Carving the Me

1. Who is choosing at five?

Children of about five were asked which of two children they would choose as a friend, after hearing each speak. They chose the child who spoke with their own accent. When accent was set against race, they chose a native-accented child of another race over a foreign-accented child of their own race (Kinzler, Shutts, DeJesus and Spelke 2009). The same research group found the preference much earlier. Infants of five to six months, who cannot speak, looked longer at a person who had previously spoken their native language with a native accent. Infants of ten months reached for identical toys offered by a native speaker: in the United States the English speaker, in France the French speaker (Kinzler, Dupoux and Spelke 2007).

The preference comes before speech and before any choice the child could describe. By five, the child's social taste has been tuned by the language community it was born into. Whether that taste belongs to the child is a small version of a larger question: how much of who a person becomes is theirs to decide.

The philosopher Harry Frankfurt (1971) separated first-order desires, such as wanting a sweet, from second-order volitions, which are wants about which wants to have. He treated the second as essential to being a person, and he gave very young children as examples of "wantons", creatures with desires who do not yet want to have one desire rather than another. A five-year-old, on this view, can want things and is not yet an author of what to want.

The sociologist Charles Horton Cooley (1902) described the self as formed partly in how others appear to see it, which he called the looking-glass self. George Herbert Mead (1934) described the self as built in stages in which the child first imitates, then plays at roles, then takes the part of a "generalised other" (secondary summaries; Mead's texts have not been read in full here). At five, the shapers are the family, the language community and tuned preferences. The child's own hand is mostly imitation.

Some parts of identity, such as sexuality and neurodivergence, are ones where the idea of carving a self is contested and can do harm. Nothing here is a claim about them. Nothing here is a claim that people are to blame for who they are: the research described below shows change that is gradual and modest, and rarely a single decisive turn.

2. Twelve and seventeen

By twelve, peers carry more of the load. Penelope Eckert's study of a US high school found that its two main student groups, the Jocks and the Burnouts, differed in speech as well as in style. William Labov's study of Martha's Vineyard (1963) found that islanders who identified with the island used more of the island's local vowel forms, and those who disliked it used fewer. A later restudy found the pattern fading as the island's economy changed. In each case a group's way of speaking marked who belonged, and young people adopted it, mostly without deciding to.

Erik Erikson placed the task of "identity against role confusion" in adolescence, roughly twelve to eighteen. James Marcia (1966) described four identity statuses from two dimensions, exploration and commitment: diffusion (neither), foreclosure (commitments handed down, usually by parents or religion, with little exploration), moratorium (exploring without yet committing) and achievement (commitment after exploration). Foreclosure is the status in which someone else has done the carving. Textbooks note that the order of the statuses may be culturally biased.

At seventeen the law sets its age limits, and they do not agree with one another. In a multinational sample, cognitive capacity reached adult levels around age 16, while psychosocial maturity, the ability to restrain oneself in the face of emotional or risky stimuli, reached adult levels beyond 18. The authors argued that different age boundaries for different legal purposes are therefore reasonable (Icenogle et al. 2019). In England and Wales the age of full legal capacity fell from 21 to 18 under the Family Law Reform Act 1969, and in 1992 the criminal law's category of "young person" was widened to include seventeen-year-olds (Criminal Justice Act 1991). The same person can be an adult for one purpose and a minor for another. The YFL essay How the Law Moved the Line follows how the age limits of United Kingdom law moved.

3. What parents provide

Biographies often tie a famous person's identity to a parent. Behaviour genetics urges caution. Twin and adoption studies, including a meta-analysis of nearly all twin studies published over fifty years (Polderman et al. 2015), find that genes and the individual's own non-shared experiences account for most variation in adult personality, and the shared family environment accounts for little. Judith Rich Harris (1998) argued from this that peers matter more than parents. Plomin and Kawakami (2025) found that the shared-environment share of cognitive ability falls from about 20 to 30 per cent in childhood to near zero by early adulthood, while years of schooling keep a lasting shared influence of about 30 per cent. Parents still matter for access, for safety, and at the extremes: in rats, low maternal care and, in human post-mortem brain tissue, childhood abuse are linked to lasting changes in the stress-hormone system (section 4). Famous cases also suffer from survivorship, and their biographies were mostly written later by adults.

