The table above (which is available as a poster from Sanderson's website) probably gives away one of my main examples, the Mistborn trilogy. But who could resist including something that amazing? Not I.
In recent years, certain terms have been bandied about a lot - and occasionally, I was one of those doing the bandying. (Yes, I am aware that is a weird verb.) The terms in questions are those relating to magic systems, and particularly the new, Sanderson-esque wave of them: scientific systems, rule based systems, and the mysterious, unsystematic magic. Through writing my column on magic systems over at Grasping For The Wind, I've been forced to clarify what some of these terms mean in my mind. And because of this, you've probably noticed that scientific magic has become distinct from rule-based magic in my articles. And this is my attempt to explain why.
I've written articles on rule-based versus mysterious systems before - including one here. Rule-based magic systems, to my mind, is what it says on the tin. There's a scale, of course, but at its extreme, rule-based systems operate only according to fixed laws - and while these aren't always immediately known to readers or characters, they are explained eventually, or hinted at. We know how the magic functions, what it can do, and because of that the possibility for high-magic stories and non-deus ex machina resolutions which involve magic come about. That's the far end of the scale, but a lesser degree of 'rule-based' gives us magic not with strict rules, but at least with limitations or costs. Something akin to Jim Butcher's Dresden Files magic in the first book, before the reader learns more. And sometimes, this magic seems scientific - but to my mind, that's a separate category.
This isn't to say that they are mutually exclusive. Far from it. A 'scientific' magic system progressed to any degree is rule based, as far as I can conceive of it. But what would my - subjective! - definition be? Scientific magic systems, simply put, follow the scientific method. For much of rule-based magic, the rules are simply there, unconditional on the areas in which they've been tested, and if they had an origin, it's historical. In other words? There's no hint that these rules were researched; no trial and error; no extrapolation that's labelled as such. Conversely, a scientific system would be one where the rules are best guesses, approximated by testing and a hefty dollop of induction. And where 'universal' laws are just predicted to be universal, but aren't necessarily. Magic like early science, in other words.
Oddly enough, Brandon Sanderson's systems are some of the only rule-based I'd consider 'scientific': and as you might have guessed, I'm using Mistborn as an example once again. It's quasi-scientific - the initial rules are just presented as laws. But later on, characters make deductions based on these rules, and they test these predictions - combining new alloys of metals to test if they are 'burnable' via allomancy. And some of those initial rules and discoveries are disproved, just in the same way antiquated scientific theories are. It's not just limited to alloys, either - certain other predictions are put to the test. And yes, I'm being vague. But Mistborn is far from a universal read, so I refuse to spoil it. ;)
That's my take on the definition, at least - so here's hoping that we see more of this interesting brand of system.