The individual words are being sorted rather than the list of words, and N here would correspond to the number of words. The sorting of each word would traditio

The individual words are being sorted rather than the list of words, and N here would correspond to the number of words. The sorting of each word would traditionally be O(M log M) where M is the number of characters in the word. Note that under certain assumptions it's probably possible to sort a word faster than that because we know extra information about the elements being sorted (that is, the letters of the alphabet form a totally ordered set of small cardinality). Specifically, there exist algorithms like counting sort which exploit this extra information.