A more cautious use of the cases treats parents as tool-makers and gatekeepers who widen or narrow a child's room to act, not as proof of a parenting effect. The YFL framework of Warmth and Governance, set out in Family Climate, describes what children's home lives feel like from the inside, and In Other Words... Growing up happens at home gives the plain-speak account.

Wolfgang Mozart. Leopold Mozart taught his son and took him on tours of European courts from early childhood. At 25, in 1781, Wolfgang stayed in Vienna against his father's wishes, and he married Constanze Weber on 4 August 1782. Leopold's consent reached him after the wedding. The father supplied the tools, and the son later used them against his father's wishes.

Joan of Arc. At her trial in Rouen on 12 March 1431, Joan was asked whether she had done well to leave without her parents' permission. She said she had obeyed her father and mother in everything except her departure, that she later wrote to them, and that they forgave her. She also said she obeyed them in everything except in the case at Toul, a suit for her marriage. She reported that her mother had often told her that her father had dreamed of her going off with soldiers, and had said he would have her drowned if it came true. The account reaches the reader through the court's record and through what Joan said her mother said her father said.

Plato. It is assumed that Plato's father, Ariston, died while Plato was a boy. His mother, Perictione, is thought to have married her uncle Pyrilampes, and Plato was probably brought up chiefly in his house. Her brother Charmides and her cousin Critias, leaders in the oligarchic terror of 404 BC, were friends of Socrates, and through them Plato must have known Socrates from boyhood. Plato's own early ambitions were probably political. Socrates became the most important formative influence, though Plato may not have been among his intimates (Britannica). Here the family supplied the milieu and a chosen model supplied the direction.

Each case shows the same pattern: a parent or family sets what is available, and the young person's own choices then follow or oppose it.

4. What can be carved: four ways of inheriting

The biologists Eva Jablonka and Marion Lamb (2005, revised 2014) describe four systems of inheritance that all affect evolution: genetic, epigenetic (non-DNA transmission within and between cells), behavioural and symbolic (transmission through language and other symbols). They are a useful way to sort what a person can and cannot change.

Genetic: the sequence of DNA. An individual cannot change it. Twin studies find a substantial genetic contribution to personality, and the genes set an inherited range. Over many generations a population can change which genes are common, because organisms modify their environments and so change the selection pressures acting on them and their descendants. This is called niche construction (Odling-Smee, Laland and Feldman 2003). The best-studied human case is dairying, which created the selection pressure that spread the alleles for lactase persistence into adulthood (Gerbault et al. 2011). That is a population-level effect over many generations, and one person's choices do not change that person's own genes. People also tend to seek out environments that suit their inherited tendencies, which Plomin calls the "nature of nurture", so conditions that look chosen are partly shaped by genes (as reviewed in Tredoux 2019).

Epigenetic and hormonal: how genes are used. Epigenetic marks are chemical changes that affect whether and how a gene is read, without altering its sequence. In rats, mothers that groomed their pups more produced offspring with different epigenetic marks at a stress-hormone receptor gene in the hippocampus. The differences appeared in the first week of life, persisted into adulthood, followed the rearing mother when pups were cross-fostered, and were potentially reversible (Weaver et al. 2004). In post-mortem human hippocampus, suicide victims with a history of childhood abuse showed more methylation of the same kind of gene and lower expression of the receptor than other groups (McGowan et al. 2009, a small, associative study). People conceived during the Dutch Hunger Winter of 1944 to 1945 had less methylation of the IGF2 gene six decades later than their same-sex siblings (Heijmans et al. 2008). The timing of puberty also responds to conditions: the average age at menarche fell from 16.5 years in 1880 to about 12.5 in western societies (Hochberg, Gawlik and Walker 2011), and the move from juvenility to adolescence has been described as adaptive plasticity in response to energy resources and other cues (Hochberg and Belsky 2013).

Hormone release is mostly driven by context and is not under direct voluntary control. A widely publicised claim that holding "power poses" for two minutes raised testosterone and lowered cortisol (Carney et al. 2010) did not replicate in a larger study (Ranehill et al. 2015) or in later attempts, although the feeling of power did. Choice reaches hormonal settings indirectly, through conditions and company.

