who first asserted that scientists should conduct narrow experiments to answer specific questions