This is the letter that causes the most offence, because the popular version implies that one group is rational and the other is emotional. That framing is wrong, it is insulting to both ends, and it survives because it flatters the people repeating it.
Everybody uses logic and everybody has feelings. The dimension is not about which faculty you possess. It is about which one your decisions are audited against when the two disagree.
What the dimension actually measures
The criterion you check a decision against before committing to it.
A Thinking decision is audited for consistency: is this coherent, does it follow, would the same rule produce the same answer next time, is it defensible independent of who is involved.
A Feeling decision is audited for effect on people: what does this do to the people it lands on, is it consistent with what I actually value, can I stand behind it as a person rather than as a process.
Both are evaluations. Both can be done well or badly. What differs is what gets treated as the tie-breaker.
The accusation each end makes
Worth stating both, because they are each half right.
Thinking says Feeling is soft — that decisions made on how people feel are inconsistent, unfair to whoever is not in the room, and vulnerable to whoever argues most emotionally.
Feeling says Thinking is cold — that decisions made purely on coherence ignore the fact that they land on people, and that a consistent answer which produces a terrible outcome is not actually a good answer.
Both of those are accurate descriptions of the other end’s failure mode, and neither is an accurate description of it working properly.
The bias nobody at the Thinking end notices
Here is the part usually left out, and it is the most useful thing on this page.
Treating emotion as noise is itself a bias. From inside, it is indistinguishable from objectivity — which is exactly what makes it dangerous.
How somebody feels about a decision is a fact about that decision. It predicts whether they will implement it, argue it, quietly sandbag it, or leave. A plan that ignores it is not more rational than one that accounts for it. It is less well specified: it has dropped a variable that is going to affect the outcome whether or not anybody modelled it.
Once stated that way, most Thinking types accept it immediately — which is the point. The correction is not “care more”. It is that the input was always data and was being discarded on a category error.
The mirror, for the Feeling end
The equivalent, and it deserves the same directness.
Empathy is biased. It is stronger for people who resemble you, stronger for one identifiable person than for a larger unseen number, and stronger for whoever is in front of you. A decision made on empathic response will reliably favour the vivid case over the greater harm, and the person making it will experience themselves as compassionate rather than as selective.
That is not an argument against caring. It is an argument for not using empathy alone as a compass, and for occasionally asking who your concern is currently pointed at.
The one real asymmetry
The dimension is close to evenly split overall — and it is the only one of the four with a substantial sex difference. Roughly two-thirds of women score toward Feeling and roughly two-thirds of men toward Thinking, across samples and cultures.
Two things follow. It is a difference of averages with enormous overlap, so it tells you nothing about any individual — there are a great many Thinking women and Feeling men, and both frequently report having been read badly for it their whole lives. And it is the reason this letter picks up so much cultural baggage that has nothing to do with what it measures.
The practical part
If you are toward the Thinking end: ask what each person will feel about a decision before finalising it — not for their sake, for the accuracy of your estimate. And spend the two seconds before a difficult message on delivery rather than on content. The content is already right; the tone is what will be remembered.
If you are toward the Feeling end: ask whether you can state the impersonal case for the decision you have reached. If not, you may be right and you cannot yet defend it, and in rooms that run on argument that is the same as being wrong.
And in either direction, the most useful move is the same: state which criterion you are using. Most arguments on this dimension are two people using different tie-breakers and each assuming the other is using theirs badly.