Whether such marks pass to the next generation is a separate question. Heard and Martienssen (2014) conclude that the inheritance of epigenetic marks occurs in plants and some animals, and that proof of a transgenerational epigenetic basis is generally lacking in mammals, where marks are largely reset in the early embryo and the germline. Whitelaw (2015) called the evidence for "Lamarckian" inheritance in mammals scant. The physiologist Denis Noble (2015) argues that findings in epigenetics mean the Modern Synthesis of evolution "requires either extension or replacement", and that in networks of interaction there is no privileged cause. In reply to critics he argues that the barrier between body and germline proposed by Weismann in 1892 has been broken, while critics reply that neo-Darwinian theory has never denied inherited epigenetic factors. The question remains disputed.

Behavioural: what a person does. Habits, skills, company and routines are the most carvable. In one pair of studies, people who wanted to raise a Big Five personality trait, such as extraversion or conscientiousness, tended to do so over sixteen weeks, and their daily behaviour changed to match it. The changes were modest but statistically significant (Hudson and Fraley 2015). Repeated practice seems to be the main route. Over a lifetime, personality becomes more stable with age: a review of 152 longitudinal studies found stability rising across the life course (Roberts and DelVecchio 2000), and a meta-analysis of 92 samples found people on average becoming more conscientious, emotionally stable and socially confident from their twenties to their forties (Roberts, Walton and Viechtbauer 2006). In early childhood the wiring itself is being completed. Brain volume rose by 101 per cent in the first year of life and 15 per cent in the second (Knickmeyer et al. 2008), and in rats, juveniles deprived of social play showed lasting differences in inhibitory connections in the prefrontal cortex and used simpler strategies on a learning task (Bijlsma et al. 2022). The YFL essay The Real Cost of Children sets out the biology of the juvenile period, and Play—the Brain's Natural Learning Environment looks at play in children. Care itself passes on through behaviour, since in the rat studies cross-fostering showed that mothering style was not simply inherited.

Symbolic: language, labels and accent. Accent is the clearest single example across the symbolic and behavioural systems, because it is neither inherited nor simply chosen. Infants of 6 to 8 months can distinguish sounds that are not in their own language, and by 10 to 12 months they mostly cannot, unless those sounds occur in the language they hear (Werker and Tees 1984). Vowel perception shows the home language's imprint by six months (Kuhl et al. 1992). Age matters for acquiring a new accent: six Canadian children aged 9 to 17 who moved to southern England acquired the new dialect with varying success, and Chambers (1992) concluded that age is critical. Among Italian-born speakers of English whose age of arrival in North America ranged from 2 to 22 years, perceived foreign accent rose in a linear relation with age of arrival, and many who arrived as young children still had a detectable accent (Flege, Munro and MacKay 1995). Steering by identity also operates. Speakers converge towards others to gain approval and diverge to assert identity: Adults learning Welsh emphasised their Welsh accent in English when an English speaker voiced a message that threatened their culture (Bourhis and Giles 1977). Le Page and Tabouret-Keller (1985) described speech as acts of identity, in which speakers shape their language to resemble the groups they want to be identified with. Labels and categories are carved in the same way, and section 5 returns to them.

Animals show something comparable. Sperm whales in clans have distinct dialects of clicks, which appear to be learned broadly within the clan and not only from mothers, with a bias to conform (Konrad et al. 2018; Cantor, Whitehead and colleagues). Chimpanzee "dialects" in pant-hoot calls were reported from communities in the Taï forest, with captive groups converging (Crockford et al. 2004), though a 2022 study of two communities found individual rather than group differences. These findings show learned group membership. They do not show choice.

Genetic
What it holds: the sequence of DNA; an inherited range
How much can be carved: Not by the individual. Changes across generations only through selection.

Epigenetic and hormonal
What it holds: how genes are used; stress-hormone settings; timing of puberty
How much can be carved: Set largely by conditions, not decided. Some marks are reversible in rats.

