The concept of negligence dominates tort law. Most tort cases are about negligence. Much tort law scholarship over the past several decades has been about the meaning of negligence. The new draft Restatement (Third) of Torts: General Principles ("Discussion Draft") devotes the vast majority of its first volume to negligence.1 And the idea of negligence as a liability standard is highly attractive to both the courts and commentators.2
All the attention that negligence receives is not surprising, given the unattractiveness of the alternatives. Imposing liability only when the injurer intended harm seems unduly limited, in that it absolves injurers of responsibility for harm caused by less blameworthy, but still wrongful conduct. Yet, under many circumstances, strict liability seems unduly broad, in that it risks imposing liability on innocent parties and depressing the level of desirable activities. Consequently, some form of negligence standard (however defined in particular) seems either to get it about right, or in any event to be a suitable compromise between the twin extremes of too little and too much liability.
The negligence standard is so much a focus of tort theory, however, and negligence cases occupy so large a proportion of all tort claims, that it is too easy to ignore how unusual negligence truly is among tort law's standards of conduct. Despite the extensive efforts of legal scholars to define negligence and to explore the relation between negligence and other standards of conduct, the character of negligence liability remains incompletely recognized. In my view, close examination of the negligence standard reveals that it is more troubled than its apparently central place in tort law implies. Far from being an appropriate default rule to be used when we are unsatisfied with the alternatives, the negligence standard is often flawed even in the ordinary cases involving liability for physical damage that are at its core. These same flaws render negligence an even less appropriate standard in most cases involving intangible loss, where at least until now it has been employed only in exceptional cases. To pursue these points, I begin by examining the negligence standard as it is applied in cases involving physical harm. Twentyfive years ago, James Henderson, arguing in a powerful article that the courts were applying the negligence standard in a manner that constituted a "retreat from the rule of law,"3 began to uncover the distinctive character of negligence in such cases. Henderson relied heavily on the distinction between rules that are "polycentric," or open-endedly dependent on a variety of incommensurable factors, and those that are formal.4 I shall employ a distinction that looks at this problem in a somewhat different light. The cases readily divide, I think, into two categories: those in which the finder of fact must in effect create a norm in order to determine whether the defendant was negligent, and those governed by a pre-existing, independent norm. I contend that negligence is a far less satisfactory standard of conduct in the former set of cases than in the latter, precisely because of the problems associated with norm creation.5
Turning then to cases involving intangible loss, I argue that several of the well-known exceptions to the general rule that there is no liability in negligence for intangible loss reflect this principle of demarcation. In the exceptional cases in which there is negligence liability for "pure" intangible loss, there is often a preexisting, independent norm that serves as the reference point for the negligence determination. And there is rarely liability in negligence for intangible loss in the absence of a pre-existing, independent norm that defines the applicable standard of conduct. In light of this principle, I conclude that there should be a restrained-and certainly not expanding-future for the negligence standard, whose flaws even in physical injury cases should be more clearly recognized, and whose application to new claims for pure intangible loss should be resisted. …