The construction of logical, testable hypotheses is integral to the scientific method. Testing a hypothesis is a way to learn about a phenomenon in an objective way.
The concept of using multiple hypotheses was put forth in the 1890s by Thomas Chrowder Chamberlin, a well-known American geologist.
Using multiple hypotheses offers several advantages over working with a single hypothesis. First, constructing multiple hypotheses forces the scientist to really consider all possible explanations, and to approach the problem from several angles. It also prevents the scientist from becoming overly attached to a single hypothesis, which does happen and can introduce bias to the experiment.
If the phenomenon being studied is actually the result of multiple causes, having more than one hypothesis makes it more likely that the scientist will detect the interaction that is creating the effect.