Quote:
Is it because they all cut into the design at some point? Is it because of the very crisp cuts by scissors? I ask because it seems to me that a genuine part perf could also cut into the design.
The lack of margins is the primary reason, but not the only one.
A genuine part perf could indeed have tight margins, but without sufficient margins so as to preclude trimming, it cannot be verified as such. Same thing as with a silk paper that has the right color, thickness, ink color, and impression, but lacks a blue thread: it may very well be legitimate, but since you cannot prove it, it cannot (or should not) be sold as such (unless it is part of a multiple with stamps that clearly pass the test).
If the original clerk cut an imperf or part perf too tight, even if it was legitimate 145 years ago, but the characteristics are such that it cannot pass muster today, it falls into the category of "trimmed perforate".
The stamp has to pass the tests in order to get to the status of part perf or imperf, i.e., "climb the ladder"; you don't start with the assumption that it is real and if you can't determine its authenticity you still call it real.
The other problems with the ones I linked above are that some have the wrong ink color and several have cancel types that are all wrong. The style of circular/oval handstamp shown was not developed until late in the usage period. Since part perfs and imperfs were almost universally early printings, you would not see those types of cancels on the genuine article.
The only way any of the items I linked to could EVER be considered authentic is if they still resided on the original document and were either tied via cancel, or verified by comparing the handwriting and/or date in a manuscript cancel with those on the document itself.
After examining several thousand legitimate and faked imperfs and part perfs, you develop a sense of what looks right and what doesn't. You can tell at a glance (for the most part) whether a stamp passes or fails the smell test.
As with any other specialization, the only way you develop expertise is by looking at many, many, many good and bad examples.