When it comes to inherited scholarly categories and taxonomies, a prominent strand of modern American tort scholarship pursues a particular kind of deflationary agenda. The First and Second Restatements divided the law of negligence into sub-rules distinguished by spurious differences (for example, the section on “type of negligent acts” distinguished between “Use of Incompetent or Defective Instrumentalities” and “Want of Preparation”). The Third Restatement combined many rules that could be brought under a more general description – the laundry list of types of negligent acts has been radically pruned, leaving just a handful, such as “negligent failure to warn”.
Because the mission of the Restatement is to organize concepts latent in the common law, it is understandable that subsequent generations of reporters will see common themes between categories that were overlooked by their predecessors (and it is also possible that the law itself might evolve over time towards fewer principles as courts eliminate ad hoc categories). But reducing the number of rules, or principles, in the common law is not an unalloyed good. Debates still rage over whether the Restatement has, for all intents and purposes, removed duty as an element of the prima facie case in negligence in most cases of personal injury or property damage, and if it has, whether that move was salutary. Last year I reviewed for Jotwell an article by Prof. Stephen Sugarman calling for the merger of battery – an intentional tort – into negligence.
And now comes a proposal from Professors Ken Abraham and Leslie Kendrick to merge Chapters 3 and 7 of the Third Restatement, so that, instead of two general categories of duty in connection to physical harm, there will be just one rule of negligence for risk creation and there would be no need for a rule concerning affirmative duties. I will review Abraham and Kendrick’s arguments for the merger, suggest a few criticisms of their arguments, and conclude by evaluating the costs and benefits of pursuing yet another round of doctrinal deflation.
Abraham and Kendrick’s argument – which they recognize as somewhat radical – is that the category of “affirmative duties” ought to be abandoned. Their recommendation, in terms of the Restatement, is that §§ 37 – 43) be eliminated, and any claim by injured victims for physical harm that would have arisen under these sections come under the relevant sections elsewhere in the Restatement. Since the article is mostly a critical project, Abraham and Kendrick do not address where the orphaned causes of action would go, and this should not be held against them. Presumably, they would find a home in Chapter 3 (“The Negligence Doctrine and Negligence Liability” or Chapter 9 (“Duty of Land Possessors”).1
Abraham and Kendrick’s argument is simple and familiar to anyone who has tried to teach torts: The line drawn between act and inaction is vague, and because it is vague, the doctrinal rules that purport to rely on that line are either over-inclusive or under-inclusive. As a descriptive matter, I agree with Abraham and Kendrick, and a brief tour of their argument will induce, I suspect, in many torts professors a familiar sense of frustration with the common law rules in this area.
The Third Restatement preserves a categorical distinction between duties grounded on risks of physical harm “created” by the defendant (§ 7) and those risks of physical harm not created by the defendant (§ 37). I will note in passing that Abraham and Kendrick organize their argument by comparing § 7 and § 37, but that § 37 deals with risks of emotional harm not created by the defendant as well as with risk of physical harms. Their focus only on the treatment of affirmative duties in relation to risk of physical harms leaves open the question of how their analysis would apply to “pure” negligent infliction of emotional distress unconnected to physical imperilment (e.g. Section 47(b)). As an initial matter, I will only address the topic they have chosen for themselves – affirmative duties in relation to physical harm.
In tort law, it is commonplace for terms used frequently by laypeople to turn out be vague when those terms are put to work in tort. The “reasonable” person is a prime case in point. Modern doctrine, it could be argued, usually handles the problem of vagueness is ways that do not produce unnecessary complexity. One way tort law deals with vagueness is to delegate the final act of line drawing to the factfinder. Not so with the line between risk creation and risk non-creation. The line is drawn as a matter of law, note Abraham and Kendrick (P. 53), and the conclusion that a risk was created by the defendant – or not – has significant consequences for the parties. If the court holds that the risk was created by the defendant then it falls under the “standard” rules of negligence for physical injury, which impose a very broad duty on the defendant. As the Third Restatement says in § 6, comment f, “[i]n cases involving physical harm, courts ordinarily need not concern themselves with the existence or content of this ordinary duty. They may proceed directly to the elements of liability.” However, the reverse is true if the court determines that the defendant did not create the risk. Unless the plaintiff can demonstrate that the risk falls into one of three limited categories, the defendant will be held to have had no duty to reduce the risk at issue, or mitigate any harm it may cause.
