”Males have more teeth than females in the case of men, sheep, goats, and swine; in the case of other animals observations have not yet been made.”
Bertrand Russell has made much of this:
“Observation versus Authority: To modern educated people, it seems obvious that matters of fact are to be ascertained by observation, not by consulting ancient authorities. But this is an entirely modern conception, which hardly existed before the seventeenth century. Aristotle maintained that women have fewer teeth than men; although he was twice married, it never occurred to him to verify this statement by examining his wives’ mouths.”
This has generated some significant discussion and the need to "rescue Aristotle":
The problem is that, as we have seen, Aristotle did not arrive to his conclusion through logic, but via faulty data.
And... arriving at bad conclusions because of faulty data is not as bad as through failed logic?
We, as scientists, value being "smart" and "clever" and "insightful" to an extreme. Sometimes, we perceive bad reasoning as worse than bad data. Bad data could be someone else's fault, as was Aristotle's problem. But bad reasoning or logic is a fatal flaw in a society that values thinking.
Two problems exist with this line of thought. Firstly, bad data implies bad thinking at some level. If one puts one's conclusions, based on bad data, out there in the literature, it is our responsibility to make sure that the data are right. There are lots of reasons data can be bad. Sloppiness and a rush to publish. Bad technique by trainees or others collecting the data. Errors in methodology that are not to know to the data collector, or the PI, or anyone for that matter. Willful ignorance, abject confusion, or even genuine lack of knowledge: these are things that need to be considered.
Secondly, if the end product is what counts, then it doesn't matter why one gets the wrong answer. The data and the logic are partners and collaborators in the results, and I am not sure why one should have the priority over the other. This is a discussion I have had with a lawyer friend, at great length with much scotch. Lawyers accept, nay promote, the idea that motivation is important. Most people do. Taking the other side (which I do not totally believe, but am happy to for the sake of a good argument with a better scotch), it is defensible to argue that motivation is irrelevant. What those results are is what counts. Sometimes I think that the knowledge that it could be an "honest mistake", sitting in the back of a PI's mind, gives them permission to less rigorous, less exacting, and push that data out even sooner. My lawyer friend says that I'm a cynic.