Emanuel Derman

All I want for New Year’s …

By Emanuel Derman
December 30, 2011

Lately it has become fashionable to disparage intuition in favor of careful statistical analysis.

Me, I’m still a fan of intuition. Intuition isn’t merely quick thinking. It’s an arrow to the heart of things. Sometimes, maybe often, it’s an arrow that misses, but when it doesn’t it’s the deepest form of perception. Once the arrow tells you where to look, slow thinking will help confirm or negate its mark. But without the arrow, there’s no target to think slowly about.

Dave Edwards recently wrote:

There are three ‘styles of knowing’:

1. Euclidean: Basic principles are clear and precise and consistent with one another; examples: Euclidean geometry, classical mechanics, von Neumann’s quantum mechanics.

2. Heuristic: Basic principles are clear and precise but inconsistent with one another; examples: chess strategies, legal systems, expert systems in A.I. using non-monotonic logic.

3. Intuitive: Basic principles don’t exist or are ineffable or tacit. Mastery usually involves apprenticeship, imitation, massive practice; examples: philosophy, mathematics, science, carpentry, language, chess-almost all human skills!

Similarly, Spinoza wrote that the highest endeavor of the mind, and the highest virtue, is to understand things by the intuitive kind of knowledge.

Therefore, I can’t get too excited about intuition’s flaws.

If you give me a choice, as a scientist or an artist, as to which kind of knowledge I’d like more of, I’ll have more intuition.

It’s all we’ve got, really.


2 comments so far | RSS Comments RSS

If my intuition (life-long best friend) isn’t mistaken Emanuel, you really aren’t wanting anything at all…for this year or the next. Flawed? Indeed! Wouldn’t have it any other way.

Best to you in the months ahead.

Posted by sullivandb | Report as abusive

A different take on those “flaws”.

If “careful” statistical analysis requires considering a ‘sufficient’ number of variables – perhaps a big, aleph lot of them – then a “not so careful” analysis could be gleaned from considering just two of them. That’s the territory that Lowy & Hood stake out in their book, The Power of the 2×2 Matrix.

A 2×2 matrix is a minimal set for figuring, if this goes here, then that goes there. No side conditions, additionally considered constraints – just A and B, do this then that happens. Which is precisely what many of us would call “having an intuitive sense” of what’s going on. Seductive, and it works until it doesn’t, but not a wild and crazy first step. If the complete problem space is a manifold, then intuition is the tangent space, and curvature gives us a sense of when “we’ll get stupid”.

Net-net: ‘intuition’ is a viable / necessary evaluative step for understanding what we “understand” about a problem. Not always right, always ‘interesting’, and quite often, enough to drive by.

Posted by zach49 | Report as abusive

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/