The Causal Mechanism Theory of Legal Causation

• Published date: 01 March 2021
• Author: Peter Bach‐y‐Rita
• Date: 01 March 2021
• DOI: http://doi.org/10.1111/raju.12305

© (2021) John Wiley & Sons Ltd.

Ratio Juris. Vol. 34 No. 1 March 2021 (57–73)

The Causal Mechanism Theory

of Legal Causation

PETER BACH- Y- RITA

Abstract. Theories of legal causation that identify the concept with probability- raising generate

false positives where the defendant’s conduct raises the probability of harm but brings about

that harm in the wrong kind of way. Moreover, what the law seeks to deter is not conduct that

raises the probability of harm, but rather conduct that is dangerous. A legal or proximate cause

is one that harms the plaintiff through the causal mechanism that sustains the lawful generaliza-

tion that the defendant’s conduct was dangerous.

1. Introduction

The causal mechanism theory is a theory about the meaning of ordinary causal

ascriptions: It says that statements of the form “A was the cause of B” imply the

existence and operation of a law- governed causal process connecting A and B. For

example, to say that smoking was “the cause” of a victim’s death is to say that there is

a potentially repeating law- governed process through which smoking causes death,

and, furthermore, that smoking led to the victim’s death through that particular pro-

cess. (In the case of smoking, we know a causal mechanism exists because it is heav-

ily studied by medical researchers and it explains the morbidity of smoking.) The

causal mechanism theory is a live and vigorously debated theory within the fields

of philosophy, psychology and cognitive science (e.g., Ahn and Kalish 2000, 199– 201;

Fugelsang and Thompson 2003, 800– 2; Glennan 1996, 49– 71; Strevens 2007). In the

psychology literature, it is typically distinguished from the “covariation” or “regu-

larity” theory, which says that subjects make causal ascriptions when cause and effect

are of types systematically seen in temporal succession, regardless of any connecting

causal process.

Carrying over the causal mechanism theory into the law has both normative and

empirical motivations. The causal mechanism theory is already widely accepted

as a theory about the meaning of ordinary causal ascriptions, making it at the very

least plausible as a theory about the actual meaning of causal language in the law.

Moreover, the purpose of this article is to show that the causal mechanism theory

is the only theory of legal causation compatible with the optimal liability rule in

a decentralized tort system. If that is correct, then the causal mechanism theory is

unique in being both descriptively adequate and welfare- optimal. And for this rea-

son, I believe that the causal mechanism theory of legal causation is quite likely to

be true.

Peter Bach-y-Rita

58

Ratio Juris, Vol. 34, No. 1© (2021) John Wiley & Sons Ltd.

While the primary goal of this article is to defend the causal mechanism theory as

a theory about the meaning of causal language in the law, a secondary goal is to bring

about a partial reconciliation between supposedly competing academic theories of

liability. Theories of law and economics are often characterized as giving central im-

portance to the role of optimal deterrence while depreciating the role of causation,

while theories of corrective justice place greater emphasis on compensating victims

whose injuries are caused by a defendant’s wrongful conduct, whether or not the

cost thereby imposed is optimal from the standpoint of efficiency and welfare max-

imization. This article takes seriously the assumptions of both schools of thought. I

assume that optimal deterrence is an important goal of the tort system, and also that

tort liability must be coupled with causal responsibility for the victim’s injury— two

plausible assumptions that, when taken together, provide two pillars of support for

the causal mechanism theory of legal causation.

2. Dangerousness in the Tort System

A basic tenet of the tort system is that we should not deter conduct unless it is dan-

gerous. This idea is imbedded in the very concept of negligence, which could be

thought of as unreasonable inattentiveness to the possibility that one’s actions will

expose others to danger. It is reflected in the “scope of the risk rule,” which requires

the plaintiff’s injury to “arise from the risk of danger created by” the defendant’s

negligence.1 Discussing the role of luck in the law, Arthur Ripstein (2008, 61– 76) has

argued that “the basis for [negligence] liability is the dangerousness of [the tortfea-

sor’s] conduct, not its defective moral character.” Similarly, strict liability is reserved

for categories of conduct that are dangerous by their very nature, including “inher-

ently dangerous” conduct,2 the sale of “unreasonably dangerous” products,3 and the

keeping of animals with “dangerous propensities.”4

One goal of the tort system, among competing priorities, is to deter conduct that

is dangerous (recognizing that this is not necessarily best achieved by sanctioning

all instances of dangerous conduct). Moreover, the tort system should, and to a large

extent does, deter conduct precisely to the extent that it is dangerous. This further

claim is supported by the basic structure of the tort system, which imposes deterrent

cost only when tortious conduct actually causes some injury, and then, by attaching

a deterrent cost equal to the victim’s injury. As a result, over time, the total deterrent

cost imposed on tortfeasors should emerge as a quantitative measure of the danger

of all negligent acts— subject to the fact that the actual amount of tort damage that

actually occurs over time is a function of both the dangerousness of tortious conduct

and how lucky or unlucky victims and potential tortfeasors turn out to be. Moreover,

from an ex ante standpoint, the expected deterrent cost imposed on any given act of

tortious conduct is equal to the total expected damage proximately caused by the act.

If the goal is for the tort system to attach a cost to tortious conduct that approaches

a true quantitative measure of its dangerousness, then a pressing problem for the

tort system is to avoid imposing deterrent cost when the plaintiff’s injury does not

1 Great Lakes Docks & Dredge Co. LLC v. Louisiana, 624 F.3d 201, 213 (5th Cir. 2012).

2 Majestic Realty Associates, Inc. v. Toti Contracting Co., 153 A.2d 321 (1959).

3 Restatement 2d Torts § 402A (1965).

4 Restatement 2d Torts § 509 (1965).