There is a lot that's left ambiguous in this chapter, necessarily so, as it's only intended as a sweeping overview of the rest of the book. Rest assured, we'll going into these things in quite some detail in the coming chapters.
The meanings of "physical" and "physicalism" are discussed at length in chapters 2 and 3. We'll discuss this in more detail later, but briefly, physical properties are understood to be the types of properties included in the fundamental ontology of physics. (As physics is not yet complete, its ontology is bound to change, but the types of things in its ontology will share some basic similarities with the ones it has now.)
Physicalism is the view that all facts about the world are either facts about physical properties, or are directly entailed by these physical facts. Once we specify what we mean by "physical," it is not circular to define physicalism in this way.
It should be clear after chapter 2 what counts as physical. When we get into the second half of the book, we'll see that physical theory is essentially a theory of what Rosenberg calls effective properties. Physics is not a theory of receptive properties or carriers, because neither of these is included in its fundamental ontology, and neither of these is entailed by its fundamental ontology. Thus, only one of the proposed three aspects of causation can rightfully be called 'physical,' and so the others are non-physical. Again, this will make more sense later on.
We'll be going into causation extensively in part II of the book, which is still a ways off. For now, if you're interested, you can skip ahead and read sections 9.1 - 9.6 for a discussion about causation.