Well, that’s a reasonable way to look at it, but it’s only half the story. There are some conventions in mathematics and physics that are truly arbitrary (defining “right-hand” as opposed to “left-hand” axes in Cartesian space comes to mind), where everyone does it one way just because everyone else does it the same way. However, defining the angle between vectors in the usual way is not arbitrary, because this definition leads to more natural mathematical manipulation of vectors.
You could, if you wanted to, build up a set of rules for vector manipulation based on your definition of the angle between vectors. Everything would work out correctly. However, the same rules make more “sense” when based on the conventional definition of an angle between vectors. Example: vector dot product multiplication. A dot product is proportional to the cosine of the angle between two vectors. If you redefine “the angle between two vectors” your way, then the dot product is proportional to the negative cosine. No problem mathematically, but kind of weird to remember that (vector A dot vector A) requires a negative sign.
Say that vector B is points slightly upward from A, so that, if you lay them end to end, it kinda looks like this:
(note, vector b is supposed to continue upwards from the initial /, but proportional spaced fonts and space-munching makes that pretty hard to draw)
---->/
The reason why you can’t pick a point on vector A, the meeting point, and vector B and calculate it that was is because, well, wait. You do. But the place that they meet has to be where you center your protractor, and then you line up the protractor with your point on vector A, and read the angle of vector B. If you do this any other way for any vector addition sum other than two vectors pointing the same or exactly opposite direction, you’ll get different answers depending on which points you choose.
But if you measure it the way of the rest of the world, you put your protractor at the meeting point, keep the left hand side of the protractor level with vector A, and measure vector B. In the case of my beatiful ASCII art drawing, this would give you about 45 degrees. Given the original question, it would give you the answer of 0.
(disclaimer, also note that the “resultant vector is greatest” in my example might actually be vector B + vector A, and the angle would then be 135 degrees. But I’m not sure on that wording, and I don’t have your textbook in front of me…)
Note that part of the confusion here is caused by the inadequate phrasing of the question. Using the phrase “the resultant vector” leads one to wonder “the vector resulting from what?” There is no vector resulting from finding the angle between two vectors.
What you really want to know is, given two vectors of fixed lengths, what should the angle of one relative to the other be in order for their sum to have the greatest length?
You can ask, “why is it that we put vectors head-to-tail in order to add them, but we put them tail-to-tail in order to measure the angle between them?” but the answer is that we don’t do either. Those are just how we visualize it. The actual calculation of these things doesn’t require that they even touch each other (and in fact, lacking position, they can’t touch each other, but that’s beside the point :)).