Behavioural
What it holds: habits, skills, company, routines; much of personality over time
How much can be carved: Most carvable, slowly, mainly by repeated practice.

Symbolic
What it holds: language, accent, labels, narrative
How much can be carved: Carvable within what the community offers, with windows that narrow.

5. Who defines the self: ascribed, achieved, labelled

The anthropologist Ralph Linton (1936) distinguished ascribed status, which a person is born into or given, from achieved status, which is won. In his era the ascribed statuses included sex, age, caste, class and race. Anthony Giddens (1991) argued that in late-modern societies the self has become "a reflexive project", open and made by the person through continual revision. Pierre Bourdieu argued that people acquire durable dispositions in childhood, including a linguistic habitus, from their social setting, which is embodied and acts as a kind of second nature, though one that is mutable and can generate new behaviour and not only reproduce old structures.

The two accounts disagree about how much of the self is made by the person. One makes self-construction an ongoing task, and the other makes the early settings of class and community run deep. Both are consistent with the evidence above: accents tuned in infancy, vowels chosen to signal belonging, and traits that shift modestly with practice.

Who gets to label whom is a further question. A child's category (a place, a class, a group, a label) is first applied by others, and Cooley's looking-glass self describes how it is taken in. Marcia's foreclosure describes the case where it is never examined. Public arguments about identity are partly arguments about who may assign a category, who may change it, and what changing it costs.

6. How much autonomy?

Autonomy here has two parts. One is capacity: what the person can understand and do at that age. The other is room: what the setting allows. Neither is fixed, and both change with age. Choosing conditions also need not be conscious or deliberate, as niche construction theorists note.

Early experience can reshape even a strong drive without the animal choosing anything. Kuo (1930) reared kittens with rats and concluded that, depending on how they were raised, a kitten could be made to kill a rat, befriend it, fear it or play with it. In a later study the kittens left their adult cage-mates alone but still killed and ate newborn and shaved rats, so the exception covered familiar individuals and not rats as a kind (Kuo 1938). Livestock guardian dogs show a similar pattern over a longer time: puppies are bonded to sheep during a critical period in early life, and the breeding lines have been selected for thousands of years so that they lack much of the predatory sequence (Coppinger and Feinstein 2015). The shaping was done to the animal. Early bonds also supply room to explore. In a test modelled on infant attachment research, about 64 per cent of kittens and 66 per cent of adult cats used their owner as a secure base, becoming less stressed when the owner returned and alternating between the owner and exploring the room (Vitale, Behnke and Udell 2019). The secure base of infant attachment research (Bowlby; Ainsworth) is the early bond from which exploration starts, and it is one source of the "room" described above. What a young person can add, and nothing in these studies suggests the animals do, is Frankfurt's second-order volition: wanting to have one want and not another.

The evidence supports a graded answer, not a number. Autonomy rises with age, and it varies with what is being carved. It is largest over habits and company, smaller over traits, set largely by conditions for hormonal and epigenetic settings, and smallest over genes. It is limited in two further ways. Conditions that seem chosen partly reflect inherited tendencies. And choosing conditions is often not a decision at all.

William James (1890) separated the "I" that knows from the "Me" that is known, and divided the Me into material, social and spiritual parts. Read this way, the "I" carves the "Me". Mead (1913, 1934) used the same words differently: the "Me" as the attitudes of others taken in, and the "I" as the response to them. Frankfurt's second-order volitions give a test for when carving becomes authorship: a person is the author of a want to the extent that they want to have it.

7. Back to the five-year-old

The five-year-old who prefers friends with a familiar accent has a taste that was tuned before the child could choose. By twelve the same child is a member of a group whose speech and style mark who belongs, and is trying commitments. By seventeen, many of the dispositions have set, the law has begun to treat the person as an adult for some purposes and not others, and the person has more room to act and more capacity to choose than at any earlier age.

The carved Me also shapes what comes next. Niche construction theory, and Jablonka and Lamb's four systems, describe how organisms alter the conditions their descendants develop in. For a person, this works largely through the early environment, which has epigenetic effects within a life, and through behaviour and culture, which are passed on by example and by language. Genes carry forward only through slow selection across generations, and the transmission of epigenetic marks across generations in humans is not established. What a person carves, then, travels mainly through the conditions of the next carving and not through the sequence of DNA.

