Once upon a time, a philosopher came upon a man on the roof of a house that's under construction, and observes the man taking a shingle and nailing it to the half-completed roof.
"Hello there!" calls the philosopher. "Have you noticed that your skyscraper is incomplete?"
"Yes, it's just a house, but it's incomplete," grumbles the man.
"Have you noticed that you are using a hammer?"
"That's right," agrees the man.
"You presuppose that the only tool is a hammer and that every problem is a nail," says the philosopher. "This is why your skyscraper will not stand."
The man scratches his head. "No, nothing like that. I have a hammer, so I'm using a hammer. Do you have something else I could use as well?"
"Well, of course, if you could molecularly fuse all the components of the skyscraper, then it could stand. Yours does not."
"That sounds useful. How do you molecularly fuse things? Do you have a fusing tool?"
"Oh," scoffs the philosopher. "It's impossible! But you think every problem is a nail, as I said, so your skyscraper simply cannot stand. Your hammerism blinds you to how limited is your view of reality."
"Well, if you don't have anything useful, I've got work to do," says the man, and he nails on another shingle.
Meanwhile, it begins to rain; the man goes inside where it's dry to have a cup of tea and wait for the rain to pass.
The philosopher walks off, sighing, saying, "He doesn't even recognize that he's getting soaked and has no skyscraper."
This is what 99% of philosophical discourse about science sounds like to most practicing scientists. Most scientists don't care about metaphysics, and most philosophers don't have ideas of metaphysics that mesh well with the outrageous practical success of science. Most scientists care about ontology but pragmatically (i.e. accepting whatever limitations are placed by epistemology), and most philosophers don't have ideas of ontology that mesh well with the outrageous practical success of science. Most scientists care a great deal about epistemology, and have in practice been doing a pretty good job at it since Francis Bacon, though Popper helped a good deal to tighten up some thinking--to the point where philosophers mostly can't keep up.
So I think Davies is almost entirely off-base here. Carroll is, admittedly, making statements that require some presumptions that go beyond what is supported...but in general (I don't know about Carroll specifically), these sorts of ideas are held quite tentatively by actual practicing scientists. It seems like the universe behaves as if it's following a few simple rules. You can just go, "Oh, huh, that seems like the way it is...let's assume so, keep an eye on it to make sure it is, and keep learning stuff." There is no strong statement about any brute status of anything. It's just--hey, looks like things are this way. Let's go with it!
If it looked another way, well, fine, go with that instead. The degree of fundamental committment is, at least among people who are devoted to empirical confirmation, quite low. But we have the hammer of trying to analyze things this way, and like zero other tools that work at all, so, off we go with hammer and nails.
No miracle. Just "we don't know for sure that things are this way though it seems like they are; we don't know why they are this way if they are, and we don't even know that 'why' makes sense as a question". No zombie living-death (is this some sort of vitalism in disguise?! Or a meaningless turn of linguistic phrase?). There's no popping of laws out of nowhere because there's no mechanistic (or ontological) description of this at all. It's just, "Huh, this seems to work really well."
Indeed, an astute physicist might notice that the philosopher tends to use an awful lot of assumptions based on human-sized temporal and spatial scales...ones that may not hold up so well at radically different scales. For instance, "creative act" implies both cognition and temporal or causal sequencing, but the former seems unfounded and the latter does not seem entirely safe to assume either.
To be a good scientist you're almost required to be a bad philosopher. You have to get very, very comfortable with the constraints of epistemology, to the point where once you can't robustly learn something, you just stop making strong claims and go, "I don't know" over and over.
Of course, as a practical matter it's easier to just provisionally accept logical positivism and go from there. But aside from a few people who have forgotten that it was only ever a good idea due to cognitive expediency, as soon as you push on it, most people back off to a pragmatic statement of models-fitting-observation without any strong claim about the models being reality.