Here, we have two words from Hachijo in the same cognate sets, but they differ (!). We can argue that for correspondence patterns, it is impossible for strictly cognate words to differ. So a preprocessing can in fact arbitrarily decide for one of them.
The pattern is difficult to detect. In CoPaR, only one word of the two is used, the other word is ignored, but this will shrink the alignment, since the one word causes all the gaps.
We might need some basic checks whether a correspondence pattern analysis is useful, since I detected one pattern that causes huge problems:
Here, we have two words from Hachijo in the same cognate sets, but they differ (!). We can argue that for correspondence patterns, it is impossible for strictly cognate words to differ. So a preprocessing can in fact arbitrarily decide for one of them.