The Dilettante’s Practice

How does a dilettante turn the 10,000 hour rule into a viable practice for herself?  The dilettante’s dilemma is that she is too interested in too many things to devote herself to only one of them.  She is a liberally educated vocational ignoramus.

Take heart, those of you insulted by my last remark; I walk among you, am one of you.

In one of Ericsson’s papers on deliberate practice and expertise, he gives a graph of expertise which looks logarithmic to me.  That is, the levels of skill that result from the accumulated hours of deliberate practice appear to me to be logarithmically related to the number of those accumulated hours of practice.  Intuitively this makes sense to me; our skill level tends to increase rapidly in the beginning, but then begins to plateau.  Even so, the logarithm is a monotonically increasing function.

Suppose that this function is in fact logarithmic.  The base ten log of 10,000 hours is four, while that of 1,000 hours is 3.  In other words, if this relation were in fact to hold true, then 1,000 hours of deliberate practice would yield 75% of the skill level of an expert.  Might this level of skill be competent?

Now consider our dilettante again.  Suppose she were to practice in domains orthogonal to one another; that is, one domain casts little, if any shadow upon another.  For example, math casts quite a long shadow on physics, but hardly any at all on dance apart from rhythm.

Suppose she had four or five of these orthogonal domains of practice in which she was “competent,” that is, had one thousand hours of deliberate practice in each; say math, dance, drawing, history, and meditation.  Consider what her experience of the world is like, its richness and depth, versus the world of the violinist who has given 10,000 hours of her life to the deliberate practice of the violin.  Could it be like the difference between Flatland and the world we live in?

What price glory?

How Many Hours of Practice?

How many hours of practice does it take to be good at something?

Since Malcolm Gladwell’s book, Outliers, came out in 2008 the 10,000 hour rule has been quoted more times than Poor Richard’s Almanac.  But that rule refers to the six sigma experts that are on the right hand edge of the bell curve.

What if you don’t want to be an expert at something, just competent?

For example, say you want to learn a new language well enough to be conversant, but not so well that you could write a piece of literature in that language.  How long would that take?

I’ve seen rules of thumb kicked around on blogs and in books, but never any research to back them up, apart from Ericsson’s 10,000 hour rule.

For instance one fairly common rule of thumb seems to be the thousand hour rule: that is, it takes one thousand hours of “practice” to achieve competence at some skill.  I found an article online that seems to confirm this for learning a second language.  The article refers to research done by the Foreign Service Institute, but without an actual citation.

Family Fortunes by the Bonner brothers asserts that it takes 1,000 hours to become competent, and 5,000 to become really good at some skill or other.  But again, they don’t cite any research to support these claims.

There is a graph in KA Ericsson’s article, The Influence of Experience and Deliberate Practice on the Development of Superior Expert Performance, that plots expert performance as a function of experience (see figure 38.1).  It looks logarithmic to me, but the article doesn’t make that assertion.

But suppose expert performance is in fact a logarithmic function of experience.  The log base 10 of 10,000 is 4, while the log base 10 of 1000 is 3.  In other words if this relationship is logarithmic, a person with 1000 hours of deliberate practice would have 75% of the skill of a person with 10,000 hours of practice.

If “really good” is halfway between the competent level at 1,000 hours, and the expert level at 10,000 hours, How many hours of practice does that translate to?  Well 103.5 = 3162, so about 3200 hours to achieve 87.5% of the skill of an expert.