As we all know quite well, and as is discussed in other posts here, any two people will always disagree on some level on any interesting topic or decision. As we discuss in the post on “Leadership and Hierarchy”, the result is the need for hierarchy in order to resolve these differences and take the necessary decisions to move the organization forward. Several problems of hierarchical decision-making structures were discussed in that post, one of which was the decline (or disappearance) of learning. The role of opinions in decision-making was a critical factor. Why does learning disappear when opinions are tolerated in decision-making?
In the classroom, I like to ask: how many different opinions are there, here in the room, on any interesting issue or topic. The answer comes back quickly from participants: as many as there are people in the room. I then pose the question: logically, how many of those must be wrong? The answer is quickly supplied: at least all but one. At which point I pose the question: what is the likelihood that the final opinion is also incorrect? Answer: very high, given that the exercise would still work even if we had included all of humanity in the room. On this basis, we are quite confident in stating that, with few exceptions, opinions are incorrect.
When we add to this the fact that people tend to love their opinions, we begin to see the danger of using them in decision-making. When your strongly held opinion meets compelling data and/or logic which refutes it, which wins: the opinion or the data? Answer: few people have difficulty pointing out why any set of data is not sufficient to override their opinion. They often point out that the data is wrong (which is certainly true), it’s from the past (so it’s probably not relevant since we are making a decision for the future), it’s incomplete (and unfortunately the data which would’ve refuted the opinion isn’t among the data provided), and it’s likely compromised (the people who provided it probably have ulterior motive – were they “finance people”?). On this basis, it is perfectly reasonable to ignore the data and keep the opinion. This does not change the above logical argument which demonstrates that each opinion is almost certainly wrong.
The problem is the emotional connection we have with our opinions, and the fact they are “supported” at the sub-conscious rather than conscious level of our mental awareness, and are therefore a blend of both objective (what is true) and subjective (what we would like to be true). Because opinions exist at the subconscious level, we are unable to separate whether, or the degree to which, the opinion is based upon what is true (objective) or is based on what we would like to be true (subjective). Moreover, because they are “supported” at the subconscious level of our awareness, we are unable to explain fully why we hold them, we simply know that we do. With the emotional connection, we will stick to them long after the evidence and logic would have caused us to move on. When two of us find our opinions in conflict (which is inevitable when we get to a certain dimension on any interesting issue), we have no way to resolve this since neither of us is fully aware of why we have the opinions we have, nor can we provide objective data or logic to “win” the debate, and we become very emotional when our opinions are challenged because they reflect the world as we would like it to be (subjective). At this point, arguments ensue which sometimes escalate to silence, violence, or even murder. The one thing which is nearly certain to not occur is learning.
Consider the conceptual difference between hypotheses as used in a learning process (following the scientific method), and opinions as commonly used in business (which typically results in what we refer to as an “anti-learning” process).
Would you consider it a good day, or a bad day, to find out that your opinion, which you’ve openly espoused in public in support of a particular business decision, is not supported by the data (i.e., is wrong)? For most people, this would be a bad day. In the scientific method, is it considered a good day or a bad day when the null hypothesis is proven wrong? If the hypothesis is supported by experiment, then no real learning takes place – we simply continue with the hypothesis we already held. However, if experiment refutes the hypothesis, then it’s time to develop a new hypothesis and test it – which leads to learning. Thus, a good day (from a learning perspective) is when our hypotheses are proven wrong and we undertake to develop a new hypothesis. If decision-makers are using opinions, and are openly rejecting new data, which might have refuted the opinions and led to new learning, simply to “save face” or protect their career/position, etc., then the organization will systematically fail to benefit from new learning opportunities. In a world changing as rapidly as ours, learning is the key to long-term survival (though it doesn’t guarantee it, unfortunately). Thus, an organization which tolerates the use of opinions in decision-making is making a decision to sacrifice learning.