Over the years, several students of the assassination have tabulated testimony of witnesses who heard the shots in Dealey Plaza.  This sounds like a precise and scientific way of going about analyzing evidence, but in reality the tabulations have differed radically in the results they have produced.  Not surprisingly, students who believe there were multiple shooters have found more witnesses who heard shots from the Knoll, and fewer who heard shots from the Depository.  People who believe that Oswald was the only shooter have consistently found more witnesses who heard shots from the Depository than from the Grassy Knoll.

The table below shows the assessments of four different scholars who have tabulated earwitness testimony.  They are:

• Stewart Galanor, from an appendix of his book Coverup.
• Joel Grant, from a tabulation widely circulated among researchers.
• Josiah Thompson, from an appendix of his book Six Seconds in Dallas.
• Surell Brady, who tabulated the witness data for Bolt Beranic and Newman who were working under contract for the House Select Committee on Assassination.

In making any tabulation, a scholar must face a variety of problems and challenges. In producing my own assessment, I have used the following rules.

• Descriptions of the source of the shots have been assimiliated to "Knoll" or "Texas School Book Depository" where at all possible. Thus, witnesses who said the shots came from "the railroad underpass" or from "the railroad yards" or from "that monument there" (the Pergola) are all classified as "Knoll" witnesses.  These are very different locations, but -- especially for witnesses on the east side of the Plaza -- it would be asking too much to expect witnesses to accurately distinguish them. Likewise, witnesses who said the shots came from "up at Houston and Elm" or witnesses standing at the corner of those two streets who said the shots came from "overhead" are all counted as "Depository" witnesses. Many such descriptions are perfectly consistent with a shooter from the Dal-Tex Building or even the County Records Building.  So strictly speaking, "Knoll" means "anywhere on the west side of the Plaza," and "TSBD" means "consistent with the School Book Depository, and not consistent with the Knoll."
• We have excluded witness inferences from the tabulation. Where a witness somehow came to believe that the shots had come from a particular location, but didn't claim to have heard the shots from that location, their placement is not counted.  The most famous example of this is Abraham Zapruder, who inferred that the shots came from the Knoll because he saw the right side of Kennedy's head explode, and because people ran up to the Stockade Fence on the Knoll.  Likewise, Galanor counts two witnesses as "Knoll" witnesses because they saw smoke in front of the Knoll, although they didn't actually say they heard shots from the Knoll. Witness inferences aren't evidence, and the inferences we might draw -- based on more information than the witness had -- are probably superior in every case.
• Sometimes, the location of a witness and the direction in which a witness was facing are important. Where a witness said "behind" it is necessary to determine which way the witness was facing when he or she heard the shots. The Chisms, for example, were facing down Elm Street when they heard shots "behind" them. Joan Franzen said she heard shots from "an area adjacent to the TSBD building. . . ." From a distance, the Knoll is "adjacent to the Depository," but Mrs. Franzen was standing on Elm Street directly across from the Knoll.
• In a major oversight, past tabulations have included four Sheriff's deputies (Elkins, Weatherford, Mooney, and Sweatt) who were standing on Main Street some distance from the corner of Main and Houston, were not in Dealey Plaza, and had no direct line of sight to either the Knoll or the Depository. They either tried to infer the source of the shots from what they knew to be echos, or repeated hearsay testimony.