It's less about making an argument for the "correct" definition of "smart"; and more about saying that the way psychologists measure "intelligence" (eg when trying to study things like "are smarter people more or less happy") is measuring success at solving well-defined problems. That things like IQ tests as well as academic success are about solving well-defined problems.

You can think "smart people" are whatever you want, it's not about any "true" definition of what "smart" means, it's just a word that means different things to different people -- often probably not very rigorously defined or consistently applied. The OP is in fact about investigating what some of these different meanings may be with relevance to correlation to happiness.