Compared with the elongate bodies of shrimps or lobsters, crabs are characterised by a compact body organisation with a depressed, short carapace and a ventrally folded pleon. The evolutionary transformation from a lobster-like crustacean towards a crab is called ‘carcinization’ and has been interpreted as a dramatic morphological change. Nevertheless, the crab-shape evolved convergently in a number of lineages within Decapoda. Accordingly, numerous hypotheses about internal and external factors have been presented, which all try to explain these frequent convergent carcinization events despite the seemingly fundamental changes in the body organisation. However, what a crab is lies greatly in the eye of the beholder and most of the hypotheses about the lobster/crab transformation are biased by untested assumptions. Furthermore, there are two meanings of the word ‘crab’ within decapods: one, the phylogenetic meaning, refers to the clade Brachyura; the other, more general and typological use of the word crab, describes decapods with a certain body shape. These two meanings should not be confused when the issue of carcinization is discussed. Here, I propose a definition of what a crab is, i.e. what is meant when we speak about carcinization. I show that not all Brachyura are crabs in the typological sense. Carcinization occurred at least twice within the clade. Among Anomala there is further evidence that crab-shaped Lithodidae derived from a hermit-crab like ancestor. Carcinization is not restricted to Anomala plus Brachyura (Meiura) but is also found in Achelata, namely in slipper lobsters. A deconstruction of the crab-shape reveals that parts of it appear in various combinations among all decapod groups. Only a certain threshold of number and quality of crab-features makes us call an animal a ‘crab’. This reveals that carcinization does not involve such dramatic changes in morphology as has been suggested. Moreover similar alterations of body shapes appear frequently in other crustacean taxa and in various animal groups as diverse as sharks and sea urchins. Hence morphological constraints, macroevolution, trends, tendencies, or underlying synapomorphies of any kind are not necessary assumptions for the explanation of the evolution of crabs.