Two Steps Forward, No Looking Back: Confronting the Problem of the Indeterminate Plaintiff

• Author: James Zaitsoff
• Pages: 240-267

240

TWO STEPS FORWARD, NO LOOKING

BACK: CONFRONTING THE

PROBLEM OF THE INDETERMINATE

PLAINTIFF

James Zaitsoff*

A. INTRODUCTION

When a deficiency in our legal system systematically denies a certain

category of person access to fairness, justice, and compensation, imme-

diate attention is required. Such a deficiency currently exists in the area

of Canadian class action law. The problem is what John G. Fleming

identifies as the “indeterminate plaintiff,” in which “the claimant is one

of several victims, only some of whom have been injured by a single

tortfeasor but who are unable to say which one among them.”1 Current

approaches to causation deny recovery in a majority of such situations,

as it is impossible to prove, on a balance of probabilities, that a defendant

is responsible for each individual injury. As a result, claims simply fall

though the cracks.

The courts are aware of this ongoing weakness in the law of causa-

tion. In Snell v. Farrell, Sopinka J. observed that

[t]he traditional approach to causation has come under attack in a num-

ber of cases in which there is concern that due to the complexities of

proof, the probable victim of tortious conduct will be deprived of relief.

This concern is strongest in circumstances in which, on the basis of

some percentage of statistical probability, the plaintiff is the likely victim

of the combined tortious conduct of a number of defendants, but cannot

* Associate at Owen Bird Law Corporation in Vancouver. Prior to articling at

Owen Bird, James gained experience at Hanson Wirsig Matheos, a small firm

specializing in plaintiff's personal injury and class action litigation. James com-

pleted his LL.B. at the University of British Columbia in 2008 and was the win-

ner of the 2008 Harvey T. Strosberg Essay Prize.

1 John G. Fleming, “Probabilistic Causation in Tort Law” (1989) 68 Can. Bar

Rev. 661 at 679 [Fleming “Causation”].

VOL UME 5, No 2, august 2009 241

prove causation against a specific defendant or defendants on the basis

of particularized evidence in accordance with traditional principles.2

While this concern is directed towards what has been coined the problem

of the “indeterminate defendant,”3 the remarks are similarly applicable to

that of the indeterminate plaintiff.

Amendments to causal approaches should not be suggested with-

out caution. However, as Lord Bingham notes in Fairchild v. Glenhaven

Funeral Services Ltd., it would be contrary to tort principles to insist on

the application of a rule that is known to yield unfair results.4 When

innocent plaintiffs are being systematically denied recovery by the rigid

application of an ill-suited test, the courts must take another look at the

test that is being applied.

This paper proposes that a two-step test for causation be adopted in

the context of class action lawsuits to replace the traditional “but for” test.

First, a general causative connection must be shown to exist between the

tortfeasor and the class as a whole. It must be established, on a balance of

probabilities, that the defendant caused injury to at least some members

of the class. Second, the court must determine, on a probabilistic basis,

the proportion of the injur y suffered by the whole class for which the

defendant is causally responsible. The defendant is then responsible for

compensating each class member for that proportion of their damages.

The proposed two-step test would allow the law to adapt to changing

times, abolishing the requirement of a specific causal link between each

individual class member and the defendant, and adopting a less onerous

probabilistic approach to causation. This paper will analyze each of the

two steps in some detail, as well as consider the particular suitability of

the proposed test in the class action context, the broader use of statisti-

cal evidence to satisfy the proposed test, and the practical difficulties of

adopting the proposed test. The analysis and examination of various ele-

ments and aspects of the test will show that such an approach ought to

be adopted in class action cases to solve the problem of the indeterminate

plaintiff.

2 [1990] 2 S.C.R. 311 at 320, 72 D.L.R. (4th) 289 [Snell cited to S.C.R.].

3 Fleming, “Causation,” above note 1 at 663.

4 (2002), [2002] 3 All E.R. 305 at para. 13, [2003] 1 A.C. 32 (H.L.) [Fairchild].