Carol Gilligan’s 1982 book In a Different Voice reshaped how psychologists and philosophers think about moral reasoning by arguing that the dominant frameworks of the field systematically undervalued an entire mode of ethical thinking. Gilligan proposed that alongside the well-studied “justice orientation,” there exists a distinct “care orientation” focused on relationships, responsiveness, and responsibility to others. The idea that people reason morally in fundamentally different voices has since rippled through fields far beyond psychology, from nursing ethics to artificial intelligence design. And the evidence, while not landing exactly where Gilligan predicted, has opened up questions about culture, context, and identity that remain unresolved.
The Care Voice and the Justice Voice
Before Gilligan, the dominant model of moral development treated reasoning about rights, fairness, and abstract principles as the pinnacle of ethical maturity. Gilligan challenged this by identifying a second orientation: the care voice, which prioritizes maintaining relationships, attending to the needs of specific people, and accepting responsibility for those who depend on you. She argued that this care orientation had been systematically devalued in moral psychology because the field’s key theories were built almost entirely on studies of men and boys.
Gilligan’s framework positioned these two voices not as opposites but as distinct lenses through which people interpret moral problems. The justice voice asks, “What is the fair rule here?” The care voice asks, “Who will be hurt, and what do they need?” She contended that the care orientation was more commonly found in women, not because of biology, but because of how girls are socialized to attend to relationships.1PubMed. Justice and care: the implications of the Kohlberg-Gilligan debate for medical ethics This reframing was enormously influential. Within a few years, researchers across moral philosophy, political theory, and applied ethics had begun developing the “morality of care” as a full-fledged alternative to the justice tradition that had dominated Western moral thought for centuries.2Criminal Justice Review. The Morality of Justice and the Morality of Care: Are there Distinct Moral Orientations for Males and Females?
What the Data Actually Show About Gender
Gilligan’s thesis made a testable prediction: women should lean toward care reasoning, and men toward justice reasoning. Decades of research have not strongly supported that prediction. A meta-analysis published in Psychological Bulletin that pooled the results of numerous studies on gender and moral orientation found only small differences. Women scored slightly higher on care reasoning, and men scored slightly higher on justice reasoning, but both effects were modest. The analysis concluded that these findings “do not offer strong support for the claim that the care orientation is used predominantly by women and that the justice orientation is used predominantly by men.”3PubMed. Gender differences in moral orientation: a meta-analysis
This does not mean Gilligan was wrong about everything. The existence of two distinguishable moral orientations has held up reasonably well. Most researchers now accept that care and justice represent real, somewhat independent ways of thinking about ethical dilemmas. Where the evidence gets thin is in tying those orientations tightly to gender. The differences exist on average, but they are small enough that knowing someone’s gender tells you very little about which moral voice they will use in a given situation. Context, the specific dilemma, and cultural background all seem to matter more.
Culture Shapes the Moral Voice More Than Gender Does
One of the more revealing challenges to Gilligan’s original framing has come from cross-cultural research. When researchers tested Gilligan’s ethic of care across different societies, the pattern shifted in ways that gender alone could not explain. A study comparing students from Brazil and Norway using a standardized measure of care reasoning found that care orientation scores were unrelated to gender in both samples. What did predict differences was culture: Brazilian participants scored lower on care reasoning, and this was linked to their higher scores on collectivism.4Journal of Moral Education. Note on a cross‐cultural test of Gilligan’s ethic of care
That finding is counterintuitive at first glance. You might expect collectivist cultures, which emphasize group harmony, to score higher on care. But the researchers suggested that the specific framing of Gilligan’s care orientation, with its emphasis on individual relationships and personal moral growth, may itself reflect Western individualist assumptions about what “caring” looks like. In collectivist cultures, care can be embedded in duty, hierarchy, and community obligation rather than in the kind of personal relational negotiation that Gilligan described.
Broader reviews of cross-cultural moral reasoning reinforce this picture. Societies develop moral systems that emphasize different orientations depending on their history, religious traditions, economic structures, and social organization.5PubMed Central. Universality and Cultural Diversity in Moral Reasoning and Judgment The “different voice” Gilligan heard may be less about being female and more about being embedded in certain kinds of social relationships, wherever those relationships happen to be emphasized.
