Sunday, August 30, 2009

Book Review: "Successful Intelligence" by Robert Sternberg. Very important ideas, but weak presentation

Link to posting on Amazon.

Leading intelligence researcher Robert Sternberg dedicates this book to the woman who turned his life around in the 4th grade, his teacher Mrs. Alexis. Bless her and teachers like her. She was the first one to have the sense to ignore his poor standardized test results and expect and encourage him to do better than the tests predicted. Sternberg's own career is a poignant and dramatic example of his own amazing discovery, that human capability is not a matter of fixed traits at all, contrary to what many have claimed based loosely on the programs of standardized testing and heritability research.

I picked up this little book in a bargain bin, and a bargain it truly was. Robert Sternberg is one of the great scholars of psychology research with a number of superb books on topics of fundamental importance like love, intelligence, success, and wisdom. These are not cute self-help books, they are mostly very readable scientific treatises that offer original technical models and expert analysis, and sometimes also include practical advice.

This book, Successful Intelligence, is not one of Sternberg's more technical books and definitely not among his best work from a technical perspective but it is one of his most relevant and important as far as its argument.

The thing I most want you to know about this book that it is a compelling argument for the importance of the dynamic ability to build on our strengths and compensate and correct for weaknesses.

The things that are fixed in human development are far less interesting and important to us that the things we can learn to build on to continually improve ourselves. If you already get that point, you may want to skip this introduction and go straight to his more technical books and others on metacognition, emotional intelligence, and so on that are in the same spirit as this book.

Sternberg's overall idea is a BIG deal because it is not now how education generally works. For example it is very much contrary to the major traditions of education in the United States: (1) the tradition of using standardized testing to predict ability, and (2) the anti-elitist tradition of presumably democratic equality that throws all students together as if they all learn the same way. This third philosophy of education is not wishful thinking, it is the result of analysis of real data about how people learn.

The book builds on Sternberg's formal triarchic model of intelligence, which sees mental power in terms of three things: analytic, creative, and practical intelligence, and our ability to make use of all three. Sternberg proposes that "successful intelligence" is the master quality that results from our learning to make best use of these three different kinds of ability. Successful intelligence is what really counts, according to Sternberg. And I think he makes his case convincingly.

Analytic intelligence is what psychometricians consider the g factor, the common factor among various cognitive subtests that tends to go up and down together. Analytic intelligence is what the application of IQ testing and other standardized testing for the prediction of outcomes is based upon. This is real, but we don't actually know what it is measureing and we rely far too heavily on it. Analytic intelligence seems to let us solve problems and judge the quality of ideas. It appears to let us perceive patterns in complexity.

Most crucially, Analytic intelligence is often used to predict further test taking ability and school performance and varyingly predicts some kinds of job performance. But what does weak to moderate predictability in these areas really tell us?

Where Sternberg agrees with the late Stephen Jay Gould's infamous critique in "Mismeasure of Man" is that correlation does not imply causation in this particular case. One of the central points of this book is the enormous gap between the very low to moderate ability these tests predict life outcomes and what people are actually able to do under the best conditions. Sternberg's emphasis is on what it takes to create the best conditions. It isn't clear that there is any way yet to manufacture enormous changes in analytic intelligence, but it is increasingly clear that boosting IQ should not be our main concern regarding human intelligence, beyond simply ensuring that we have the best developmental environment.

Then there is creative intelligence, which we need in order to formulate good problems and ideas in the first place, to ask the right questions. Standardized testing does not measure this at all.

Then there is practical intelligence, which is the rich background each of us needs in order to apply ideas and analysis effectively in everyday life. This is not measured by standardized testing either.

The last chapter of the book consists of the high level skills we need in order to have successful intelligence and make best use of analytic, creative, and practical intelligence. There are 20 of these, which are frustratingly generic and mostly should be commonsense I think, but they are not emphasized in education very much and they give a good flavor of Sternberg's philosophy. These include:

Self-motivation skills,

Impulse control skills,

Knowing when and how to persevere,

Knowing how to make the most of your own abilities,

Knowing how to translate thought into action,

Learning to focus on the products of your efforts, not just the process,

Learning to both initiate and complete tasks and to follow-through,

Learning to get past fear of failure,

Learning to conquer procrastination

Willingness to accept fair blame for mistakes


Finding ways to surmount personal challenges

Learning to focus and concentrate to achieve goals

Learning to delay gratification

Learning to see both the forest and the trees

Reasonable self-confidence but not excessive (in contrast to the dismal failure of extreme versions of"self-esteem"psychology)

Learning to balance analytic, creative, and practical intelligence in thinking.

This is obviously not a curriculum, it is a direction. This is not a great book in terms of scholarship or details but it is a compelling introduction to the ideas. A wonderful, research-based philosophy based on human success and satisfaction. And a novel direction for education.


  1. Hi Todd,

    I think Sternberg is a very interesting thinker. He has been one of the first to create a counterweight to the dominant thinkers on intelligence among psychologists.

    Not only his ideas have been very refreshing but also his approaches to research (for instance his taking seriously laypeople's views on intelligence.

    In this book I found the first chapters very interesting. His personal revelations were inspiring, I thought. The rest of the book I found a bit confusing here and there. The basic idea I found useful and powerful but in the way he operationalized it I thought he was mixing personality and intelligence. Maybe I should reread the book.

    Another book by Sternberg which I also found interesting was Defying the crowd. Have you read that too? In this book too I found the basic idea very appealing but the operationalization a bit confusing. This could, of course, imply that I just failed to graps it well.

    All in all, a very interesting thinker who has been very influential.

    Have you also read Outsmarting IQ by David Perkins? This is a book in the tradition of Sternberg and very interesting

  2. I think I know what you mean. Sternberg often starts out very strong with a clear model of abilities and a direction for what he wants to measure to validate it, and then it's almost as if he gets tired of it by the time he goes to establish the measurements to operationalize his concepts.

    I suspect that's part of his way of thinking, he is afterall very much against using measurements to pigeonhole people by predicting their outcomes through fixed traits. So I think he is a little conflicted when it comes to figuring out ways to do research on his models without assuming fixed traits. It's hard to find ways to measure something that is supposed to by dynamic and skill-based. You can mostly only measure its outcomes or observe its process. That's why the g factor was so compelling in psychometry afterall, it gives you something to measure.

    When I've read his followup on how to teach for successful intelligence maybe I'll have a better since of where he was going with this.

    Haven't read Perkins, I'm adding him to my wish list. Thank you!!