In an column at The Week, Taegan Goddard raises an interesting idea: The 2016 Republican nomination might not be settled at the ballot box, but negotiated at the GOP convention.

He makes three points.

No. 1, the Republican field is so big and has enough viable candidates that a failure to get a majority of delegates seems like a real possibility. An April Quinnipiac poll showed four candidates at or above 9 percent. When the pollsters asked for second choices among those who picked Jeb Bush and Scott Walker, four candidates got 8 percent or more as a second choice to Bush, and six got 8 percent or more as a second choice to Walker.

No. 2, that none of those candidates seem likely to be able to lockdown a constituency cleanly. We've looked at the overlap of support from what are often thought of as distinct voting blocs -- religious voters, the tea party and conservatives -- who, so far, are not as distinct as you might think.

That "so far" does a lot of work in that sentence. It's so, so early in this race, and so many likely candidates haven't yet jumped in. We've seen new candidates get a bump in the polls after their announcements; that Quinnipiac poll has Rubio in the lead almost certainly in part due to his having announced shortly beforehand.

But it's Goddard's third point that I think breaks this thing wide open. No. 3 is that the existence of Super PACs can make up a lot of empty space where votes should be.

In January, we wrote about how several 2012 candidates stuck around thanks to external PAC support. Newt Gingrich in particular had the robust backing of Sheldon Adelson, allowing him to stay out on the trail longer than he might otherwise have been able to. The 2016 candidates are embracing PACs in so-far-unseen ways, including rumors that Jeb Bush will outsource much of his effort to his PACs. PACs supporting Ted Cruz raised $31 million in a week. Given how little of their fortunes the wealthy need to spend in order to have a significant presence in a campaign, it's easy to think that a few big donations from a few big donors could result in a lot more Gingriches.

So let's say that three candidates divvy up 80 percent of the delegates, with no one getting a majority. Two or three others make up most of the remaining 20 percent. (This is far, far harder than it sounds, given how many states distribute delegates on a winner-take-all basis, but stick with us.) Sticking around longer can make this more likely, since candidates could focus on smaller states (with massive, uncoordinated super PAC air support) to pick up as many delegates as possible.

The party goes into its convention, and things are unsettled. At that point, the goal is to cobble together a majority, somehow. Let's say Walker is 40 delegates away from the majority and Cruz has those 40 delegates. What then? And: What if the super PACs are still outside the convention hall, hammering away on which candidates are most electable and which are most true to the Republican platform?

On the list of "things that might actually happen," this is probably a bit below "huge weather phenomenon postpones convention." But it's conceivable. The big-name candidates will have big-name PACs backing them up that might crowd out the also-rans; the change we see from 2012 is that the arms race has been joined.

But, we'll remind you: In 2012, the Republican convention was delayed due to a hurricane. Which is to say: Weird things happen.