Work, love and play

By Martin E.P. Seligman, PhD
APA President

As my department was writing the ad for a social psychologist last fall, I got an e-mail note that gave me a sleepless night.

'Marty,' my colleague Jon Baron wrote, 'do you think it would be all right if one of the interests listed was industrial psychology?'

I wrote back, 'Of course?the three great realms of life are work, love and play, and industrial psychology is the psychology of work. Remember that Morris Viteles [who had just passed away at age 98] was a distinguished member of our department for 75 years and a founder of industrial psychology.'

Then I couldn?t get to sleep. It bothered me that my department did not have anyone whose main interest was work, love or play. As I mentally ran through the rosters of several other fine academic departments, I couldn?t think of anyone whose primary research was work or love or play. I happened to see Jerome Bruner the next day. Jerry, at 82, is still yeasty and going strong and is a walking history of modern psychology.

'How did this come to be?' I asked him.

'It actually happened at a moment in time,' he replied. 'About 60 years ago the chairmen [the gender is intentional] of Harvard, Princeton and Penn got together at a meeting of the Society of Experimental Psychologists and agreed that they would hire no applied psychologists! This set the hiring pattern of many of the great departments to this very day.'

A need for better synthetic thinking

This is a symptom of a larger problem. Complete scientific activity requires both analysis and synthesis. Synthesis is needed when you are not sure that the elements of your analysis 'carve nature at the joints.' The only way to test the validity of your elements is by reconstructing natural phenomena with them. What entitles academic physics and chemistry to be almost solely analytic is their long, accompanying history of synthetic activity, which demonstrated the validity of their elements: predicting eclipses, tides, synthetic fibers, rockets, the green revolution and computers. What those august chairmen forgot was that psychology did not have a history of engineering, and so there was real danger that its laboratory elements might not be the elements out of which mind, emotion, behavior and psychopathology are made.

Around 1970, the National Institutes of Health began to require that grant applicants write a 'significance' section. We wrote perfunctory ones and then were vexed when we found they were actually being read and reviewed and grants were being turned down because of poor synthetic thinking. Psychology needs better synthetic thinking. But not only to get grants. We need better synthetic thinking to do better science. Our journal and academic structures now strongly reward only analytic thinking. We select students for analytic ability; then we hire assistant professors and promote them for high-quality analytic work.

A suggestion

Perhaps some structural changes will promote higher quality synthetic thinking. Here?s a suggestion for all the APA journals: that the editors in chief encourage synthetic thinking strongly in all empirical journal articles. One way to do this would be a concluding section called 'Synthesis' or 'Significance.' This would allow the authors to address the following three issues:

? Problem choice. Why did the authors choose the problem they did? What larger issues, issues that transcend the particulars of the subfield, hinge on the findings? What larger propositions are confirmed or disconfirmed by the extrapolation of these findings to such molar issues.

? Levels of analysis. How can the findings be linked to phenomena at more complex and less complex levels of analysis?

? Application. If the findings are valid and replicable, what 'real-life' psychological phenomena might be explained, modeled or reconstructed? What else would have to be known for such a reconstruction to be testable? To be true? What applications seem feasible?

The benefits for psychology of rewarding synthetic thinking to complement analytic thinking are several: It will move significance claims from hand-waving toward rule-governed, rigorous discourse. It will test whether our analyses are valid. It will help link our findings to more complex phenomena as well as to simpler systems. It will make our journal articles more readable. It will help our scientists communicate effectively with the press, the public and Congress. And not least, it will bring science and practice closer together.