The article itself addresses this in detail at the very beginning. Quoting it a bit: "This statement makes no sense in Haskell (or lambda calculus) since all

The article itself addresses this in detail at the very beginning. Quoting it a bit: "This statement makes no sense in Haskell (or lambda calculus) since all functions take one argument, and functions with zero arguments are illegal. Does that mean that the ideal function can not exist in Haskell? No, this is an unfair judgment. Uncle Bob wrote this book with an OOP (Object-Oriented Programming) perspective with examples in Java. The point is that many arguments can be hard to use, understand, and can be a smell of too many dependencies. "