Avoidant attachment style
Avoidant attachment style: Why the phrase ‘He’s an avoider’ puts an end to any further reflection
Avoidant attachment style: Why the phrase ‘He’s an avoider’ puts an end to any further reflection
DESCRIPTION: Attachment styles have become a tool for relationship diagnosis on TikTok. Those labelled as ‘avoiders’ can hardly explain themselves: every answer merely confirms the label. What Bowlby and Ainsworth actually studied, and why the Strange Situation test does not measure relationships.
The Avoider
“He’s simply an ‘avoider’.” The phrase sounds like a realisation and feels like drawing a line under it. It transforms a person who withdraws in certain situations into a type about whom nothing more needs to be said.
What it’s all about:
· what attachment research has described,
· how this became a weapon, and
· why the same mechanism works in reverse for the woman who writes too much.
The statement against which there is no rebuttal
Anyone labelled as ‘avoidant’ has already lost.
If he says he needs some time to himself, this is seen as a retreat and confirms the label. If he says the relationship feels too close, it is seen as a rejection of intimacy and, in turn, confirms it. If he says he doesn’t love her and doesn’t want any commitment, this too is interpreted as a symptom: that’s how avoiders talk; they simply don’t realise it. Even the admission that he is probably an ‘avoider’ serves as evidence.
A statement that is confirmed by every conceivable response explains nothing. It labels. From that moment on, the other person is judged rather than listened to, and their view of their own situation is treated as material for a ‘diagnosis’.
That is precisely where the appeal lies. The sentence puts an end to the uncertainty about whether the other person is afraid, unable, unwilling, or asking too much. This uncertainty is hard to bear and involves hurt, shame and the possibility of having been wrong. And that is why the sentence strikes back, in three words.
What Bowlby and Ainsworth investigated
John Bowlby described a system that kicks in under stress: in the face of danger, separation, illness or being overwhelmed. Through repeated experiences, children form expectations about whether someone is available, whether help will come and whether they are allowed to express their own needs. Bowlby called these expectations ‘internal working models’. The term refers to models in the literal sense: assumptions that can be adjusted in the light of new experiences.
Mary Ainsworth investigated these expectations in the ‘Strange Situation’, a standardised procedure involving brief separations and reunions in an unfamiliar room. From the child’s behaviour upon the primary caregiver's return, she identified patterns, including an ‘avoidant’ one.
‘Avoidant’ described a strategy adopted under stress. A child who shows little distress upon reunion has learnt that openly expressing distress does not elicit a reliable response. They suppress their expression whilst retaining the need. Physiological measurements show increased arousal in these children, whilst they appear calm on the outside.
This is a statement about an adaptive response. It does not say anything about the character of an adult man.
The Strange Situation
This is where pop psychology falls short.
The same child may be classified differently in the Strange Situation when with their mother than when with their father. The meta-analysis by van IJzendoorn and De Wolff on father-child bonds shows that the classifications are only weakly correlated with either parent. What can be observed is a child’s behaviour towards a primary caregiver.
This means the category is already shaky at its very foundation. If a child reacts securely with one parent and avoidantly with the other, there is no such thing as an ‘avoidant child’. There is a relationship in which distress was responded to, and one in which it was not.
This pattern continues into adulthood. People experience varying degrees of security with parents, friends and partners. The same person may appear aloof in one relationship and affectionate in the next. In addition to general orientations, attachment research explicitly recognises relationship-specific models.
Urie Bronfenbrenner highlighted the methodological aspect as early as 1977, when he accused developmental psychology of having become the ‘science of the strange behaviour of children in strange situations with strange adults for the shortest possible time’. This statement highlights the issue of generalisation: a twenty-minute encounter in an unfamiliar room does not allow for any conclusions to be drawn about forty years of relationship life.
Social media
On TikTok, Instagram and in dating guides, these patterns have been distilled into four types of person: secure, anxious, avoidant, and anxious-avoidant. Anyone who messages a lot is anxiously attached. Anyone who doesn’t reply for three days after an intense weekend is avoidant. Anyone who does both is anxiously avoidant.
