Ethics

Immanuel Kant (1724–1804) posed a radical question: can a moral act be determined by its consequences? His answer is a categorical no. If we lie for good and it leads to positive outcomes, our action does not become moral. Morality lies in intention, in the principle we act by, not in the outcome.

Kant builds ethics on the concept of duty. The only unconditionally good thing is a good will: a will acting from duty, not from inclination or calculation. A person helping others because he enjoys it acts kindly, but not morally in the strict sense. It is moral to help because it is your duty, ...

Kant proposes the categorical imperative—the supreme principle of morality from which all concrete duties are derived. It has several equivalent formulations.

First formulation (formula of universal law): “Act only according to that maxim whereby you can at the same time will that it should become a universal law.” The test: take the principle of your action (the maxim) and ask—what happens if everyone acts that way? If a contradiction arises, the acti...

Definitions

Act utilitarianism

— each specific action is evaluated according to its consequences. This is maximally flexible, but creates a problem: sometimes calculation requires actions that seem clearly immoral.

The problem of distribution

— utilitarianism cares about the sum of happiness, not its distribution. A society where a million people are happy at the expense of a thousand suffering slaves may have a high “moral balance.” John Rawls called this a fundamental flaw: utilitarian...

The problem of rights

— utilitarianism does not recognize rights in the strict sense. Any right may be violated if it maximizes good. Yet rights make sense precisely when they protect against the “tyranny of the majority”—when the majority wants to infringe on a minority...

The problem of measurement

— how can we measure and compare the happiness of different people? How to compare the happiness from music to that from food? Utilitarian calculation presupposes what is practically impossible—precise interpersonal comparison of utility.

Jeremy Bentham (1748–1832) and John Stuart Mill (1806–1873) created utilitarianism—the most influential ethical theory in the Anglo-Saxon world. Its central principle: the right action is that which produces the greatest happiness for the greatest number of people.

Bentham proposed the hedonistic calculus—a quasi-mathematical instrument for measuring pleasures and pains according to their intensity, duration, certainty, propinquity, fecundity, purity, and extent (how many people are affected). Add up all pleasures, subtract all pains—and you get the "moral ...

Mill refined the theory: not all pleasures are equal in quality. “It is better to be a dissatisfied Socrates than a satisfied pig.” Higher pleasures (intellectual, moral, aesthetic) surpass lower ones, even at lesser intensity. This makes utilitarianism more in tune with our intuitions.

Act utilitarianism: each specific action is evaluated according to its consequences. This is maximally flexible, but creates a problem: sometimes calculation requires actions that seem clearly immoral.

Definitions

Ethos

— literally "character" or "habit." Ethics is the science of character. A virtuous person does not struggle with themselves to act well (this would be a sign of incomplete education). They act well easily, with pleasure—because that is their character.

While Kant asks, "What ought I do?", and the utilitarians, "What are the consequences?", Aristotle poses a different question: "What kind of person should I be?" This is a shift from actions to character, from rules to virtues.

Aristotle begins the "Nicomachean Ethics" with an observation: everything strives toward its own good, and the highest good for a human is eudaimonia (εὐδαιμονία). This word is translated as "happiness," but more accurately as "flourishing," "well-being," "a life lived well." Eudaimonia is not a ...

A person achieves eudaimonia by living in accordance with their nature—and the nature of a person as a rational being implies a life in accordance with virtues: stable dispositions to feel, desire, and act in the right way.

Aristotle defines each virtue as the mean between two vices—an excess and a deficiency. Courage is the mean between cowardice (deficiency) and recklessness (excess). Generosity is between stinginess and wastefulness. Honesty is between lying and boastfulness.

Definitions

1. Autonomy

— the right of the patient to make decisions about their own treatment based on informed consent. The patient is not an object of treatment but a subject participating in decision-making. This requires: disclosure of information, understanding, volu...

2. Beneficence

— to act in the best interests of the patient. Medicine exists for the benefit of the sick. But who defines "benefit"? The conflict between what the doctor considers beneficial and what the patient wants is the main bioethical collision.

3. Non-maleficence

— "first, do no harm." The principle requires avoiding actions whose risks exceed the expected benefits. But medical interventions almost always involve risks—the boundary is blurred.

4. Justice

— the fair allocation of medical resources. Who receives a transplant when there are fewer organs than those in need? On what basis—medical necessity? Social value? Lottery?

Euthanasia

— passive (cessation of treatment) and active (administration of a lethal dose). The Netherlands, Belgium, and Switzerland have legalized one form or another. Arguments for: respect for the autonomy of the suffering patient, prevention of senseless ...

Prenatal diagnosis and abortion

— the possibility of discovering genetic abnormalities in the fetus before birth raises the question of "designing" children. Eugenic connotations are not a historical curiosity but a real issue.

Organ distribution

— transplant waiting lists are an ethically sensitive area. Criteria: medical compatibility, urgency, waiting time, adherence to regimen, social factors? Each criterion carries a value judgment.

Bioethics emerged in the 1960s–70s in response to the rapid development of medical technologies, which raised questions going beyond the boundaries of traditional medical ethics. Organ transplantation, life-support machines, DNA tests, genetic engineering—all this created situations where medicin...

