rotmanmagazine.ca / 23in decision-making teams. Among the tactics of the best teams
are that “they work with more, rather than less, information;
they develop multiple alternatives to enrich debate; and they
resolve issues without forcing them to consensus.”
Even the U.S. Supreme Court provides evidence for the
value of dissent, as it has been found to increase what is called
the ‘integrative complexity’ of the Court’s decisions. This con-
cept is akin to divergent thinking. Integrative complexity is the
ability to see both sides of an issue and their trade-offs. One
study of Supreme Court decisions analyzed its written opinions
for integrative complexity and found that they showed more of it
when there was a dissenting opinion. When the Court’s opinion
was written on behalf of a unanimous group, it showed very low
integrative complexity; when the opinion of the Court was writ-
ten on behalf of a majority who confronted a dissenting opinion,
integrative complexity was high. It was in the latter case that
the justices showed an awareness of alternatives and were more
likely to consider both sides — at least in their written arguments.
Given the evidence that dissent stimulates the kind of think-
ing that makes for powerful decision-making, we must ask why
so many people fail to embrace it — and even resist it.
Groups and Consensus
High on many people’s list of techniques for achieving diver-
gent thinking is the idea of playing ‘devil’s advocate’ — a form of
‘pretend dissent’. The origins of the this technique lie in the prac-
tice of the Roman Catholic Church when considering a candidate
for sainthood. Understandably, the Vatican didn’t want to make
a mistake and learn later that a candidate for sainthood behaved
in ways that were not saintly. Thus, in 1587, it instituted a practice
of exploring everything negative about each candidate.
Intuitively, this technique would seem to counterbalance
a bias in favour of the candidate held by most of the decision-
makers, including the Pope. In modern uses of the technique,
the premise is still that it will challenge consensus. This was the
antidote favoured by Yale Psychologist Irving Janis in discussing
ways to counter groupthink. The greater hope is that this tech-
nique will stimulate some form of divergent thinking, the kind
that, as we have seen, occurs with dissent and is directly linked
to good decision-making. Minimally, the hope is that devil’s ad-
vocate, like dissent, will encourage teams to evaluate risks and
benefits with equal vigour.
Now for the bad news: It doesn’t work. Available research
on the devil’s advocate technique provides support for its value
— but only when compared to groups where there is no challenge
whatsoever. Compared to these groups, it can provide benefit, as
can its sister technique, ‘dialectical inquiry’, in which a person
doesn’t just argue all the negatives of a proposed position, but
also offers a counter position. By and large, these two techniques
are better than no challenge at all; but my colleagues and I have
found something that works much better: authentic dissent.
We first conducted a study comparing the devil’s advocate
technique to authentic dissent in 1998. I had been skeptical
about the devil’s advocate approach for years. It seemed to me
to be a contrivance. Anyone can undertake an intellectual game,
but that is not the same as really thinking through all the pros and
cons of something we believe.
I had studied the power of authentic dissent for years and
had witnessed its ability to stimulate a broader search for infor-
mation, a consideration of more alternatives, the use of multi-
ple strategies, and more original thinking. Before my collegues
and I undertook our studies of the devil’s advocate, I had never
believed that you can role-play dissent and get those kinds of
reactions.
Our study was quite simple. We assembled groups consist-
ing of four individuals. In one set of groups, there was no dis-
sent; in a second set, there was authentic dissent; and in a third
set, there was devil’s advocate ‘dissent’. The groups deliberated
on a personal injury case and were to agree on compensation
for pain and suffering. They could choose any one of eight op-
tions in $75,000 increments. The lowest was $1 to $75,000, then
$75,001 to $150,000; the eighth option was over $525,000.
Each participant typed in their own position. From pretest-
ing this case, we knew that they would all pick a low compen-
sation, and they did: All took the first or second position. Each
person saw the arguments of each of the others, but in fact all