In trying to reconstruct the past and understand it, often one of two issues arise: too much or too little evidence. In both cases, the historian has to make decisions and evaluate the evidence in order to come to a conclusion. How does the coherence of different pieces of evidence influence the degree of certainty and hence the reliability in gaining knowledge? If evidences cohere i.e. many sources have the same information, an event seems very likely and it can be said that a truth is found. An example for this is the attacks of 9/11, especially the attacks on the World Trade Center. As it happened in 2001, it is an event of the past. However, testimony of all the people on the streets will say the same information: two planes hit each of the two towers of the World Trade Center. A piece of knowledge has been found and the past has been reconstructed.

However, there are also events where a coherence of evidence did not result in a true piece of knowledge, even in my personal experience. For example in Kindergarten, a group my friends and myself broke a flower vase in the garden of the Kindergarten while playing. When the teachers found out, they questioned the whole class. Another child that had seen us broke the vase told on us, hence making the teacher especially suspicious on us. However, we cohered on our testimony in saying it was not us who broke the vase, but a cat that ran through the garden (as we agreed on beforehand). The teacher believed us and we did not get into any trouble. Although this is on a quite unsophisticated level, it is an example where coherence was an obstacle in trying to reconstruct the past. In fact, because coherence often seems reliable, it can quite frequently result in untrue knowledge. In contrast to human or natural sciences, where coherence is almost always evidence of truth, verifying knowledge, coherence in history can lead