These categories promise something that research does not actually provide: insight into the future. Anyone who has identified the ‘type’ believes they know how the relationship will unfold and considers any further conversation a waste of time.
In doing so, they lose sight of the malleability inherent in the concept of the working model. Attachment styles shift with life circumstances, within stable relationships, following losses, and during psychotherapy. Review articles on the stability of attachment describe a relationship between continuity and change. A template that remains valid for forty years does not feature in them.
Withdrawal has many causes.
Withdrawal hurts, but it requires explanation. Someone withdraws because conflict immediately triggers shame and a loss of control. Because they experience a demand for commitment as a restriction. Because the other person is pressuring, controlling or belittling them. Because the relationship has failed to find a language for dealing with differences.
Someone may also withdraw because they do not want to invest. Because the matter is not important enough to them. Because they want to keep other options open. Because they want the benefits of a relationship without taking on its obligations.
For neurodivergent people, there is another possibility. Sensory overload, exhaustion following intense closeness, delayed processing of conflicts, a lack of time awareness or difficulties with self-organisation can manifest in relationships in the same way as indifference. An unanswered message is not in itself proof of attachment avoidance. This connection is complex enough to warrant a separate article. The only thing that matters here is that the same outward behaviour can have several causes.
The action itself does not answer the question of its function or the motives behind it. Whether a withdrawal serves to regulate an overload, to ward off dependency, to protect against overstimulation, or to exert control over the other person is determined by follow-up questions or, failing that, by subsequent behaviour.
The counterpart
The avoider rarely appears alone. A second figure is associated with them: the woman who writes too much, presses too soon, takes a lack of reply badly, and seeks contact all the more urgently after the other person has withdrawn.
She is initially labelled ‘anxiously attached’. As her protests grow louder, she is labelled ‘emotionally dependent’, then ‘controlling’, then ‘dramatic’. Ultimately, the most serious label looms large: ‘borderline’.
The division of roles is fixed. His aloofness is psychologised and thus explained; her protest is psychiatrised and thus dismissed. He cannot help himself; she cannot control herself. What used to be called hysteria is now referred to as a dysregulated attachment system.
Yet the underlying facts are the same. He gets in touch irregularly, alternates between closeness and distance in an unclear pattern, and avoids conversations about commitment. She asks for clarification. Who is considered disturbed in this scenario depends on what behaviour the culture deems acceptable in the first place.
From attachment style to personality disorder
There is a link between attachment insecurity and borderline traits, and it is unevenly distributed. A meta-analysis of romantic attachment styles and borderline pathology finds a moderate correlation of approximately r = 0.48 for anxious attachment and approximately r = 0.30 for avoidant attachment.
No diagnosis can be drawn from these figures. A correlation of this magnitude describes a clustering of cases across sample groups. It says nothing about any person, and certainly nothing about someone known only through chat histories and the account provided by the other party.
Borderline personality disorder requires a persistent, cross-situational pattern, significant distress or functional impairment, and the consideration of alternative explanations. Numerous messages, jealousy, the despair of a break-up and the desire for swift clarification fulfil none of these criteria. A conflict-ridden romantic relationship is not a clinical setting.
The harm caused by such ‘kitchen-table diagnoses’ is well documented. A borderline diagnosis is one of the most heavily stigmatised of all, and this stigmatisation is gendered: it reduces women to being dangerous, manipulative and unreasonable. Anyone who uses it in an argument is reaching for the harshest word available. The angry accusation of narcissism levelled at men is not much better.
Why the diagnosis provides relief
Diagnostic labels serve the same purpose for whoever applies them.
Anyone who labels the other person as an ‘avoider’ does not have to examine their own position. The questions that would be uncomfortable are sidestepped: Do I use closeness as insurance against anxiety? Does the other person experience my desire for clarity as pressure? Do I react to withdrawal with pursuit, thereby reinforcing precisely what I fear? Am I clinging to an unattainable person because their unattainability keeps a familiar form of hope alive?
