Arnold Kling  

On Being Certain

Obama on How Markets Reduce Ra... Forecasting...

Robert Burton writes,

The message at the heart of this book is that the feelings of knowing, correctness, conviction, and certainty aren't deliberate conclusions and conscious chohices. They are mental sensations that happen to us.

That is from the conclusion to On Being Certain.

Somehow, it seems to relate to the idea of what Tyler calls the Hansonian journalism of Julian Sanchez.

Burton's model of conscious thought reminds me of an arcade game in which there are 10 toy horses lined up to race. Each human contestant has a squirt gun and and a target to shoot at. When a bell rings, you try to squirt your target. The closer to the center of the target you hit, the faster your horse goes. The best shooter will win the race.

For Burton, a thought or belief is like one of the horses in the game. The squirt gun is the various emotions and sensations that move that thought into consciousness.

The thoughts that win the horse race in an individual's head depend in part on how the individual's squirt-guns are configured. If certain squirt guns are stronger than others, that affects the likelihood that a particular belief will be held by that person.

One point is that the statement "I am certain of X" is nothing more than the statement "I believe X" with an emotional kicker. He suggests that we try to drop the emotional kicker.

My guess is that the book is somewhat of a catch-22. The lower the likelihood that one reads it, the greater would be its value to that person.

Comments and Sharing

COMMENTS (16 to date)
Matthew C. writes:

Excellent stuff, Arnold. This can really help bring about some epistemic humility, which is a precursor to distinguishing reality from one's existing prejudices.

Bork O. Balmer writes:

The subconscious mind does most of our thinking. This is an advantage for those of us in politics, since it is easier to sway the subconscious than to appeal to reasoning.

I have achieved particular success on environmental topics, which seem to be particularly prone to emotional appeal.

Clayton writes:

Wow, that last paragraph is wildly insightful!

Paul Zrimsek writes:

Feelings of knowing, correctness, conviction, and certainty might not be deliberate conclusions and conscious choices. Then again, they might be. Gosh, who can tell?

Mattyoung writes:

Too difficult a topic to grasp in a single post.

Snark writes:

One point is that the statement "I am certain of X" is nothing more than the statement "I believe X" with an emotional kicker. He suggests that we try to drop the emotional kicker.

Yet, many of us accept that the domain of our physical existence is governed by certain immutable laws of nature. No amount of emotionally-reinforced belief alters this perception in those who we would describe as rational. I would also argue that there are moral absolutes of which we can be certain, regardless of what we may believe. The statement, “I am certain of X”, thus distinguishes itself and remains valid in either context.

“I believe X” with an emotional kicker sounds like a subjective psychological belief, about which Bayesian analysis enables us derive degrees of certainty before we take a leap of faith.

Dr. T writes:

"The subconscious mind does most of our thinking."

No, the subconscious mind does most of our dreaming and emoting.

Some people confuse simple decision-making (should I get a cola or a root beer) with thinking, but there's almost no thinking involved. One's subconscious desires often determine the choice.

Thinking requires that we give directions and information to our conscious minds. Otherwise, we're just daydreaming. We think about what we write in a blog, we think about whether parallelism is appropriate, and we think about whether an argument will be persuasive. There is little input from the subconscious in those processes. Problem-solving also uses the conscious brain almost exclusively. The subconscious has little to say in determining which disease best fits a patient's symptoms, history, and lab test results.


I agree with Snark about the "I am certain of X" statement. It is quite appropriate in many circumstances and cannot be replaced by "I believe X (because it fits my emotionally distorted view of the universe)."

steakknife writes:

See A.J. Ayer on emotivism.

Mattyoung writes:

Did I get that Arnold's squirt gun analogy concerned neuronal firing?

In the primitive animal, before fish, for example, the olfactory function would be directly connected to muscular function, so the identification of food reaches a threshold and the grasping action is triggered [Knowing]. The basic triggering function would survive evolution but be heavily inhibited by the limbic system, an evolutionary wrapper probably evolved to control the urge to grasp. But the function still is a requirement, without which we would not be able to act.

