O, what to do with instincts and lexicography...

I'm beginning research on how to study a noun's semantic potential and came across Atkins and Rundel's (2008) Oxford Guide to Practical Lexicography and decided to jump to chapter 5, titled "Linguistic Theory meets Lexicography". Here's the first couple sentences, which already have my mind humming.

"By the nature of the work they do, lexicographers are applied linguists. Yet many people working in the field have no formal training in linguistics. Does this matter? Our experience as editorial managers suggests that good lexicographers operate to a large extent on the basis of instinct, sound judgment, and accumulated expertise" (Atkins and Rundel 2008:130)

While people making dictionaries of modern languages may be able to get away with this reliance on instinct, we who study ancient languages are not so fortunate.

  • What precautions do we take?
  • How much do we trust our (next-to-nothing?) intuitive powers?
  • How did lexicographers of the past account for this?
  • Or did they make their lexicons in much the same way as current lexicographers of modern languages?
    • If so, should we question their conclusions with more than the regular amount of skepticism that should accompany any lexical semantic investigation?

Don't be afraid to chime in. What other questions should be asked?