Sacking Old Records
Actually the census records weren't destroyed until some 21 years later when, over the objections of genealogical and historical organizations, Congress authorized their destructon on February 21, 1933 rather than deal with the issue of preserving the dried out documents which were still pretty much in tact, though water damaged. The only portion of "special schedule" that was systematically preserved was the information pertaining to surviving Civil War veterans. Seems the GAR had more clout in those days than the NGS or even the DAR. Disposition of Useless Papers in the Department of Commerce, 2d sess., No. 2080; Thurber, "The 1890 Census," p. 8; Note, n.d., signed E.L.Y, folder "Census of 1890," box 9, Alphabetical Subject File, entry 160, RG 29, NA. E.L.Y. is presumably Evelyn L. Yeomans, on the staff of the Geography Division from 1899 to 1941, who "apparently maintained the Division files and answered requests for information from and about the old census schedules." See Katherine H. Davidson and Charlotte B. Ashby, comps., Records of the Bureau of Census: National Archives Preliminary Inventory 161 (1964), p. 53. Cited in ""First in the Path of the Firemen" The Fate of the 1890 Population Census" by Kellee Blake in Prologue, Spring 1996.
Of course "our past" wasn't lost. History exists on its own in the continuum of time with or without records to document what occurred. Sort of like the sound of a tree falling in the woods with no one around to hear it. Well maybe a little different since the definition of "sound" requires a listening ear. But it's a similar idea.
I agree completely about the the tragedy that occurs when important human knowledge (or information) is lost, whether talking about the loss of the 1890 census in 1921, the sacking of the Iraqi Museum in 2003, or of the great pagan library at Alexandria around 642 CE by Arab conquerers.
The real tragedy of such losses is that they most often occur as a deliberate act, whether of war or carelessness.