4. The negation of the conclusion is easy to prove. When a=b, it must be that (a−b)∣[p(a)−p(b)], so ∣a−b∣⩽∣p(a)−p(b)∣. Taking a,b as xi,xi+1,i=1,2,⋯,n (with xn+1=x1), we get ∣x1−x2∣⩽∣x2−x3∣⩽⋯⩽∣xn−1−xn∣⩽∣xn−x1∣⩽∣x1−x2∣. Therefore, ∣x1−x2∣=∣x2−x3∣=⋯=∣xn−x1∣. When n⩾3, if i makes xi−1−xi and xi−xi+1 have opposite signs, then from the above equation, we know xi−1−xi=−(xi−xi+1), thus xi−1=xi, which contradicts the problem's condition. Therefore, for any i, we have xi−1−xi=xi−xi+1. At this point, if x1⩾x2, then x1⩾x2⩾⋯⩾xn⩾x1, so x1=x2=⋯=xn, which is a contradiction. If x1<x2,x1<x2<⋯<xn<x1, it is also a contradiction. Thus, the proof is complete.