What is negligence? Our answers to this are frequently misleading, because we turn all our attention on to what doctrine says rather than asking how it is actually used. So we routinely talk of personal liability, even though we know very well that individuals (as opposed to organisations, typically insurers) do not pay damages. We think of negligence doctrine as if it were applied automatically and without bias, when in fact the complex insurance arrangements involved have rather decided biases. And we treat the development of the law purely as a matter of evolving judicial thought, when in fact legislatures and insurers also routinely modify the system in response to new realities. The result is that fundamental change has occurred under the very noses of theorists, who still tell us that negligence holds individual defendants responsible for their wrongdoing (it does not) or that the economic effect of tort rules is to deter defendants (there is not much reason to think this is so). Why are the leading theoretical justifications of negligence – corrective justice, responsibility theory and economic theory – so dependent on myths? Is it because the myths do not matter? Or is it that, in fact, we have no good justification for the system as it works in practice at all – we have no good theory of why negligence makes sense because it does not, in reality, make sense?