Self-Esteem: Why It Doesn't Change Through Argument

Someone tells him, "You did brilliantly." He smiles, says thank you, and silently adds a footnote: they don't know how hard I worked to make it look easy. Or: they're just being kind. Or, most often: if they knew me better, they'd think differently. The praise runs off him like rain off a waxed jacket. It rolls away, and underneath he stays dry.

The people around him have tried everything. They remind him of his achievements, list his qualities, tell him sincerely how much they care about him. He listens, often agrees with the logic, and feels exactly the same afterwards. This is the most puzzling part of low self-esteem: it isn't short of arguments. It simply speaks a language that arguments cannot reach.

Self-esteem is the steady, often unconscious sense that you are worth something, far more than an opinion you hold about yourself. It is built within relationships, from the way important people saw you, and it works like a relational memory. Memories of this kind change through new experiences of relationship, not through argument.

Self-esteem and self-confidence: what's the difference?

A surgeon walks into theatre and knows exactly what she is doing. Her hands are steady, her decisions quick, her colleagues trust her. That same evening, at a dinner with friends, she feels that everyone else is more interesting, that her presence is tolerated rather than wanted. The same woman, a few hours apart.

The two words are often used interchangeably, and the confusion is costly. Self-confidence is about "I can": trust in a particular ability, in a particular area. It grows with practice and is confirmed by results. Self-esteem is about "I am worth something": a broader sense of having a place in the world and in your relationships, even on the days when you fall short. No result can prove it, because it was never built on results.

That is why the two can sit at opposite levels in the same person. People with considerable professional competence often carry a quiet certainty that they don't deserve what they have, and live in fear of one day being "found out". This is the experience described as impostor syndrome: success keeps accumulating while the sense of worth refuses to follow.

There is a subtler version too. When a person's value is tied entirely to performance, confidence ends up doing the job that self-esteem should be doing. While things go well, nobody notices the difference. It shows the first time something fails, and the whole structure shakes with it.

A gauge that watches other people

In the mid-1990s the social psychologist Mark Leary put forward an idea that changed how research thinks about self-esteem. He called it sociometer theory. In Leary's account, self-esteem acts as an internal gauge that monitors how accepted and valued we are by others. What it measures is relational value: how much we seem to matter to the people who matter to us.

The idea has solid empirical support. In experiments, self-esteem drops with real or imagined rejection and rises with acceptance. One finding from the same line of research is especially revealing: rejection lowers self-esteem more than acceptance raises it. The gauge is lopsided. It records losses in darker ink than gains.

The research also contains a detail that matters a great deal to psychodynamic thinking. The gauge doesn't only read what is happening now. It also draws on memories of earlier relationships and on what we expect to happen next. This is where the language of social psychology meets the language of the consulting room: the needle was set long before we walked into today's room. A person can be loved by everyone around them and still have a gauge reading from a relationship that ended thirty years ago.

Neuroscience has offered some interesting but limited evidence. In one brain-imaging study (Onoda and colleagues, 2010), 26 participants played a computer game of catch in which, at some point, the other "players" stopped throwing them the ball. Those with lower self-esteem reported more social pain and showed stronger activity in the dorsal anterior cingulate cortex, a region also involved in the distress of physical pain. The finding is striking, but it needs careful handling: the sample was small, the rejection was staged in a laboratory, and a correlation doesn't tell us what comes first. Mostly it tells us that, for some people, the same scene hurts more.

The self finds itself in another person's gaze

A baby cries. Her mother leans in, and a version of the baby's distress appears on her face: the brows draw together, the voice softens. She doesn't panic along with the baby, nor does she stay blank. She shows something recognisable yet slightly different, a "marked" version of the feeling, which says at once "I see what you feel" and "this can be survived".

Peter Fonagy, György Gergely and their colleagues described this process as marked mirroring. Parents who can think about their child's inner world respond to the child's feelings in a way that gradually lets the child form a representation of its own experience. In a sense, the child finds itself in the mind of another person. From this root grows mentalisation: the capacity to understand behaviour, our own and other people's, in terms of thoughts, wishes and feelings.

