I think the basic problem here is that the classification of oranges is not a physical pattern in my brain, it's a concept that instantiated as a physical concept in my brain. Since it has many, many, different instantiations in multiple physically unconnected locations, it seems strange to call the concept physical. You can do it, of course, but all that really serves to do is confuse the idea of what physical is.
The initial point by
ryan was that what we can broadly call "structures" seem to have physical effects. Since structures are not matter, it seems clear that we need both matter and structures to explain the material world. However, what you are talking about seems to be a different issue.
That could well be.
You make frequent appeal to 'a realist sense' but I'm concerned that what you mean here is actually 'a physicalist sense'.
The point I'm making is that an immaterial concept of ownership and sale is an accurate, useful and reliable way of dealing with the phenomenon, while a discourse on the neural changes made in the buyer's and seller's brain does not, in practice, accurately capture what is going on.
Now you can make a conscious choice to try and base everything in physical interaction. So to represent a concept, you imagine a theoretical set of physical interactions that would populate all the necessary mental, physical and social interactions, from words on paper, to computers to people's heads, in which that concept is instantiated, or could ever be instantiated, in all of space and time.
It's possible, but there are some objections. A few off the top of my head...
The first is that this isn't a particularly useful way of representing a concept. The concept of ownership neatly categorises certain kinds of social relationship. A vast sprawling collection of physical states that are physically unrelated to eachother, while it may cover the same ground, doesn't do the same work.
The second is that you're seriously violating Occam's Razor here. Insisting on the 'reduction' of a fairly simple and straightforward concept to a vast array of otherwise unrelated and unconnected physical processes is about as far from the concept of maximum parsimony as it's possible to get. This is important largely because Occam's Razor is the most common reason given for adopting some form of physical explanation in the first place.
The third is that it's not really an adequate replacement. Replacing a conceptual category with the instantiation of everything that could fall into that category is not an equivalency. A category is not equivalent to its contents, a set is not equivalent to its members. One does not accurately replace the other.
A fourth objection is that you lose the distinction between fact and fiction. If 'the scientific method' is merely a collection of physical descriptions or conceptions of that concept, and 'Santa Claus' is similarly a collection of physical descriptions or conceptions of that concept, then on what grounds are they treated differently?
So while you certainly can decide to replace all mention of immaterial things with physical things, it's not clear to me that it's somehow desirable, or accurate to do so.