Don't think so. Kernighan et al. are the first to say code should be clear first and made fast only if it matters. This is clear code. It may (or may not) be faster than an array-based version but that's not the point of the exercise in K&R -- the point is to teach how pointers work.