A few decades earlier, John Bowlby had described something related from a different angle. Through thousands of repeated moments of care, the child forms "internal working models": quiet expectations about how others will respond and, alongside them, about whether he or she is someone worth caring for. These models aren't stored as sentences. They are stored as ways of waiting, approaching and withdrawing. Heinz Kohut would speak of the gleam in the mother's eye, which confirms the child's existence before the child has done anything to earn it.

Here lies the answer to the puzzle at the start. If self-esteem were an opinion, better evidence would change it. It is something older and deeper: a relational memory, written at a time before words, and stored at a level that logic doesn't visit. Nobody can persuade a memory that it got things wrong. What we can do is add new experiences alongside it, enough of them and steady enough that eventually they carry more weight.

As a psychotherapist, I often meet people who remember very few specific events from childhood and describe their early years as "normal". And yet their relationship with themselves carries, with real precision, the quality of that early gaze: hurried, tired, conditional. The body and our relationships remember what the memory of events did not keep. What stayed silent in those years often speaks again in adult life, as explored in the article on childhood trauma and its silent expressions in adulthood.

Why don't kind words convince you?

In 2009 Joanne Wood and her colleagues at the University of Waterloo tested one of the most popular pieces of self-help advice: repeating positive statements about yourself. They asked participants to repeat "I am a lovable person". Those with high self-esteem felt slightly better. Those with low self-esteem felt worse than people who repeated nothing at all.

The finding is only surprising if we think of self-esteem as an opinion. If we think of it as a memory, it is almost predictable. A sentence far removed from what someone feels provokes the opposite voice: "that's not true", "if only they knew". Each repetition becomes a small proof of failure, since even this doesn't work.

A second finding is less comfortable still. William Swann and his colleagues showed, across dozens of replications in different laboratories, that people with a negative view of themselves prefer the company of those who also see them negatively. They called this self-verification. It sounds absurd until you ask what it offers: a world that makes sense. When others see you as you see yourself, you know what to expect. When they overrate you, you live with the feeling of a mistake waiting to be corrected.

In clinical practice this turns into repetitions that look like bad luck. The partner who belittles, the manager who never acknowledges, the friendship where only one person gives. A psychodynamic view sees neither masochism nor poor judgement here. It sees the pull of the familiar: relationships that confirm the gauge's old reading feel, paradoxically, safer than relationships that question it. A related dynamic is explored in the article on dependency in relationships, where the need not to be alone leads a person to accept less than they need.

None of this means kind words are useless. It means that, on their own, they don't reach where they need to. Wood's research found something hopeful as well: when participants with low self-esteem were allowed to consider the ways in which the positive statement wasn't true for them, they felt better. Change seems to pass through honesty rather than insistence. Later research on values affirmation points in a similar direction: when people write about what genuinely matters to them, instead of repeating flattering phrases, they often perform better under pressure. The difference is subtle and important. One exercise asks a person to believe a verdict about themselves. The other invites them to meet something they already love.

High but fragile: when self-esteem becomes armour

Public conversation treats self-esteem as a quantity: too little is a problem, plenty is health. Clinical experience doesn't bear this out. We have all met the person who speaks about themselves with total certainty and collapses, or turns aggressive, at the first criticism.

In 2003 Michael Kernis proposed a distinction that helps make sense of this. High self-esteem, he argued, can be secure or fragile. Fragile self-esteem is defensive, conditional, fluctuates sharply from day to day, and often doesn't match what the person feels about themselves at a less conscious level. Secure self-esteem is stable and doesn't need constant confirmation. Kernis saw authenticity as its core feature: being able to be who you are without having to prove it.

From a psychodynamic perspective, fragile high self-esteem often works as armour around exactly the same wound we see in low self-esteem. The strategy changes; the root stays the same. One person withdraws to avoid rejection. Another keeps achieving so as never to have to ask whether they would be accepted without their achievements.

The line between healthy pride and defence isn't always clear, though, and here I must be honest. Some people with a real sense of worth react strongly to an unfair criticism, and some with fragile self-esteem have learnt to hide their vulnerability flawlessly. Research measures trends in groups, not the story of a particular person. The question "how much of my confidence is real and how much is protective?" can't be answered by a questionnaire. It takes time, and often another person willing to explore it with you.

