What Agency and Communion Mean in Psychology

Two people describe the same year. Both ended a marriage, both moved cities, both started over in a rented room with someone else's furniture in it.

The first one says: I decided I was done, I found the flat in three weeks, and I have hit every target I set myself since. The second one says: it was a bad year, but my sister flew out and stayed a month, and the people at the new job were kinder than they had to be.

Neither account contradicts the other. They are not opposite personalities, and they are not the good outcome and the bad one. They are simply high on two different thematic dimensions, and narrative psychologists code those two separately — in the same story, at the same time. Agency is the degree to which the teller affects change in their own life or influences the people around them, through mastery, achievement, or the sense of controlling their own fate. Communion is the degree to which that same teller experiences connection with others — love, friendship, dialogue, belonging (McAdams & McLean, 2013).

The mistake almost everyone makes

The agency theme has a long lineage. Adler lists Bakan's 1966 book on the duality of human existence among the sources standing behind it (Adler, 2012).

It is easy to read the pair as a single spectrum, with the driven achiever at one end and the warm connector at the other — and that reading distorts everything downstream.

Pick your side.

That is not how they are scored. They are two separate constructs, each rated as a degree in its own right (McAdams & McLean, 2013). Nothing in the definition of one makes the other smaller — the two do not trade against each other.

So there are four possibilities, not two. Your story can run high on both. It can run low on both. It can run high on one and quiet on the other. Someone describing a company they built with people they love is not splitting a fixed budget between two accounts — that story simply scores high twice.

Even where the two collide, they collide as separate things. McAdams describes a life-story prototype, found among highly generative American adults, in which the protagonist experiences real conflict between the desire for agency and the desire for communion (McAdams, 2011). The conflict is possible because both are present and both are strong. That prototype also belongs to a specific culture — it is a model of the good life circulating in American midlife, not a universal human shape.

Agency is a fact about mastery, not mood

Here is a rough way to hear it in what you say. Notice who is doing the verbs.

"I decided to leave" and "things fell apart" can describe the identical event — the same month, the same ending, the same flat. In the first, you act. In the second, a situation happens and you are where it lands. That difference is usually where you hear it first, but what gets scored is the mastery and control the story claims, not the grammar (McAdams & McLean, 2013).

It is not optimism.

Those two get confused constantly. The high-agency example McAdams and McLean give is a story about a divorce — a narrator who reports that since the marriage ended, they have challenged themselves to the limit and accomplished virtually any goal they set (McAdams & McLean, 2013).

The event is a loss. The telling is agentic anyway. Agency does not ask whether your year went well. It asks whether you were affecting things or being carried by them.

Communion is a fact about whether other people are real

Communion has its own tell, and it is not a headcount.

Other people can appear in your story as people, or they can appear as scenery. A narrative can be crowded with colleagues, exes, siblings and landlords and still run low in communion, if every one of them is furniture — named, present, and doing nothing to you. What gets coded is the connection itself — warmth, dialogue, care, belonging.

The matching high-communion example achieves nothing at all. A narrator describes a night of being warm, surrounded by friends and positive regard, and feeling unconditionally loved (McAdams & McLean, 2013). No goal is met in that sentence. None needs to be.

The finding that runs forward in time

Most of what is known here is correlational — which means it cannot tell you which way the arrow points. One study gets closer than the rest, because it followed the same people forward in time instead of measuring them once.

Adler followed forty-seven clients at one outpatient clinic in Chicago through psychotherapy, collecting a written account of how treatment was affecting them before they began and after every session, across twelve assessment points (Adler, 2012). Each account was coded for agency and for coherence. Agency rose over the course of treatment; coherence did not. Increases in agency tracked improvements in mental health. And in lagged models, the change in agency came first — the shift in how people told their story preceded the improvement in their symptoms rather than following it.

The ordering is what makes it interesting. The simplest deflating reading is that people who already feel better narrate themselves as more powerful afterwards, and in that sample the reverse pattern did not appear. That weakens the deflating reading rather than settling it: lagged models show the direction of an association, not a cause. Adler is also careful that the weight his study puts on agency may be an American expectation rather than a universal one.

Writing about this literature, McAdams and McLean note that results like Adler's are consistent with narrative change contributing to wellbeing, and that considerably more research is needed to disentangle the causal relations (McAdams & McLean, 2013).

Neither one is a virtue

This is where the pair gets misused, so it is worth being blunt. High agency is not better than high communion. Neither is a number to raise.

A life story dense with agency and empty of other people is not a success. Read the two openings again. The first teller acts on the world; the second is held by people who showed up. Nothing here makes one of those the better person. Agency is the theme with mental-health findings attached to it, and communion has not been studied the same way (McAdams & McLean, 2013).

The distribution is not a merit ranking either. McLean, Pasupathi and Pals note that in fully formed life stories, women are more likely than men to voice themes of communion (McLean, Pasupathi, & Pals, 2007). A theme tracks what a person has been taught to say out loud — not only what they have lived through.

And a coded theme is a feature of a telling, not a measurement of a life. Those same authors define what they study as situated stories — accounts produced in a particular situation, by particular people, for particular audiences, to serve particular goals (McLean, Pasupathi, & Pals, 2007). The room moves the result. In the studies they review, attentive listeners draw out more elaborated tellings than distracted ones. Change who is listening and the story changes with it.

So none of this describes who you are. It describes how one account, told once, was built.

What to listen for the next time you tell it

You do not need an instrument to hear either dimension. You need to notice two things in your own sentences.

The verbs. When you describe the hardest year you have had, are you the one acting, or the place where events landed? Both can be honest accounts of the same twelve months. They are not the same story.

The people. Notice who appears, then ask whether any of them are doing anything to you beyond standing in the background. If your narrative moves people around like furniture, that is information. If you can barely describe a decision without naming who was there for it, that is information too.

Neither answer needs fixing. That is the whole point of noticing — a description is not a prescription. The assessment on this site reads these two dimensions back to you and stops there. That makes it a mirror. It has nothing to say about which of the two you ought to want more of.

Tell the story once, on purpose, and listen to yourself do it. Most people never have.