Abraham and Kendrick claim that there are descriptive and normative dimensions to the vagueness problem described above. (P. 10.) Yet, according to Abraham and Kendrick, the point of their article is not to criticize the ultimate holdings of any of the many courts that have applied the distinction between risk creation and risk non-creation. They are careful to emphasize that their project does not entail a normative argument that any particular case (including such chestnuts as Tarasoff orMoch) ought to have been decided differently. Their critique is conceptual. Abraham and Kendrick believe that there is no bright line separating affirmative duty cases from negative duty ones. Dispensing with the distinction might bring valuable clarity to our legal categories but it would not necessarily change the outcome of any of the cases they discuss.
The conceptual critique can be seen best in their treatment of cases involving the duties owed by landlords to tenants and other entrants in regard to criminal assault. (Pp. 12 – 14.) These cases, such as Kline v. 1500 Mass Ave. Apartment Corp., are treated as affirmative duties by the Third Restatement in § 40. Abraham and Kendrick’s problem with Kline is not with the outcome – a finding that there was a duty – but with the rationale for the duty provided under § 40. § 40 describes a situation where (1) the defendant did not create the risk that harmed the plaintiff but (2) there was a special relationship between the plaintiff and the defendant. Abraham and Kendrick’s conceptual critique is twofold: first, the Restatement’s reasons for holding that a special relationship exists between landlords and tenants and other entrants are conclusory and, second, that the conclusion that a landlord doesn’t create risk in cases like Kline is question-begging. Abraham and Kendrick think that the landlord, by maintaining the property, creates the opportunity for criminal assault on that property (“landlords . . . create the very conditions under which risk to the individuals with whom they deal may arise” P. 52.) and hence participates in creating the risk of criminal assault.
If the landlord in Kline created the risk, then the case should have been analyzed under § 7 or § 39. Under § 7, the landlord would owe a duty if his or her unreasonable conduct created the risk of criminal assault. Under § 39, the landlord would owe a duty because, despite the exercise of reasonable care, the relevant conduct created a continuing risk of physical harm that the landlord could prevent or minimize. In neither case would the court in Kline have to resort to finding a special relationship between the landlord and the victim. According to Abraham and Kendrick, given the breadth of § 7 and § 39 (negligent risk creation and non-negligent prior risk creation), duties based on special relationships (§ 40) end up being a fifth wheel.
Abraham and Kendrick’s treatment of the third category of affirmative duties, the duty to perform a gratuitous rescue reasonably (§§ 42 – 44) is extremely brief. This is probably because they see these duties as clear cases of risk creation, and therefore not true cases of affirmative duty. The bulk of their argument is devoted to showing that despite its many subsections, only one special rule for affirmative duties is needed to explain the doctrine – the rule concerning duties arising from prior risk creation (§ 39).
It must be observed that Abraham and Kendrick choose a rule for affirmative duty that is often viewed as having the most limited scope of application, as compared to the other two rules (affirmative duties based on special relationships and the duty to perform a gratuitous rescue reasonably). And although they concede that courts don’t actually hold that landlords in the sort of cases discussed above are found to have affirmative duties under § 39, they insist that this is a mistake. Abraham and Kendrick point to Illustration 1 of § 39 to prove their point. This illustration describes a golfer who, after carefully scanning the course and carefully hitting a long drive, sees a stranger walking towards the point where her ball is likely to land. (P. 17.) There is a duty to warn says the Restatement, even though the golfer was careful in every respect. If the golfer has a duty to warn because she non-negligently created a risk which imperils a foreseeable victim, then the landlord in Kline also owes a duty, and for the same reason.
The overall conclusion of Abraham and Kendrick’s article is that almost all of the cases that the Restatement breaks off from §7 are really instances of what they call “prior risk creation” described in § 39. For Abraham and Kendrick, Chapter 8 contains two mistakes. The first is its failure to see that if it needed to exist at all, it would only need one rule – the duty to take reasonable precautions to protect others from the consequences of prior risks one has created non-negligently. The second mistake is to fail to see that there is no need to have a special rule for affirmative duties based on prior risk creation at all. In other words, § 39 can be collapsed into § 7 (and vice versa). They say this quite clearly on P. 18.
“Prior risk creation” threatens to transform most negligence cases into affirmative duty cases. Many standard negligence cases assume that the defendant had a duty to anticipate that his conduct might impose risks on others and to take reasonable precautions to prevent this. If this constitutes an “affirmative” duty of anticipatory risk reduction – if it essentially turns every case into a rescue case –then there is no distinction left between negative and affirmative duties.
