The Special Verdict Paradox: Part I

If a jury finds that a civil plaintiff has proven each element of its claim, does it necessarily follow that the plaintiff has proven that claim?

The intuitive response that most lawyers have to this question is "yes, of course" and perhaps to wonder why anyone would even pose it. At first blush, the question seems tautological and hardly worth asking.

But, as a matter of logic, there are two ways in which a rational California jury could find on a special verdict form that a plaintiff has proved the elements of a cause of action, where the same jury would find on a general verdict form that the plaintiff has not proved the same cause of action. One of these ways is a mathematical and abstract function of probabilities; the other is more concrete.

The more abstract possibility was discussed in the California Court of Appeal's recent opinion in Ryan v. Crown Castle NG Networks Inc., 2016 DJDAR 12296 (Dec. 13, 2016). While not identified as such, the court's brief discussion is a variant of a philosophical construct sometimes known as the "lottery paradox."

To make clear the issue, consider what "general verdicts" and "special verdicts" are. On a general verdict form, a jury - which has been instructed on the elements of the cause of action - simply answers yes-or-no to whether a plaintiff has proven liability on the particular claim under the appropriate standard, usually a preponderance-of-the-evidence (that the claim more likely than not is valid). If the answer is yes, it is assumed that the jury has found each of the elements satisfied, even though the jury did not expressly state as much.

A special verdict form, however, is the reverse of a general verdict form as to findings on the elements and overall liability. On a special verdict form, a jury typically answers yes-or-no to whether a plaintiff has proven each element of the cause of action by a preponderance of the evidence. If the jury answers yes to each such question, finding each element satisfied, the law treats liability on the cause of action as proven, even though the jury did not expressly state that conclusion.

The abstract way in which affirmative answers on a special verdict form might not "add up" to an affirmative finding on a general verdict form is illustrated by the lottery paradox used by philosophers of epistemology. The lottery paradox illustrates how probabilities can, on occasion, mean that affirmative answers to specific ("element") questions such as those on a special verdict form do not necessarily equate to an affirmative answer to an overall ("liability") question encompassing the specific ones.

In the form in which it was first articulated by logician Harry E. Kryberg Jr., the paradox assumes a lottery in which 1,000 tickets were issued and one would be drawn as the winner.

With these assumptions, any rational person would provide a confident answer to the question: "Is Ticket Number 1 is more likely than not to win?" That answer is, of course, no. The ticket has only a 0.1 percent chance of winning.

The answer also is no to the question, "Is Ticket Number 2 more likely than not to win?" as well as "Is Ticket Number 3 more likely than not to win?" Because each ticket has only a slight chance of winning, the answer is no to all 1,000 questions about whether each particular ticket is likely to be the winner.

The lottery paradox is that - with clear negative answers established as to all these questions about whether individual tickets will win - it might seem rational to conclude that that the answer to the question "will one of the 1,000 tickets win?" should also be no. And yet... the answer is yes. It is certain that a ticket will win, at least under the assumptions given in the question. The sum of the individual negative answers does not "add up" to a negative general answer.

Just as with the lottery paradox, a jury's answers on a special verdict form do not, as a mathematical matter, necessarily add up to liability, when probabilities are considered.

In footnote 17 of Ryan, the court identifies the interplay between probabilities and a special verdict form as an effect that "exaggerates a logical fallacy pervading much of our factfinding jurisprudence." The court points out that "the probability of multiple independent statements being true (or multiple independent events occurring) is the product of their individual probabilities. This means that several statements may each be probably true - i.e., more likely than not - and yet their combined probability may remain well under 50 percent."

Using an example taken from a journal article, the court posits a three-element tort where each element has a 70 percent likelihood of being true. "Assuming the elements are independent of each other, the likelihood of all three being true is the product of their individual probabilities (.7 x .7 x .7), or .343 - slightly over one in three." This means that even if a jury believed that each of the three elements was more likely than not true, it nevertheless believed there was only a 34.3 percent chance that the entire tort was committed, so liability for the tort would not be found on a general verdict form. In such a situation, on a special verdict form, the jury would find each element proven and never would make that "overall" determination that there is no liability. Instead, the trial court simply would enter judgment for the plaintiff by finding that there is liability, because the jury found each element.

Ryan's hypothetical example is a variant of the lottery paradox. To make the lottery scenario closer to Ryan - with three specific questions with affirmative answers - the classic lottery paradox question could be posed as a hypothetical lawsuit where the plaintiff prevails upon proving that all the tickets in a three-ticket lottery are losers.

A rational juror would answer yes if asked (as an "element" question) whether Ticket Number 1 is likely to lose. That ticket has a 2-in-3 chance of losing.

The answer would also be affirmative as to Ticket Number 2 and Ticket Number 3, each of which are individually likely to lose. Since each ticket is likely to lose - as per findings on a special verdict form - it might appear that one should conclude from those three answers that the answer to the overall "liability" question is affirmative: The plaintiff has proven that no ticket is a winner.

This is not the case. The answer to the general question - equivalent to a finding on a general verdict form - would be no, it is not the case that all three tickets are likely to lose, as some ticket will be drawn and will win.

The existence of the lottery paradox does not mean that it is common for jurors to return a special verdict form on which they have found all the elements but would not have found liability on a general verdict form. Nor - quite importantly - does it mean that it can ever be known when this has happened.

Indeed, in its footnote, Ryan correctly identified its probability-based concern with special verdicts as an "abstract" one. For one thing, individual jurors typically do not assign a precise percentage of certainty to their views as to particular elements.

The lottery paradox, and Ryan's example, further become more complex when it is considered that a jury finding is not that of a single individual making a probabilistic determination as to each cause of action, but several individuals making that determination. Even if each juror agrees that an element is proven more likely than not, one may think that is 55 percent certain and another that it is 90 percent certain.

But Ryan, and the lottery paradox, do illustrate, contrary to most lawyers' initial instincts, one way in which it is logically possible for a rational juror to find each element of cause of action proven (if a special verdict form were used) but to not find liability on the entire cause of action (if a general verdict form were used.) That is, it would not be inherently illogical for a juror to think "I am sufficiently persuaded that the plaintiff has proven each of the three elements of the tort, but, when I consider the tort as a whole, I'm not quite persuaded it was committed."

In the course of civil cases, there is no way to ever know whether any juror actually thought that way. However, there is a second - and more concrete and discernible - way in which a rational jury can find all the elements a special verdict form where it would not find liability on a general one. This second way, which is distinct from the lottery paradox, applies particularly in California and has more legal significance. I will explain it in an article next week.