FCP will add a shift fields filter if your source media is upper and the sequence is set to lower, or vice versa.
AS the established international standards defined an interlaced frame as field 1 (which has certain defined aspects to it) followed by field 2 (with slightly different properties) and the standards are easily available (e.g., ITU-R rec.601), no one has ever adequately explained to my why the DV format chose to encode fields the “wrong way round” or why manufacturers deliberately obfuscated the issue by inventing terminology such as up/down, top/bottom and even/odd for what was perfectly easy to understand beforehand. I hope that doesn’t sound like a rant, I don’t really care that much about it, I’m just still slightly curious after all these years…