This second mistake is based on the naïve belief that tort law should (and can) distinguish between conduct and non-conduct. While some might think that the line exists but is vague, Abraham and Kendrick don’t think there is a line at all. They assert: “All conduct creates some risk of harm” (P. 48, emphasis added.) The real meaning of this statement is that by choosing any course of action (including inaction), one is engaging in risk producing conduct, since at every moment one is potentially, no matter how remotely, in a position to affect the risk of harm faced by another. The “conventional duty of care [§ 7] often involves minimizing or protecting another against a risk of harm whose source is not the defendant’s negligence” (P. 45.) Of course, in much of a person’s daily life the practical opportunity to reduce the risks faced by others is so minimal as to be almost invisible, but that does not undermine the point that, in theory, there is always a potential opportunity to mitigate another’s experience of risk. The choice to do nothing is “conduct”. The choice to do something is “conduct”. All choosing is conduct, and since “all conduct creates some risk of harm,” every choice may, in theory, may create a risk to which another is exposed.
The viewpoint that “all of life is conduct and all conduct produces risk to others” is illustrated by Abraham and Kendrick with the following example. Driver, through no fault of her own, is confronted with a pedestrian entering into the street. Sidewalk User sees the pedestrian at the same time. Abraham and Kendrick insist that there is no reason to evaluate Driver’s response to the pedestrian under § 7 (with all the attendant issues of comparative fault) and Sidewalk User’s response under § 39. They have no objection to the likely outcome that today’s doctrine would produce (Driver likely liable; Sidewalk User likely not liable) but they insist that, since Sidewalk User’s “conduct” potentially created risk (since Sidewalk User could reduce the pedestrian’s risk of injury) the question of duty should be analyzed under the part of the Restatement at the question of duty for Driver.
The elegance of Abraham and Kendrick’s approach is that it deflates so many categories with a single blow. Of course, there is no reason to have three exceptions to the no affirmative duty rule covered in Chapter 8: Once a defendant’s ex ante opportunity to reduce another’s risk of injury is deemed to be an act of potential risk creation, the obligations imposed by special relations, gratuitous undertakings and the non-negligent creation of a continuing risk can be fused into a single class of conduct. Further, since the test of conduct is purely based on circumstance – whether one is in a position to reduce the risk of injury to another – any focus on conduct prior to the decision to reduce risk to another is irrelevant, and so the distinction between conduct and non-conduct at the point of that decision is erased.
Obviously, it is a serious question whether the concept of “conduct” adopted by Abraham and Kendrick has much currency in everyday language or the moral conventions of the societies to which the Restatement is addressed. In moral philosophy, their position is most closely associated with act-utilitarianism. In tort theory, there is a superficial connection between their account and a simplistic version of Calabresi’s cheapest cost-avoider. After all, Calabresi did, for rhetorical purposes, hypothesize that anyone who could most cheaply reduce the cost of car accidents ought to be held liable in tort for those accidents. But Calabresi was making a point about the relationship between strict liability and negligence, and Abraham and Kendrick claim to be offering an interpretation of negligence law. Therefore, it makes sense to evaluate their proposal from the perspective of their target, which is a reconstruction of negligence in the common law. I want to conclude with a much more limited critical observation.
As mentioned above, Abraham and Kendrick do not explicitly call for significant changes in liability judgments. They are calling for a significant change in legal process. They note that, once the ad hoc segregation of a case like Pedestrian v. Sidewalk User is abandoned, the case against Sidewalk User will be handled by the same process as the case of Pedestrian v. Driver. In the modern era of the Third Restatement, that means that, unless § 7(b) is invoked, the liability decision will depend on a factual judgment about breach, causation, and proximate causation. This is the clear implication of the statement that “prior-risk-creation cases can be more simply and more accurately be seen as asking whether the defendant exercised reasonable care under all the circumstances” (P. 54, emphasis added.) Abraham and Kendrick mention that in some cases, like Pedestrian v. Sidewalk User, the court will exercise its “tools for policing the outer-boundaries of fact-finding” to routinely find for Sidewalk User. (P. 53.)
My point is not that there is any reason to distrust the capacity of courts to use their supervisory power over fact-finding to preserve the status quo’s consensus that the defendants who today are found not liable under Chapter 8 will continue to be found not liable. It is rather, to observe that it is just as likely that many of the defendants who today are found not liable under Chapter 8 will be diverted into § 7(b). § 7(b) states “[i]n exceptional cases, when an articulated countervailing principle or policy warrants denying or limiting liability in a particular class of cases, a court may decide that the defendant has no duty or that the ordinary duty of reasonable care requires modification.” It was added, of course, in response to the concern that, by adopting the view that there was a “ordinarily” a duty to all with regard to the creation of unreasonable risk (§ 7(a)), judges needed some way to block plaintiffs from asking whether the defendant “exercised reasonable care under all the circumstances” as a matter of law in some familiar classes of cases.
Even if one were persuaded that the conceptual deflation of the word “conduct” is defensible, I suspect that it will not achieve Abraham and Kendrick’s aim of removing complexity from the Third Restatement. It is just as likely that they have done nothing more than moved a difficult problem from one chapter to another.