How much of a life is authored by the person living it depends on the age and on what is being asked. Societies mark the point at which a person is credited with that authorship in different ways, through rites, through law and through custom, and the marks have shifted. The YFL essay Who Decided the Child Was Grown? follows those marks for girls and boys, rich and poor, the YFL essay How the Law Moved the Line follows the law's, and the YFL essay Counting a Child's Age looks at how a child's age has been counted and proved, which decides when those marks apply.

---

References

• Bijlsma, A. et al. (2022). Social play behavior is critical for the development of prefrontal inhibitory synapses and cognitive flexibility in rats. Journal of Neuroscience, 42(46), 8716–8728.
• Bourdieu, P. Writings on habitus and linguistic habitus, including Language and Symbolic Power (1991). (page-level citations not given; seen through secondary summaries)
• Britannica. Entry "Plato".
• Carney, D. R., Cuddy, A. J. C. and Yap, A. J. (2010). Power posing: brief nonverbal displays affect neuroendocrine levels and risk tolerance. Psychological Science, 21(10), 1363–1368.
• Chambers, J. K. (1992). Dialect acquisition. Language, 68(4), 673–705.
• Cooley, C. H. (1902). Human Nature and the Social Order. Scribner's.
• Criminal Justice Act 1991, s. 68 and Sch. 8; Family Law Reform Act 1969 (c. 46).
• Crockford, C., Herbinger, I., Vigilant, L. and Boesch, C. (2004). Wild chimpanzees produce group-specific calls: a case for vocal learning? Ethology, 110, 221–243.
• Eckert, P. (1989). Jocks and Burnouts: Social Categories and Identity in the High School. Teachers College Press.
• Erikson, E. H. (1968). Identity: Youth and Crisis. Norton.
• Flege, J. E., Munro, M. J. and MacKay, I. R. A. (1995). Factors affecting degree of perceived foreign accent in a second language. Journal of the Acoustical Society of America, 97(5), 3125–3134.
• Frankfurt, H. G. (1971). Freedom of the will and the concept of a person. Journal of Philosophy, 68(1), 5–20.
• Gerbault, P. et al. (2011). Evolution of lactase persistence: an example of human niche construction. Philosophical Transactions of the Royal Society B, 366(1566), 863–877.
• Giddens, A. (1991). Modernity and Self-Identity: Self and Society in the Late Modern Age. Polity.
• Bourhis, R. Y. and Giles, H. (1977). The language of intergroup distinctiveness. In H. Giles (ed.), Language, Ethnicity and Intergroup Relations, 119–135. London: Academic Press.
• Harris, J. R. (1998). The Nurture Assumption. Free Press.
• Heard, E. and Martienssen, R. A. (2014). Transgenerational epigenetic inheritance: myths and mechanisms. Cell, 157(1), 95–109.
• Heijmans, B. T. et al. (2008). Persistent epigenetic differences associated with prenatal exposure to famine in humans. Proceedings of the National Academy of Sciences, 105(44), 17046–17049.
• Hochberg, Z. and Belsky, J. (2013). Evo-devo of human adolescence: beyond disease models of early puberty. BMC Medicine, 11, 113.
• Hochberg, Z., Gawlik, A. and Walker, R. S. (2011). Evolutionary fitness as a function of pubertal age in 22 subsistence-based traditional societies. International Journal of Pediatric Endocrinology, 2011(1), 2.
• Hudson, N. W. and Fraley, R. C. (2015). Volitional personality trait change: can people choose to change their personality traits? Journal of Personality and Social Psychology, 109(3), 490–507.
• Icenogle, G. et al. (2019). Adolescents' cognitive capacity reaches adult levels prior to their psychosocial maturity: evidence for a "maturity gap" in a multinational, cross-sectional sample. Law and Human Behavior, 43(1), 69–85.
• Jablonka, E. and Lamb, M. J. (2005; rev. 2014). Evolution in Four Dimensions. MIT Press.
• James, W. (1890). The Principles of Psychology, ch. 10. Henry Holt.