Anyone who treats the other person as an ‘anxiously attached’ or ‘borderline’ individual likewise need not examine anything. Their protest becomes a symptom, and the question of whether one’s own behaviour was unclear, contradictory or hurtful is sidestepped.
Amidst such labels, the relationship disappears: its promises, its power dynamics, its mutual reinforcements. The more one side seeks reassurance, the further the other drifts away; the further they drift away, the more urgently reassurance is sought. This cycle does not require two ‘types’; it requires only two people and enough ambiguity.
That does not mean that both sides always contribute equally. There are relationships in which one person takes and does not give, initiates closeness but cannot tolerate contact, disappears during conflicts and then expects everything to carry on as before. The fact that a relationship can be asymmetrical does not, however, make the ‘type’ diagnosis any more accurate. It simply makes it more convenient.
Attachment does not relieve one of responsibility.
Early experiences explain why closeness is hard to bear. They do not negate the fact that everyone is responsible for how they treat others.
Those who cannot be available need not pretend to be so. Those who need time can say so. Those who do not want a relationship are entitled to say so. Those who are currently unable to address a conflict can acknowledge its existence and suggest a time to do so.
The same applies from the other side. Fear of loss is allowed. How this translates into demands, control, or constant contact remains a choice. A need for closeness does not justify surveillance or threats.
What matters is whether all this can still be communicated. Whether distance is a boundary or a punishment. Whether contact is a request or a compulsion. Whether a response is given after a conflict.
What takes the place of categorisation
The alternative to a hasty diagnosis is not endless conflict avoidance and indulgence. No one has to stay in a relationship in which they are constantly ignored, controlled or devalued.
The alternative lies in questions that can be answered. What was promised? What was given? What remained unanswered? What kind of closeness was possible, and what was not? What keeps repeating itself? Which actions cause harm, regardless of how they are explained? And why do I cling to a relationship in which my reality finds so few answers?
These questions are more uncomfortable than a ‘type’ label. They do not pass judgement on the other person’s character. Instead, they describe the situation one actually finds oneself in, and allow for a decision to be made.
‘He’s just an avoider’ offers none of that.
The most important points in brief
• The statement ‘He’s an avoider’ cannot be refuted by any response. Any reply that addresses what is meant confirms it. A statement of this kind merely categorises; it explains nothing.
• Ainsworth’s Strange Situation does not measure a relationship. The same child can be classified differently with their mother and father; the classifications are only weakly correlated.
• Even among adults, alongside general attachment orientations, there are relationship-specific working models. Attachment shifts with life circumstances, relationships and psychotherapy.
• Withdrawal may be a defence against dependency, a lack of interest, or a consequence of sensory overload and exhaustion in neurodivergent people. The behaviour alone does not determine this.
• Regarding the counter-figure: Fear of commitment is, on average, more strongly associated with borderline traits (r ≈ 0.48) than attachment avoidance (r ≈ 0.30). A correlation observed across samples does not justify a diagnosis in an individual case.
• Both labels absolve the person who assigns them and cause the relationship itself to disappear.
Sources
· Bowlby, J.: Attachment and Loss, Volume 1: Attachment, 2nd edition, Basic Books 1982
· Ainsworth, M. D. S., Blehar, M. C., Waters, E. & Wall, S.: Patterns of Attachment. A Psychological Study of the Strange Situation, Erlbaum 1978
· Bronfenbrenner, U.: ‘Towards an experimental ecology of human development’, American Psychologist 32 (1977), 513–531
· van IJzendoorn, M. H. & De Wolff, M. S.: In search of the absent father. Meta-analyses of infant–father attachment, *Child Development* 68 (1997), 604–609
· Spangler, G. & Grossmann, K. E.: Biobehavioural organisation in securely and insecurely attached infants,* Child Development* 64 (1993), 1439–1450
· Romantic attachment style and borderline personality pathology: A meta-analysis, Clinical Psychology Review 2020, PMID 31918217: https://www.sciencedirect.com/science/article/abs/pii/S0272735819302934
· World Health Organisation, ICD-11 for Mortality and Morbidity Statistics, 6D10 Borderline pattern: https://icd.who.int/browse11/l-m/en
Related
DESCRIPTION: Attachment styles have become a tool for relationship diagnosis on TikTok. Those labelled as ‘avoiders’ can hardly explain themselves: every answer merely confirms the label. What Bowlby and Ainsworth actually studied, and why the Strange Situation test does not measure relationships.