What changes in psychotherapy

If self-esteem is a memory of relationship, changing it has to pass through a relationship. That is why psychodynamic work rarely sets out to convince someone that they are worthwhile. A therapist who praises persistently soon ends up in the position of the friends at the start of this article: kind, and ineffective.

What therapy offers is a different kind of experience. One person pays steady attention, week after week, without needing the other to perform, to be pleasant or to get better quickly. They tolerate disappointment, anger and silence, and they stay. At some point the person in therapy begins to notice something strange: they haven't been rejected, even though their gauge predicted it with certainty.

In more recent work, Fonagy and Elizabeth Allison speak of epistemic trust: the capacity to trust that what another person conveys is true and relevant to you. People with deeply low self-esteem have often lost exactly this kind of trust towards any kind word. They don't regain it through explanations. They regain it when they feel that someone has genuinely understood them, including their difficult sides.

The path is slow and not linear. There are periods when everything seems to slide back to the beginning, especially after a rejection or a failure. Nobody can promise a timetable. What usually changes is not that a person suddenly starts to admire themselves. Something quieter shifts: the voice that accuses them loses its absolute certainty.

In therapy that loss of certainty often takes a specific form. The person stops treating their inner critic as an objective description and starts hearing it as a voice with a history. They begin to wonder who used to speak like this, and at what age that tone was first heard. The question isn't asked to find someone to blame. It is asked because a voice with a history is no longer the law of nature it seemed to be. This is, in essence, mentalisation turned towards oneself: the capacity to think about what you feel rather than simply being subjected to it. More on how this approach works can be found in what psychodynamic psychotherapy is.

One note of care. When low self-esteem comes with persistent sadness, loss of interest in things that used to bring pleasure, disturbed sleep or thoughts that life has no value, it may be depression. In that case, talking to a professional is a necessity rather than a luxury.

In simpler terms

Self-esteem is the sense that you are worth something even when you fall short, and it is different from trusting your abilities. It forms very early, from the way the people who raised you saw and felt you, and it stays stored like a memory rather than a thought. That is why positive affirmations and logical arguments often fail to reach it, and can sometimes hurt it further. Impressive confidence doesn't always mean secure self-esteem either. New, steady experiences of relationship are what shift it, and psychotherapy is one place where such an experience can happen safely.

To be seen, again and again

Low self-esteem is neither a character flaw nor a lack of willpower. Very often it is a faithful record of an old relationship, still being read as though it were present. That may sound heavy, but it carries its own hope: what was written within relationships can, with time, be rewritten within relationships.

You don't need to be convinced of your worth for this change to begin. What it usually needs is that you let someone see you as you are, often enough that the old needle slowly begins to move.

Theoretical sources

For anyone who would like to read further, the main sources behind this article, in alphabetical order. Where legitimate free access exists, the title links to the text:

Inge Bretherton, The Origins of Attachment Theory: John Bowlby and Mary Ainsworth (1992), free PDF: the clearest introduction to Bowlby's thinking and to internal working models.

Peter Fonagy, Elizabeth Allison, The Role of Mentalizing and Epistemic Trust in the Therapeutic Relationship (2014), free PDF from University College London: mentalisation and epistemic trust within the therapeutic relationship.

Peter Fonagy, György Gergely, Elliot Jurist, Mary Target, Affect Regulation, Mentalization, and the Development of the Self (2002), book with no free edition: marked mirroring and the birth of the self in the mind of another.

Michael Kernis, Optimal Self-Esteem and Authenticity: Separating Fantasy from Reality (2003), free PDF: secure and fragile high self-esteem, and the role of authenticity.

Mark Leary, Making Sense of Self-Esteem (1999), abstract on the publisher's page: sociometer theory from its originator.

William Swann, The Trouble with Change: Self-Verification and Allegiance to the Self (1997), free PDF from the author's lab: why we seek confirmation even of a negative view of ourselves.

Joanne Wood, Elaine Perunovic, John Lee, Positive Self-Statements: Power for Some, Peril for Others (2009), abstract on PubMed: the study showing that positive statements can harm the very people who need them most.