Matthew C. writes:

Yet, many of us accept that the domain of our physical existence is governed by certain immutable laws of nature. No amount of emotionally-reinforced belief alters this perception in those who we would describe as rational.

Actually, I find that self-described "rationalists" are among the least able to discern how their neurons lead them around by their nose (although certainly capable of seeing that phenomenon in OTHER groups of people).

Having accurate models of reality requires taking into account all the facts and "rationalists" tend to discard vast swaths of them because they do not fit into their belief system. "Rationalism" is like all the other "isms", political philosophies, and religions -- reality takes a back seat to group membership, group norms, group "heroes" and "villains", group mythology and the like.

The best thing you can do to become more aligned with truth is to see how your thought processes, beliefs, and group affiliations are very much like those practiced by every other human being on earth. Especially those that are flavored with neurological "certainty" -- feelings of being "less wrong" then those benighted fools with different beliefs.

fundamentalist writes:

I don't get why this is considered deep or important. I haven't read the book, but based on reviews it seems that all Burton is saying is that certainty is an emotion. So?

The important question is how did you arrive at that certainy? Did you take your someone's word for it? Or did you have solid empirical evidence or good reasoning without a bunch of fallacies? And did you consider alternative evidence? Everyone is certain about a lot of things, but we know that most people have no reason to be certain. Most certainty is built on prejudices.

Even when you arrive at certainty by evidence and logic, intelligent people can disagree because they place different weights of importance on evidence and aspects of logic. Still, the most common source of failure among academics and really really smart people is their clinging to prejudices. Hazlitt wrote a great book on it called "Thinking as a Science."

The only cure for prejudice is to work at getting into a mental attitude in which you don't care about the results of your investigation; you only want the truth. That is very difficult for most people to do, but even harder for intelligent people.

Snark writes:

@Matthew C:

Certainty is gained through repeatable experience, in which case realist might be more appropriate than rational. From a philosophical standpoint, however, I would argue that we cannot ascertain absolute truth from the facts alone, because truth is multidimensional (i.e. logical truth, necessary truth, analytic truth). I'm probably more of a rationalist than a realist when it comes to seeking the truth. Empiricism is too mundane of an existence for those of us who believe life has purpose.

fundamentalist writes:

Matthew: "Having accurate models of reality requires taking into account all the facts..."

How do you construct models without theory. Facts don't interpret themselves. In the first place, there are too many facts, many of them contradictory. Someone has to select the facts that are appropriate for the model, which requires some sort of reasoning about which facts are important. Then someone has to organize the model according to a theory of how the facts relate to each other. As Arnold has pointed out about macro modeling, the models do nothing but reflect the prejudices of the modeler.

In other words, even though empiricists might deny it, they have a boat load of rational baggage that works in filtering and selecting and organizing the data into a model. As far as I know, no one has yet discovered self-organizing data.

rhhardin writes:

I'd recommend Wittgenstein via Stanley Cavell, either _The Claim or Reason_ or the essay ``The Argument of the Ordinary'' in _Conditions Handsome and Unhandsome_.

Most mental terms are terms used in accounts, not names of things ever present.

This matters a lot if you're going to claim to look for them one way or another.

Mattyoung writes:

I happen to have a degree in neurology, physiological psychology we called it.

But the point here is economics. What arrangement of goods, revealed to all would generate the certainty function? Because if the certainty has a biological basis, then we expect its measure or its tripping point to be aligned within a species. A collective observation of the environment, when triggering an certainty, would do so for the herd. So this puts you on a path to economic theory as constant certainty.

Biology tells us to expect a counter certainty dampening inhibition. It gets you a band of certainty, coherent across the herd, a coherence between micro and macro. Evolution figured out, early on, how to use this certainty band to get efficiency of scale from a herd, they move synchronously; maximizing mobility and herd cohesion.

So, when we go shopping, the arrangement of goods tells each individual a lot about the average person, who therefore are not strangers, but a herd sharing their own certainty measure with the group.

Bill D writes:

Read the book based on this post.

It would be nice to see how to apply some of the insights from the book to managing one's own certainty and working with others.

Perhaps a study of how certainty and personality types interact. Are rationals more likely to be certain than idealists?, etc.

Comments for this entry have been closed
Return to top