I’ll give it a stab.
I’m not sure what is obscure to you, so I’m going to elaborate on points that might be simple.
To make sure we are on the same page, I’ll define the terms here. Say apples are a dollar a pound right now in the store. A marginal buyer is someone who will buy the apples for a dollar a pound, but not for $1.01 a pound. A submarginal buyer won’t even pay 99 cents a pound. A supramarginal one will pay $1.01 or more if he had to.
Next we have to make sure we understand that there is nowhere in the real world that has the data needed to draw a demand curve, or a supply curve for that matter. If we want to draw a demand curve for yesterday, for instance, we can know that 100 people bought apples yesterday when the price was a dollar a pound. Do we know how many would have bought apples if it they were 90 cents a pound, or $1.10 a pound, or any other number besides a dollar a pound, which was the market price? No we can’t, for we are not mind readers. We cannot travel into alternate universes where the price of apples was different and see how many people bought apples in those alternate universes.
Thus, the whole supply and demand curve thing is a very abstarct theoretical construct, a visual way that helps some people communicate and/or understand certain concepts of economics.
The only things we can say with certainty are that if apples were cheaper, there would not be less people who bought them, all other things being equal, etc. [Rothbard describes this by saying the curves have to be rightward sloping].
So when we say a supply curve moves to the right, we are saying that we are going to analyze two hypothetical situations at two different times. One the first day, we know by Divine revelation not only the price of apples and how many people bought them, but also how many people would have bought apples at every price possible. On the second day we not only how many people bought apples and at what price, but again we know by Divine revelation all the might have beens.
As we examine the data, we notice that for any given price, more people would have bought apples on the second day than on the first day, no matter what price we choose to examine. This is captured on the curves [Chapter 2, Section 9, on page 144] by the fact that at any given height on the charts, the second day curve is to the right of the first day’s curve.
Obviously, to be able to say that the whole demand curve moved to the right, we have to be able to see the whole demand curve. Meaning that if Divine revelation had merely shown us a part of the curve, for example just the dot where the actual price was, we would have no idea what the rest of the curve looked like. For example, the second curve may look what you would get by moving the first curve two units to the right, and then rotating it forty five or more degrees counterclockwise, which would cause part of the second curve to be to the right of the first one and part of it to be to the left.
The situation you are describing in your original post would [I think] move a part of the curve to the right [or left], and leave the rest of the curve intact, as if someone had poked the curve right at the actual price, causing it to bulge a bit locally. That is an interesting situation, but Rothbard decided to discuss a different one, where the whole demand curve moved.