I am not an expert, but perhaps this document on WordNet might be
helpful:
http://wordnet.princeton.edu/wordnet/man/wngloss.7WN.html
Pay attention to the following terms: adjective cluster, head synset,
satellite synset.
You might also be interested in running the following code:
from nltk.corpus import wordnet as wn
for i in wn.synsets('happy', wn.ADJ):
print "Processing synset: ", i
for j in i.lemmas:
print "\tProcessing lemma: ", j
for k in j.antonyms():
print "\t\tFound antonym: ", k
You will realise that the lemmas in satellite adjective synsets do not
have antonyms.
I suppose that somebody with more expertise could shed more light on
the subject.