t plausible, then why have separate parties to begin with? There must be at least a few issues that separate the parties, but if those are the ones I care about then why would I continue to support a group of leaders that just gave up on that issue?

The only way we can know for sure if more people want to have a coalition of Liberals, Bloc Québécois, and New Democrats running Parliament and not the Conservatives is if there were an election with these two options. Of course that would never happen because as soon as an election is called each party would go back to trying to win votes for their own party. Keep in mind that it may be true that more Canadians want a coalition government, but we can’t know that from the fact that more Canadians voted for the three parties individually than for the Conservatives.

As an American, I find this whole thing really fascinating. It is also interesting to see that American politicians don’t have a corner on the ‘bad reasoning’ market.