The author's point is that anytime the butterfly effect is used in this way (identifying the singular cause of a chain) it insinuates that we can always pinpoin

The author's point is that anytime the butterfly effect is used in this way (identifying the singular cause of a chain) it insinuates that we can always pinpoint such a butterfly if we look very hard or that there is one single such butterfly causing the chain of effect in the first place. However Lorenz (at least per the author) tried to make the exact opposite point. Among many many variables if one slightly changes (e.g. flap or no flap of a butterfly's wings somewhere) it can affect a lot, but we don't know which one. I don't know it's wrong per say to use it as most people do but I see the author'a point. It's often (falsely) used when there is a chain of ovents and we know which thing caused that chain.