How Children Grow Into Moral Reasoning
If care and justice are not neatly divided by gender, where do they come from developmentally? Research tracking children from early childhood into adolescence suggests that emotional capacities and cognitive reasoning become increasingly intertwined as children grow. A longitudinal study following over 1,200 Swiss children from age six to age twelve found that sympathy in younger children predicted later increases in moral reasoning and moral emotion, but not the other way around. Both sympathy and moral reasoning in early and middle childhood were associated with stronger social justice values by age twelve.6PubMed. Developmental relations between sympathy, moral emotion attributions, moral reasoning, and social justice values from childhood to early adolescence
The picture that emerges is that the care voice and the justice voice are not competing developmental paths. They are interrelated capacities that grow together. A child who feels genuine sympathy for others is building the emotional scaffolding that later supports more sophisticated moral reasoning. This fits with broader evolutionary evidence suggesting that empathy is phylogenetically ancient, probably shared across mammals and birds, and that it serves as the foundation for more complex forms of concern and perspective-taking.7Annual Review of Psychology. Putting the altruism back into altruism: the evolution of empathy In other words, the different voice of care may not be an alternative developmental track at all. It may be the emotional bedrock on which all moral reasoning is built.
Care Ethics in Healthcare and Nursing
Regardless of how the gender debate has played out in psychology, Gilligan’s framework has had lasting practical impact in professional ethics, particularly in nursing. Nurses have long argued that the traditional justice-based frameworks of bioethics fail to capture what makes their work morally distinctive. Drawing on Gilligan and the philosopher Nel Noddings, many nursing ethicists have argued that a care orientation better reflects the moral texture of clinical practice: the attention to particular patients, the responsiveness to suffering, the relational nature of healing.
This application has not been without pushback. Some bioethicists have argued that while care in the sense of sensitivity and responsiveness to individuals is necessary for ethical practice, it is not sufficient on its own. Ethics still needs justice, the argument goes, because care without principles can become paternalistic, inconsistent, or limited to those we happen to feel close to.8PubMed. Clinical ethics and nursing: “yes” to caring, but “no” to a female ethics of care The most productive thinking in healthcare ethics now tends to treat care and justice as complementary rather than competing, which is probably where Gilligan’s framework is most useful in practice.
Care Ethics Meets Artificial Intelligence
One of the more surprising places Gilligan’s “different voice” has traveled is into the design and governance of technology. As AI systems take on roles in healthcare, eldercare, and education, researchers have begun asking whether care ethics can provide something that the dominant frameworks in tech ethics miss.
The argument is that standard approaches to AI ethics tend to focus on fairness, transparency, and individual rights, which are essentially justice-orientation concepts. These matter, but they can miss the relational dimension of what happens when a machine replaces or mediates a human caring relationship. A care-ethics perspective highlights that AI does not simply replace tasks; it reorganizes the conditions under which authentic caregiving happens. When an algorithm triages patients or a robot assists an elderly person, the moral labor does not disappear. It shifts, often in ways that reduce the space for genuine responsiveness to individuals.9PubMed Central. Care ethics and the transformation of care in an age of artificial intelligence
This line of thinking has been applied specifically to the design of care robots. Researchers have proposed that care ethics should be the central ethical framework for designing robots that interact with vulnerable populations, arguing that it provides concrete principles sensitive to the contexts of caring practices and that it aligns well with the preservation of trust between caregiver and recipient.10PubMed Central. Trust in and Ethical Design of Carebots: The Case for Ethics of Care Whether this approach will actually influence how tech companies build products is another question, but the theoretical framework is being taken seriously in academic and policy discussions.
The Literal Different Voice
Gilligan’s “different voice” was a metaphor, but the literal phenomenon of speaking in a different voice is its own rich area of research. People modulate their voices constantly and often unconsciously, adjusting pitch, pace, and tone depending on whom they are speaking to and what they are trying to accomplish.
