To be fair to them, a scientific record is in many ways just the aggregate of the agreements of the practitioners of [insert experimental / theoretical discipli

To be fair to them, a scientific record is in many ways just the aggregate of the agreements of the practitioners of [insert experimental / theoretical discipline here] who agree that a particular set of data and theories seem to be well-formed. So, scientific record is an aggregate opinion, backed by data. Sometimes the data are experiments, but these experiments require peer validation, or else any old crackpot wcould have their theories accepted. Other times the scientific consensus is less-well-backed by experiement (String theory anyone?) sometimes correctly, sometimes not. This is how induction works!