National Battlefield Parks have significant resources for researchers. Nonetheless, researchers must handle material from National Battlefield Parks with care. The materials are often excerpts from documents which exist in their complete form at other repositories. Sometimes the whole document says something quite different from the National Battlefield Park excerpt.
Two examples come to mind.
The first involves an excerpt from the writings of George S. Bernard, who was a private in the 12th Virginia Infantry, Mahone's brigade, Anderson's division, Hill's Corps. He kept diaries and wrote letters throughout the war. After the war, he compiled, edited and contributed to War Talks of Confederate Veterans (1892). He was ready to publish its sequel in 1896 but it disappeared until 2004, when it turned up at a flea market, was bought for $50 and sold to the Museum of Western Virginia History for $15,000. I was one of the co-editors of the book published by the University Press of Virginia in 2012 as Civil War Talks: Further Reminiscences of George S. Bernard & His Fellow Veterans.
The writing in question lies at a certain National Battlefield Park. The writing has been cited at least twice since 1998 for the proposition that Bernard on the night of July 2, 1863 heard Lt. Gen. James Longstreet tell Maj. Gen. Richard Heron Anderson that an improvised night attack involving Mahone's brigade should be called off.
Bernard, however, did not write the piece and was not the witness, though the account was among the papers edited into Civil War Talks. The witness was the 12th Virginia's adjutant, First Lt. William Evelyn Cameron. His account of the Gettysburg Campaign is called "Across the Rubicon" and it forms part of Civil War Talks, 155-156. There is also a copy of it in Cameron's papers at the University of Virginia.
That inaccurate citation is merely embarrassing. The next is far more substantial. It also concerns July 2, 1863.
A historian who otherwise writes very well used a quotation from a private in the 22nd Georgia Infantry, Wright's brigade, Anderson's division, to liven up an account of the charge by Wright's brigade up Cemetery Ridge. This was the charge that Brig. Gen. Ambrose Ransom "Rans" Wright in his official report and in a letter to his wife claimed to have summitted Cemetery Ridge and briefly driven off the Federals.
The problem is that the complete account of the 22nd Georgia private says that because of illness Wright did not accompany his brigade in its charge up Cemetery Ridge, but that his incomparable adjutant, Capt. Victor Jean Baptiste Girardey, led the brigade's charge. Worse than that, there are substantial grounds for believing the private, and they are both well known. First, Wright was admittedly ill earlier that day. Secondly, Girardey was a superb leader who later led troops to victory on June 23, 1864 as well as on July 30, 1864, for which he won the Confederate army's most extraordinary promotion, four ranks, to brigadier general. (He perished in command of Wright's brigade August 16, 1864.)
The historian who used the quotation from the private did not address the additional questions raised by the private's complete account. The failure to do so calls into question the history's accuracy.
Now of course I may be wrong. The historian may have had the complete account. But that situation is even worse. An historian can't duck an issue such as whether Wright actually led his brigade on July 2, 1863. If he did not accompany his brigade on the charge, as the private says, Wright's opportunity to observe comes into question. Instead of talking to General Lee himself, Wright ought to have had Girardey report to Lee.
UPDATE: It's worse than I thought. Another major historian has made the same error with the 22nd Georgia private's statement.
Names have been omitted to protect the guilty
review of 
