Large reasoning models almost certainly can think

There has been a lot of hullabaloo about the idea that large reasoning models (LRM) are unable to think. Apple argues that LRMs must not be able to think; instead, they just perform pattern-matching. This is a fundamentally flawed argument.