Emotional Intelligence

Goleman is a science journalist synthesising the work of others, principally Peter Salovey and John Mayer, who had proposed emotional intelligence as a construct several years earlier.

His contribution was popularisation and the strong claim attached to it: that these abilities matter more than conventional intelligence for many outcomes, which is where most of the subsequent argument has concentrated.

The construct is Salovey and Mayer's. The claim that it outweighs IQ is Goleman's, and it is the contested part.

Devmark’s reading — not a quotation from the book

His framework: self-awareness, self-regulation, motivation, empathy and social skill.

The first two are the foundation in his account — the ability to notice an emotional state as it arises and to modulate the response rather than being carried by it. The remaining three are treated as building on that base.

His most durable contribution to popular vocabulary. Drawing on Joseph LeDoux's research, he describes a fast subcortical pathway that can trigger an emotional response before the cortex has fully processed the situation.

The underlying neuroscience is real, though Goleman's popularisation simplifies it considerably and the strict two-pathway picture has been refined substantially since.

The practical content follows: if the response arrives before the appraisal, regulation must operate on what happens next rather than on the initial reaction.

His recommendations — naming the emotion, creating a delay, reappraising the situation — anticipate techniques that later research has supported reasonably well, particularly affect labelling and cognitive reappraisal.

He gives prominent treatment to Walter Mischel's delay-of-gratification studies and their reported association with later outcomes.

This requires an update. Large replication work found the original association substantially weaker once family background and socioeconomic factors were controlled, and the popular version of the finding considerably overstates what the evidence supports.

The delay-of-gratification association largely attenuates once family background is controlled. The popular version overstates it.

Devmark’s reading — not a quotation from the book

The organisational chapters made the book's commercial reputation, arguing that these competencies distinguish outstanding performers particularly in senior roles, and that the higher one goes in an organisation the more they matter relative to technical ability.

Much of the supporting evidence for this comes from consulting research and internal corporate competency studies rather than from peer-reviewed work, and the reported effect sizes have not held up well under independent scrutiny. This is the section of the book a reader should treat most sceptically, and it is the section most often cited in workplaces.

His sections on social and emotional learning programmes in schools are the most constructive part of the book, and this is the area where the evidence has developed most positively.

Meta-analyses of well-implemented programmes have found modest but real improvements in behaviour and academic outcomes, which is a better record than most of the book's other claims.

The central criticism is definitional: emotional intelligence as Goleman uses it is a broad mixture of traits, skills and motivations rather than a coherent construct, and much of it overlaps with established personality dimensions.

The ability-based model that Salovey and Mayer developed is measurable and more modest in its claims; Goleman's mixed model predicts less than it appears to once conscientiousness and agreeableness are accounted for. The specific claim that it outweighs IQ is not supported.

Its cultural effect exceeded its scientific standing. It made emotional competence a legitimate subject in workplaces and schools that had treated it as outside their concern, and it created demand for the social-emotional learning programmes whose evidence base is now reasonably good.

That is a real contribution even where the underlying claims were overstated.

As an influential popularisation with a strong claim that did not survive scrutiny and a set of practices that mostly did.

Take the self-regulation techniques, the affect-labelling insight and the educational material. Discard the comparative claim against IQ, treat the workplace research sceptically, and update the marshmallow chapter entirely.