In recent years evidence has accumulated demonstrating that dogs are, to a degree, skilful in using human forms of communication, making them stand out in the animal kingdom. Neither man's closest relative, the chimpanzee, nor dog's closest living relative, the wolf, can use human communication as flexibly as the domestic dog. This has led to the hypothesis that dogs’ skills in this domain may be a result of selection pressures during domestication, which have shaped dogs’ skills tremendously. One hypothesis, the so-called by-product hypothesis, suggests that dogs have been selected against fear and aggression and as a by-product this paved the way for the evolution of generally more flexible social cognitive skills, which surpassed those of their ancestor, the wolf. Another hypothesis, the adaptation hypothesis, has claimed that dogs may have been specifically selected for certain tasks for which using human forms of communication was necessary. As of yet, the mechanism underlying dogs’ understanding of human forms of communication is not fully understood. We argue here that understanding the mechanism involved will also shed light on possible evolutionary scenarios. We argue that the evidence to date suggests that dogs’ understanding of human forms of communication may be more specialized than was predicted by some and may be best explained as the result of a special adaptation of dogs to the specific activities humans have used them for.