The Avoider
“He’s simply an ‘avoider’.” The phrase sounds like a realisation and feels like drawing a line under it. It transforms a person who withdraws in certain situations into a type about whom nothing more needs to be said.
What it’s all about:
· what attachment research has described,
· how this became a weapon, and
· why the same mechanism works in reverse for the woman who writes too much.
The statement against which there is no rebuttal
Anyone labelled as ‘avoidant’ has already lost.
If he says he needs some time to himself, this is seen as a retreat and confirms the label. If he says the relationship feels too close, it is seen as a rejection of intimacy and, in turn, confirms it. If he says he doesn’t love her and doesn’t want any commitment, this too is interpreted as a symptom: that’s how avoiders talk; they simply don’t realise it. Even the admission that he is probably an ‘avoider’ serves as evidence.
A statement that is confirmed by every conceivable response explains nothing. It labels. From that moment on, the other person is judged rather than listened to, and their view of their own situation is treated as material for a ‘diagnosis’.
That is precisely where the appeal lies. The sentence puts an end to the uncertainty about whether the other person is afraid, unable, unwilling, or asking too much. This uncertainty is hard to bear and involves hurt, shame and the possibility of having been wrong. And that is why the sentence strikes back, in three words.
What Bowlby and Ainsworth investigated
John Bowlby described a system that kicks in under stress: in the face of danger, separation, illness or being overwhelmed. Through repeated experiences, children form expectations about whether someone is available, whether help will come and whether they are allowed to express their own needs. Bowlby called these expectations ‘internal working models’. The term refers to models in the literal sense: assumptions that can be adjusted in the light of new experiences.
Mary Ainsworth investigated these expectations in the ‘Strange Situation’, a standardised procedure involving brief separations and reunions in an unfamiliar room. From the child’s behaviour upon the primary caregiver's return, she identified patterns, including an ‘avoidant’ one.
‘Avoidant’ described a strategy adopted under stress. A child who shows little distress upon reunion has learnt that openly expressing distress does not elicit a reliable response. They suppress their expression whilst retaining the need. Physiological measurements show increased arousal in these children, whilst they appear calm on the outside.
This is a statement about an adaptive response. It does not say anything about the character of an adult man.
The Strange Situation
This is where pop psychology falls short.
The same child may be classified differently in the Strange Situation when with their mother than when with their father. The meta-analysis by van IJzendoorn and De Wolff on father-child bonds shows that the classifications are only weakly correlated with either parent. What can be observed is a child’s behaviour towards a primary caregiver.
This means the category is already shaky at its very foundation. If a child reacts securely with one parent and avoidantly with the other, there is no such thing as an ‘avoidant child’. There is a relationship in which distress was responded to, and one in which it was not.
This pattern continues into adulthood. People experience varying degrees of security with parents, friends and partners. The same person may appear aloof in one relationship and affectionate in the next. In addition to general orientations, attachment research explicitly recognises relationship-specific models.
Urie Bronfenbrenner highlighted the methodological aspect as early as 1977, when he accused developmental psychology of having become the ‘science of the strange behaviour of children in strange situations with strange adults for the shortest possible time’. This statement highlights the issue of generalisation: a twenty-minute encounter in an unfamiliar room does not allow for any conclusions to be drawn about forty years of relationship life.
Social media
On TikTok, Instagram and in dating guides, these patterns have been distilled into four types of person: secure, anxious, avoidant, and anxious-avoidant. Anyone who messages a lot is anxiously attached. Anyone who doesn’t reply for three days after an intense weekend is avoidant. Anyone who does both is anxiously avoidant.
