We may use hypotheses that contradict well-confirmed theories and/or well-established experimental results. We may advance science by proceeding counterinductively.He argues that such a standardized methodology of science in the status quo, where theories and hypotheses that contradict current observations, experimental results and other "facts" are discouraged. The example for his justification is very interesting, thought-provoking, and fundamental to science itself:
When we see that a table is brown, we tell ourselves (unsurprisingly) that, "the table is brown." When others come up to us and say, for instance, that the table is purple, we immediately strike the idea down.
And why wouldn't we? After all, we "know" ourselves (unless we are colorblind, of course) that the table is indeed brown. But aren't we assuming something?
As Feyerabend points out, we "take it for granted that the material medium between the object and us exerts no distorting influence, and that the physical entity that establishes the contact - light - carries a true picture." Such abstract assumptions, which many would consider as simply trivial, actually "shape our view of the world without being accessible to a direct criticism."
In fact, unless we are faced with a contrasting cosmology (in this example, the statement that the table is purple), we fail to recognize and resolve the underlying assumptions.
Thus we can agree that "a theory may clash with the evidence not because it is not correct, but because the evidence is contaminated.
Analogously, in a world where hypotheses that contradict "given" observations are rejected, such principles would be impossible to examine. This is why Feyerabend argues for a pluralistic approach of science, so that we can create an environment for clashes with the most established results and the most plausible principles.
The only way to achieve this, of course, is to remove the framework of an organized methodology of science. Only in a scientifically anarchistic nature, where "anything goes," can we be unlimited in our scientific endeavors.