According to the generally accepted theory of plate tectonics, the earth’s crust consists of a dozen or so plates of solid rock moving across the mantle—the slightly fluid layer of rock between crust and core.
Topic
the relationship between subduction and earthquakes
Framework
Old/New or Answer a Question
Main Point
Head-to-head shallow subduction is more likely to cause earthquakes, whereas head-to-tail steeper subduction is not as likely to cause earthquakes.
Summary
P1 - the Accepted Theory (the Old), the data it fails to explain, and the Question
We're told that the accepted explanation for how earthquakes happen is heavily tied to the concept of subduction (one tectonic plate sliding underneath another, into the Earth's mantle), and that most quakes occur in high subduction zones. Contrary to expectations, however, (the purpose Pivot), there are lots of areas with tons of subduction that don't experience many quakes. What's the deal?
P2 - a proposal that would Answer the Question (the New)
A group of scientists is saying,
We might be wise to keep a few causal relationships straight:
leads to quakes doesn't lead to quakes 1. head-to-head head-to-tail 2. shallow subduction steep subduction 3. underlying mantle not underlying mantle is moving much relative moving a good bit to subduction relative to subduction
P3 - implications of the New
The author seems to be implicitly accepting this new theory and explaining one interesting takeaway -- you might think you live in an area safe from earthquakes if there's little subduction, but if the subduction that does occur is usually head-to-head, then you still might see quakes.