Pitch is the most perceptually obvious dimension. Lower-pitched voices are consistently perceived as more dominant and authoritative, and meta-analytic evidence links lower pitch to higher testosterone and greater upper-body strength, though the actual relationships are modest.11PubMed. Voice pitch: a window into the communication of social power What makes this interesting is that people do not just have a fixed pitch. They actively shift it based on social context. Experimental research has shown that men who believe they are physically dominant to a competitor lower their pitch when speaking to that person, while men who believe they are less dominant raise their pitch.12Evolution and Human Behavior. Dominance and the evolution of sexual dimorphism in human voice pitch This is a literal “different voice” deployed strategically, often without the speaker being fully aware of it.
Both the fundamental frequency of the voice and the spacing of vocal resonances influence how dominant a speaker sounds, though research suggests that resonance characteristics may actually have a larger effect on dominance perceptions than pitch alone.13Evolution and Human Behavior. Men’s voices as dominance signals: vocal fundamental and formant frequencies influence dominance attributions among men In romantic and courtship contexts, people shift their voices in different ways. There is evidence that attraction and romantic interest can be distinguished through vocal tone alone, and that individuals purposely change their vocal behavior to sound more appealing to potential partners.14PubMed Central. Vocal modulation in human mating and competition The voice you use when flirting is measurably different from the voice you use when competing.
Speaking a Different Language, Becoming a Different Person
Bilingual individuals often report feeling like a slightly different person in each of their languages. Research suggests this is not just subjective. A study of Korean-English bilinguals found that participants’ self-reported personality traits shifted depending on which language they were tested in. They scored higher on agreeableness when tested in English and higher on emotional sensitivity when tested in Korean. Openness scores also shifted, and the size of that shift depended on how frequently participants used each language.15Journal of Applied Language Learning. The Language Effect: Investigating the Impact of Language Switching on Personality in Korean-English Bilinguals
Interestingly, the acoustic side of bilingual voice shifts is less dramatic than the personality side. A study of Russian-English bilinguals found no overall difference in how participants perceived their own communication characteristics across their two languages. The one significant acoustic difference was that participants spoke faster and used longer sentences in English than in Russian.16Journal of Voice. Self-perceived and Acoustic Voice Characteristics of Russian-English Bilinguals The implication is that the “different voice” of bilingualism may operate more at the level of identity and self-concept than at the level of raw acoustics. You may not sound radically different in your second language, but you may think and feel differently while speaking it.
When the Voice Reveals What Words Do Not
Voice modulation is not always strategic or social. Sometimes the way a person’s voice changes is a signal of something happening beneath conscious awareness. Research on depression has shown that vocal characteristics shift in measurable ways as depression worsens. People with more severe depression tend to speak more slowly, pause more frequently, and show greater variability in their pause patterns. Several of these acoustic markers correlated with clinician-rated depression severity, including total pause time, pause variability, and speaking rate.17PubMed Central. Vocal Acoustic Biomarkers of Depression Severity and Treatment Response
The clinical value here is that these vocal changes can be detected from recorded speech samples without the person needing to report how they feel. For conditions like depression, where self-report is unreliable and clinical access is limited, the voice itself becomes a diagnostic window. Researchers are exploring whether automated speech analysis could eventually track treatment response or flag worsening symptoms between clinical visits. The voice, in this framing, is not speaking in a metaphorically different register. It is literally, acoustically transformed by emotional state.
Gendered Voices in Machines
The politics of voice have taken on a new dimension with the rise of voice assistants. When companies design synthetic voices for AI products, they are making choices about gender, authority, and social role that embed cultural assumptions into technology. Acoustic analysis of major voice assistants has found that the vocal qualities chosen for different products align with distinct gendered personas. A breathy voice indexes a traditional domestic femininity, while a lower pitch with creaky voice qualities aligns with a more professional, authoritative femininity. These vocal features reflect cultural beliefs about what voices should sound like in different roles and contexts.18AI & SOCIETY. Gender by proxy: echoes of femininities in synthetic voice design
This is where Gilligan’s metaphor loops back to the literal. The “different voice” she identified in moral reasoning was shaped by socialization and cultural expectations about gender. The different voices engineered into our devices are shaped by market research and design ideology about what consumers expect from a helpful, trustworthy, or authoritative presence. In both cases, the voice is not neutral. It carries social meaning, encodes power dynamics, and reflects assumptions about who should sound like what. The question of whose voice gets heard, and how that voice is allowed to sound, turns out to be as much a design problem as a philosophical one.