In late fall, I attended a reading by Steven R. Stotelmyer of his paper arguing that there was but one battle at South Mountain and that Crampton's Gap was an action fought within that battle. I took some good notes and had some useful correspondence on the matter. Last month, the Bivouac Banner published an article on the same subject by the same author; I'd thought I could give you a summary criticism of this information in a few posts.
Unfortunately, I see that the Banner's article is not exactly the one I heard presented, so I have to do a close reading of these 43 pages; and I see that my beta version Gmail account no longer holds any of my November and December correspondence on the subject.
So, it's back to the fireside for some reading and thinking.
In the mean time, consider a foretaste of reasoning by analogy. The Waterloo campaign is divided into four battles: Ligny and Quatre Bras (6/16/1815); Wavre and Waterloo (6/18/1815). None of these points is more than five miles away from the other. What makes them "battles" - as close in time and space as they were - are discrete objectives and discrete commanders.
Stotelmyer has gathered the best arguments possible with which to defend the Maryland park system's historical pretensions. The matter deserves a thorough response. Please have a look at Tim Reese's views on this while I organize my own thoughts.