These categories promise something that research does not actually provide: insight into the future. Anyone who has identified the ‘type’ believes they know how the relationship will unfold and considers any further conversation a waste of time.
In doing so, they lose sight of the malleability inherent in the concept of the working model. Attachment styles shift with life circumstances, within stable relationships, following losses, and during psychotherapy. Review articles on the stability of attachment describe a relationship between continuity and change. A template that remains valid for forty years does not feature in them.
Withdrawal has many causes.
Withdrawal hurts, but it requires explanation. Someone withdraws because conflict immediately triggers shame and a loss of control. Because they experience a demand for commitment as a restriction. Because the other person is pressuring, controlling or belittling them. Because the relationship has failed to find a language for dealing with differences.
Someone may also withdraw because they do not want to invest. Because the matter is not important enough to them. Because they want to keep other options open. Because they want the benefits of a relationship without taking on its obligations.
For neurodivergent people, there is another possibility. Sensory overload, exhaustion following intense closeness, delayed processing of conflicts, a lack of time awareness or difficulties with self-organisation can manifest in relationships in the same way as indifference. An unanswered message is not in itself proof of attachment avoidance. This connection is complex enough to warrant a separate article. The only thing that matters here is that the same outward behaviour can have several causes.
The action itself does not answer the question of its function or the motives behind it. Whether a withdrawal serves to regulate an overload, to ward off dependency, to protect against overstimulation, or to exert control over the other person is determined by follow-up questions or, failing that, by subsequent behaviour.
The counterpart
The avoider rarely appears alone. A second figure is associated with them: the woman who writes too much, presses too soon, takes a lack of reply badly, and seeks contact all the more urgently after the other person has withdrawn.
She is initially labelled ‘anxiously attached’. As her protests grow louder, she is labelled ‘emotionally dependent’, then ‘controlling’, then ‘dramatic’. Ultimately, the most serious label looms large: ‘borderline’.
The division of roles is fixed. His aloofness is psychologised and thus explained; her protest is psychiatrised and thus dismissed. He cannot help himself; she cannot control herself. What used to be called hysteria is now referred to as a dysregulated attachment system.
Yet the underlying facts are the same. He gets in touch irregularly, alternates between closeness and distance in an unclear pattern, and avoids conversations about commitment. She asks for clarification. Who is considered disturbed in this scenario depends on what behaviour the culture deems acceptable in the first place.
From attachment style to personality disorder
There is a link between attachment insecurity and borderline traits, and it is unevenly distributed. A meta-analysis of romantic attachment styles and borderline pathology finds a moderate correlation of approximately r = 0.48 for anxious attachment and approximately r = 0.30 for avoidant attachment.
No diagnosis can be drawn from these figures. A correlation of this magnitude describes a clustering of cases across sample groups. It says nothing about any person, and certainly nothing about someone known only through chat histories and the account provided by the other party.
Borderline personality disorder requires a persistent, cross-situational pattern, significant distress or functional impairment, and the consideration of alternative explanations. Numerous messages, jealousy, the despair of a break-up and the desire for swift clarification fulfil none of these criteria. A conflict-ridden romantic relationship is not a clinical setting.
The harm caused by such ‘kitchen-table diagnoses’ is well documented. A borderline diagnosis is one of the most heavily stigmatised of all, and this stigmatisation is gendered: it reduces women to being dangerous, manipulative and unreasonable. Anyone who uses it in an argument is reaching for the harshest word available. The angry accusation of narcissism levelled at men is not much better.
Why the diagnosis provides relief
Diagnostic labels serve the same purpose for whoever applies them.
Anyone who labels the other person as an ‘avoider’ does not have to examine their own position. The questions that would be uncomfortable are sidestepped: Do I use closeness as insurance against anxiety? Does the other person experience my desire for clarity as pressure? Do I react to withdrawal with pursuit, thereby reinforcing precisely what I fear? Am I clinging to an unattainable person because their unattainability keeps a familiar form of hope alive?
Anyone who treats the other person as an ‘anxiously attached’ or ‘borderline’ individual likewise need not examine anything. Their protest becomes a symptom, and the question of whether one’s own behaviour was unclear, contradictory or hurtful is sidestepped.