• Kinzler, K. D., Dupoux, E. and Spelke, E. S. (2007). The native language of social cognition. Proceedings of the National Academy of Sciences, 104(30), 12577–12580.
• Kinzler, K. D., Shutts, K., DeJesus, J. and Spelke, E. S. (2009). Accent trumps race in guiding children's social preferences. Social Cognition, 27(4), 623–634.
• Coppinger, R. and Feinstein, M. (2015). How Dogs Work. University of Chicago Press.
• Knickmeyer, R. C. et al. (2008). A structural MRI study of human brain development from birth to 2 years. Journal of Neuroscience, 28(47), 12176–12182.
• Konrad, C. M. et al. (2018). Kinship and association do not explain vocal repertoire variation among individual sperm whales or social units. Animal Behaviour, 145, 131–140.
• Kuhl, P. K. et al. (1992). Linguistic experience alters phonetic perception in infants by 6 months of age. Science, 255(5044), 606–608.
• Kuo, Z. Y. (1930). The genesis of the cat's responses to the rat. Journal of Comparative Psychology, 11(1), 1–36.
• Kuo, Z. Y. (1938). Further study of the behavior of the cat toward the rat. Journal of Comparative Psychology, 25(1), 1–8.
• Labov, W. (1963). The social motivation of a sound change. Word, 19(3), 273–309.
• Le Page, R. B. and Tabouret-Keller, A. (1985). Acts of Identity. Cambridge University Press.
• Linton, R. (1936). The Study of Man. Appleton-Century.
• Marcia, J. E. (1966). Development and validation of ego-identity status. Journal of Personality and Social Psychology, 3(5), 551–558.
• McGowan, P. O. et al. (2009). Epigenetic regulation of the glucocorticoid receptor in human brain associates with childhood abuse. Nature Neuroscience, 12(3), 342–348.
• Mead, G. H. (1913). The social self. Journal of Philosophy, Psychology and Scientific Methods, 10; Mead, G. H. (1934). Mind, Self and Society. University of Chicago Press.
• Noble, D. (2015). Evolution beyond neo-Darwinism: a new conceptual framework. Journal of Experimental Biology, 218, 7–13; and Noble's reply "Central tenets of neo-Darwinism broken" and the correspondence "Neo-Darwinism is just fine" (Journal of Experimental Biology, 2015, 218, 2658–2659).
• Odling-Smee, F. J., Laland, K. N. and Feldman, M. W. (2003). Niche Construction: The Neglected Process in Evolution. Princeton University Press.
• Plomin, R. and Kawakami, K. (2025). Nurture and nonshared environment in cognitive development. Psychological Review.
• Polderman, T. J. C. et al. (2015). Meta-analysis of the heritability of human traits based on fifty years of twin studies. Nature Genetics, 47(7), 702–709.
• Ranehill, E. et al. (2015). Assessing the robustness of power posing: no effect on hormones and risk tolerance in a large sample of men and women. Psychological Science.
• Roberts, B. W. and DelVecchio, W. F. (2000). The rank-order consistency of personality traits from childhood to old age: a quantitative review of longitudinal studies. Psychological Bulletin, 126(1), 3–25.
• Roberts, B. W., Walton, K. E. and Viechtbauer, W. (2006). Patterns of mean-level change in personality traits across the life course: a meta-analysis of longitudinal studies. Psychological Bulletin, 132(1), 1–25.
• Tredoux, G. (2019). Two new books on heredity and behaviour genetics: a review (Plomin, Blueprint, 2018; Mitchell, Innate, 2018). Online review.
• Vitale, K. R., Behnke, A. C. and Udell, M. A. R. (2019). Attachment bonds between domestic cats and humans. Current Biology, 29(18), R864–R865.
• Weaver, I. C. G. et al. (2004). Epigenetic programming by maternal behavior. Nature Neuroscience, 7(8), 847–854.
• Werker, J. F. and Tees, R. C. (1984). Cross-language speech perception: evidence for perceptual reorganization during the first year of life. Infant Behavior and Development, 7, 49–63.
• Whitelaw, E. (2015). Disputing Lamarckian epigenetic inheritance in mammals. Genome Biology, 16, 60.