“…Since, for v satisfying (1.42.1), ǫ 1 ( k i ) + v · k i − k 2 i /2 ≥ 0 for all i = 1, · · · , r, and similarly for the other excitations (excluding the umklapp excitations, which we have shown to be absent), it follows that the restriction of H N,L − E 0 N,L + v · P N,L to R c,d N,L is positive, which is (1.36.2). The fact that the subspace R c,d N,L does not shrink to the empty set in the thermodynamic limit, i.e., (1.36.3), is easily seen to be true for any r ≥ 1, see [Wre15a].…”