Open JohnAGoldsmith opened 3 years ago
analysis of "fabricated" from encarta corpus.
In split-affixes, we have 3 splits fabricate d fabricat ed fabric ated
(which is good) This gives us fabr ic ated, instead of the expected fabri-ic-ate-d
And we get the signatures: NULL=d e=ed=ing=ion
when I would have expected to see ic=??=?? (attached to the stem "fabr") NULL=ate=ation=s (attached to suffic "ic:32")
analysis of "fabricated" from encarta corpus.
In split-affixes, we have 3 splits fabricate d fabricat ed fabric ated
(which is good) This gives us fabr ic ated, instead of the expected fabri-ic-ate-d
And we get the signatures: NULL=d e=ed=ing=ion
when I would have expected to see ic=??=?? (attached to the stem "fabr") NULL=ate=ation=s (attached to suffic "ic:32")