Why is it so important to begin with only the strongest evidence?
11 February 1999

It strikes me as a truism that if we are going after the truth of the JFK assassination, we must take all possible measures to free our efforts of error. Since errors may easily enter our evidence and our reasoning, we must examine both very carefully. This essay considers the need for purity of evidence; a parallel essay will deal with reasoning.
    It also seems self-evident to me that to avoid erroneous evidence, we must avoid all data of uncertain reliability because we cannot be sure they are correct. Thus, we must test our data harshly and make sure that we retain only those pieces that are guaranteed to be reliable. It is very important to be conservative here—better to err on the side of eliminating some data that are correct than to let in any data that are wrong. The moment we include any data of dubious reliability, we can no longer be sure that any subsequent conclusions are correct. Our goal needs to be much like the physician's "First, do no harm." I see no way to avoid these harsh measures if we really want to get there.
    But this approach is really nothing new—we do the same in all matters of importance in our daily lives. When we go looking for a new car, for example, we want to be sure that we are provided with the correct information. If a salesman misleads us or offers incorrect information, we go somewhere else immediately. When we set out to buy a house, we often get it inspected so that we know the truth about it. Why then should we be any less rigorous with the JFK assassination?
    Surprisingly, few JFK researchers treat their evidence like this. They let in anything and everything, and hope that somehow the weaker information will be sorted out in the wash. Even if they give initial lip service to hard evidence, they don't follow it in practice. They just take it all in and arbitrarily emphasize what seems best to them.
    Those practices guarantee disaster. The link between evidence and conclusions cannot be beaten—include unreliable evidence and your conclusions will be unreliable. I submit that no one among us wants that.
    We should be proceeding in the opposite manner to current practice. We should first examine each piece of our evidence carefully and rigorously, and then discard any that we cannot validate. With the remaining reliable data, we can move confidently to the next step, which is reasoning properly with it.
    I repeat—the moment we allow uncertain data into our analysis, we can no longer be sure of any conclusions that follow from it.
    Of course, clearing out all the bad data will significantly restrict the number of conclusions we will be able to draw. It will also restrict the number of topics we will be able to consider. I believe that many JFK researchers consider this penalty too harsh to bear. I say the opposite—I consider it a relief to have settled this all-important question. The last thing I want is to be wrong about any part of the assassination. Far better to remain silent about some part than to be wrong about it.
    How do we put these principles into action? The answer is contained in Steps 3a,b of the pattern of critical thinking offered in "The scientific method and the JFK assassination." First we divide the available evidence into strong and weak, where strong means objectively falsifiable. (For further discussion of strong and weak evidence, see the essay "Types of evidence useful for understanding the JFK assassination.") Then we eliminate all the weak evidence from further consideration and proceed with the strong evidence.
    Step 3 is perhaps the most distinctive of the nine steps in this pattern of critical thinking. It is controversial among JFK researchers because it goes against the grain of so much of their research. And yet it is fully justified from the standpoints of logic, common sense, and tradition. It offers the fastest and most efficient path to delimiting the answer. It offers the most realistic assessment of the quality and quantity of the available evidence. It offers an immediate sense of how close to an answer we are likely to get. (See the essay on asking the right first question.) It is not too strong to state that anyone who does not follow this method is going against logic, common sense, and centuries of critical tradition, and is guaranteeing an uncertain, if not wrong, answer. Seen in this light, there is no other choice.