Taken very literally, it seems to contradict itself -- if we're assuming the reports are true, why even investigate? The authors are responding to a tendency t

Taken very literally, it seems to contradict itself -- if we're assuming the reports are true, why even investigate? The authors are responding to a tendency to assume harassment reports are false, and then not investigate. We can see there is large zone of action between these two ideas. It's probably fair to say: don't dismiss harassment reports, investigate them.