Amidst such labels, the relationship disappears: its promises, its power dynamics, its mutual reinforcements. The more one side seeks reassurance, the further the other drifts away; the further they drift away, the more urgently reassurance is sought. This cycle does not require two ‘types’; it requires only two people and enough ambiguity.
That does not mean that both sides always contribute equally. There are relationships in which one person takes and does not give, initiates closeness but cannot tolerate contact, disappears during conflicts and then expects everything to carry on as before. The fact that a relationship can be asymmetrical does not, however, make the ‘type’ diagnosis any more accurate. It simply makes it more convenient.
Attachment does not relieve one of responsibility.
Early experiences explain why closeness is hard to bear. They do not negate the fact that everyone is responsible for how they treat others.
Those who cannot be available need not pretend to be so. Those who need time can say so. Those who do not want a relationship are entitled to say so. Those who are currently unable to address a conflict can acknowledge its existence and suggest a time to do so.
The same applies from the other side. Fear of loss is allowed. How this translates into demands, control, or constant contact remains a choice. A need for closeness does not justify surveillance or threats.
What matters is whether all this can still be communicated. Whether distance is a boundary or a punishment. Whether contact is a request or a compulsion. Whether a response is given after a conflict.
What takes the place of categorisation
The alternative to a hasty diagnosis is not endless conflict avoidance and indulgence. No one has to stay in a relationship in which they are constantly ignored, controlled or devalued.
The alternative lies in questions that can be answered. What was promised? What was given? What remained unanswered? What kind of closeness was possible, and what was not? What keeps repeating itself? Which actions cause harm, regardless of how they are explained? And why do I cling to a relationship in which my reality finds so few answers?
These questions are more uncomfortable than a ‘type’ label. They do not pass judgement on the other person’s character. Instead, they describe the situation one actually finds oneself in, and allow for a decision to be made.
‘He’s just an avoider’ offers none of that.
The most important points in brief
• The statement ‘He’s an avoider’ cannot be refuted by any response. Any reply that addresses what is meant confirms it. A statement of this kind merely categorises; it explains nothing.
• Ainsworth’s Strange Situation does not measure a relationship. The same child can be classified differently with their mother and father; the classifications are only weakly correlated.
• Even among adults, alongside general attachment orientations, there are relationship-specific working models. Attachment shifts with life circumstances, relationships and psychotherapy.
• Withdrawal may be a defence against dependency, a lack of interest, or a consequence of sensory overload and exhaustion in neurodivergent people. The behaviour alone does not determine this.
• Regarding the counter-figure: Fear of commitment is, on average, more strongly associated with borderline traits (r ≈ 0.48) than attachment avoidance (r ≈ 0.30). A correlation observed across samples does not justify a diagnosis in an individual case.
• Both labels absolve the person who assigns them and cause the relationship itself to disappear.
Sources
· Bowlby, J.: Attachment and Loss, Volume 1: Attachment, 2nd edition, Basic Books 1982
· Ainsworth, M. D. S., Blehar, M. C., Waters, E. & Wall, S.: Patterns of Attachment. A Psychological Study of the Strange Situation, Erlbaum 1978
· Bronfenbrenner, U.: ‘Towards an experimental ecology of human development’, American Psychologist 32 (1977), 513–531
· van IJzendoorn, M. H. & De Wolff, M. S.: In search of the absent father. Meta-analyses of infant–father attachment, *Child Development* 68 (1997), 604–609
· Spangler, G. & Grossmann, K. E.: Biobehavioural organisation in securely and insecurely attached infants,* Child Development* 64 (1993), 1439–1450
· Romantic attachment style and borderline personality pathology: A meta-analysis, Clinical Psychology Review 2020, PMID 31918217: https://www.sciencedirect.com/science/article/abs/pii/S0272735819302934
· World Health Organisation, ICD-11 for Mortality and Morbidity Statistics, 6D10 Borderline pattern: https://icd.who.int/browse11/l-m/en
Related