Twelve defence mechanisms, in plain English

A defence mechanism is an unconscious mental operation that keeps an unwanted thought, feeling or impulse out of awareness, at the cost of distorting something. Sigmund Freud named the idea in 1894; his daughter Anna Freud produced the first systematic list in 1936. Twelve of them are defined below, followed by the four pairs that are routinely confused with each other.

• A defence mechanism is unconscious by definition. If you decide to do it, it is coping, not defence — with one exception, suppression, which is deliberate on purpose.
• Anna Freud’s 1936 book names nine mechanisms and adds sublimation as a tenth, describing it as belonging to the study of the normal rather than of neurosis.
• Rationalisation is not on her list. Ernest Jones named it in 1908, in a paper called “Rationalization in every-day life”.
• Defences are not pathological in themselves. George Vaillant’s hierarchy ranks them from psychotic to mature, and the mature ones predicted better long-term outcomes.
• The DSM-IV printed a Defensive Functioning Scale in its appendix of measures needing further study. It never left that appendix, and DSM-5 does not carry it.

What a defence mechanism is, and who first listed them

A defence is what the mind does when something is unbearable to know. It is not a decision. That is the whole point of the concept: if you can watch yourself doing it, the theory says it is not working.

Freud used the word Abwehr — defence — as early as 1894, in a paper on what he then called the defence neuro-psychoses, and returned to it throughout his career. What he never produced was a catalogue.

Anna Freud’s “The Ego and the Mechanisms of Defence” supplied one in 1936. She names nine — regression, repression, reaction formation, isolation, undoing, projection, introjection, turning against the self, and reversal — then adds a tenth, sublimation, noting that it belongs to the study of normal development rather than of neurosis. Her own contribution to the list, identification with the aggressor, is described in the same book.

The list has grown since, in ways that are not always tidy. Some later additions are clinically indispensable and none of them are Anna Freud’s.

The twelve, one sentence each

Each of these is one operation. The German names are given because several of them say plainly what the mechanism does, where the English name says nothing at all — an example of what happens when psychology crosses a language border.

Two words about that list. It is not exhaustive, and no version of it is: splitting, acting out, idealisation, devaluation, altruism and humour all appear in serious clinical use. And nothing on it is a diagnosis. Everybody uses most of these, most weeks.

The four pairs people mix up

Half of what makes these terms hard is that they arrive in near-identical pairs. Learning them one at a time is how you end up able to define each and unable to tell them apart.

Repression against suppression

Repression is not a decision; suppression is. If you deliberately put a worry aside until after the exam, that is suppression, and Vaillant classes it among the mature defences. If you cannot recall something and are not aware of avoiding it, that is repression. German keeps the same pair — Verdrängung and Unterdrückung — and confuses it just as often.

Projection against displacement

Both move something outward, and they move different things. Projection moves the attribution: the feeling is yours, and you locate it in somebody else — “he is furious with me”, when you are furious with him. Displacement moves the target: the feeling stays yours and lands on somebody safer than the person it belongs to. Shouting at your flatmate after a bad meeting is displacement, not projection.

Denial against rationalisation

Denial rejects the fact. Rationalisation accepts the fact and rewrites the reason. Someone who insists the diagnosis is a laboratory error is in denial; someone who agrees with the diagnosis and explains that the appointment can wait because work is busy is rationalising. Rationalisation is by far the more common of the two, and much harder to spot, because everything it produces sounds sensible.

Sublimation against reaction formation

Both convert an impulse, and only one of them keeps its direction. Sublimation redirects the energy into something valued — the classic example is aggression channelled into surgery or sport. Reaction formation replaces the impulse with its opposite, usually louder than necessary: the person who cannot tolerate their own hostility toward a colleague and becomes conspicuously, exhaustingly nice to them. The exaggeration is the tell.

Are defence mechanisms still a serious idea?

Yes, though not in the form most first-year textbooks present them. Three things are worth knowing.

Defences form a hierarchy, and the level matters more than the label

George Vaillant proposed one in 1971, ordering mechanisms from psychotic through immature and neurotic to mature. Working with decades of longitudinal data, he reported in 1976 that men whose characteristic defences were the mature ones — sublimation, altruism, suppression, anticipation, humour — were doing better across work, relationships and health than men relying on immature ones. He restated the case in 2000, arguing these mechanisms belong in positive psychology rather than in a list of symptoms.

They can be measured, which was the hard part

Rating scales applied to interview transcripts, and self-report instruments such as the Defense Style Questionnaire, produce usable reliability. Phebe Cramer’s 2015 review sets out the empirical case, and is direct about the conceptual muddle that comes from every author maintaining a slightly different list.

The diagnostic manuals let them go

DSM-IV printed a Defensive Functioning Scale in Appendix B, among the criteria sets and axes provided for further study, with defences sorted into seven levels. It stayed a proposal, and DSM-5 does not carry it. Defences remain central to psychodynamic training and to psychotherapy research, and they are absent from the diagnostic criteria — both of those are true at once.

There is one more reason the vocabulary is worth having straight. Rationalisation is exactly what cognitive dissonance predicts a person will do when a behaviour and a belief collide, and the two literatures describe the same everyday moment with different words. Projection has a cousin one shelf along as well: several of the cognitive biases describe the same misattribution with no unconscious machinery behind it at all. Holding twelve near-neighbours apart is not a memory problem so much as a contrast problem — which is a distinction that changes how you should study them.