The first bioethical scandals set the tone. Tuskegee (1932–1972): four hundred African Americans with syphilis were observed without treatment even after the discovery of penicillin—in order to study the natural course of the disease. This violated the principle of informed consent and respect fo...

Tom Beauchamp and James Childress, in "Principles of Biomedical Ethics" (1979, 8 editions), formulated four principles that became the standard:

1. Autonomy—the right of the patient to make decisions about their own treatment based on informed consent. The patient is not an object of treatment but a subject participating in decision-making. This requires: disclosure of information, understanding, voluntariness, capacity.

Definitions

Algorithmic bias

— machine learning systems are trained on data reflecting historical inequities. If historically certain groups were more often found guilty, the algorithm will "learn" to treat these groups in the same way—reproducing rather than correcting injusti...

Transparency and explainability

— the "black box" of neural networks is opaque. When an algorithm denies a loan, the patient or client has the right to know why. The European GDPR establishes a "right to explanation." But technical explainability of complex models is an unsolved i...

The problem of responsibility

— if a self-driving car hits a pedestrian—who is at fault? The manufacturer? The programmer? The owner? The algorithm is not a moral subject. Yet traditional legal concepts of responsibility are created for moral subjects—humans.

Artificial intelligence is not just a technology that speeds up existing processes. It is a technology that makes decisions—or imitates decision-making—in situations where people traditionally acted. When an algorithm denies a loan, recommends a sentence, or drives a car, questions arise that wer...

Algorithmic bias: machine learning systems are trained on data reflecting historical inequities. If historically certain groups were more often found guilty, the algorithm will "learn" to treat these groups in the same way—reproducing rather than correcting injustice. The COMPAS system (recidivis...

Transparency and explainability: the "black box" of neural networks is opaque. When an algorithm denies a loan, the patient or client has the right to know why. The European GDPR establishes a "right to explanation." But technical explainability of complex models is an unsolved issue.

The problem of responsibility: if a self-driving car hits a pedestrian—who is at fault? The manufacturer? The programmer? The owner? The algorithm is not a moral subject. Yet traditional legal concepts of responsibility are created for moral subjects—humans.

In his article “Famine, Affluence, and Morality” (1972), Peter Singer posed a question that changed the discussion of global ethics: if you can prevent something bad (the death of a child from malnutrition) without sacrificing anything morally significant — you are obliged to do so. The distance ...

The consequence is radical: wealthy people in developed countries are obliged to give a substantial portion of their income to help the poorest in the world — up to the point where the next dollar given to charity brings you more benefit than harm. This goes much further than traditional notions ...

Thomas Pogge suggested a different approach: global poverty is not simply a misfortune that we can alleviate. It is the result of an unjust international system that wealthy countries support: trade rules, debt regimes, support for corrupt regimes. We are not just failing to help the poor — we ar...

When a transnational corporation uses suppliers in a country with low labor standards, who is responsible for violations of the rights of workers?

Definitions

Moral realism

— Yes, there are objective moral facts. “Torturing a child for pleasure is wrong” is true regardless of whether anyone believes it. Plato (“Forms of the Good”), G. E. Moore (“non-natural moral properties”), Derek Parfit (“objective reasons”).

Moral anti-realism

— There are no objective moral facts. Sub-variants:

Normative ethics answers: what should we do? Meta-ethics asks: what does it mean for “something to be right”? Are there moral facts independent of our beliefs? Or is morality a construct, a projection, an emotion?

Moral realism: Yes, there are objective moral facts. “Torturing a child for pleasure is wrong” is true regardless of whether anyone believes it. Plato (“Forms of the Good”), G. E. Moore (“non-natural moral properties”), Derek Parfit (“objective reasons”).

If there is no moral realism—how can one criticize genocide, slavery, corruption from a standpoint of “objective evil”? If realism exists—how to explain moral progress and disagreement?

Kantian constructivism: moral principles are not discovered facts of nature, but that which rational beings would construct under certain conditions. The categorical imperative is not a description of reality but a principle that autonomous reason gives itself.

John Rawls (1980) developed this: “Constructivism in Moral Theory” — principles of justice are what we would construct in the “original position” behind a “veil of ignorance.” Morality is the result of a hypothetical contract, not a discovery of objective facts.

Constructivism explains why ethical principles can change and improve: we continue to negotiate about which principles we would accept. This is not relativism — some constructions are better than others because they are more coherent, inclusive, and rationally justified.

For organizations: corporate values are a construct. They are real to the extent that all stakeholders are genuinely involved in them.

Joshua Greene and neuroethics: moral judgments rely on two processes — fast intuitive (emotional) and slow rational. The trolley problem activates different neural networks depending on whether it is necessary to physically push a person.

Jonathan Haidt: “moral intuitionism” — most of our moral judgments happen quickly and intuitively, and rational arguments are retrospective justifications. We do not reason our way to a conclusion — we feel the conclusion and then rationalize it.

Moral intuitions are data. They are the result of evolutionary and cultural learning. Sometimes they are mistaken (intuition saw nothing wrong with slavery in some cultures). But completely ignoring intuition in favor of “pure theory” is also a mistake.

Rawls’s reflective equilibrium: move between theory and intuition, correcting both.

Definitions

For

— elimination of hereditary diseases (Huntington’s, cystic fibrosis, sickle cell anemia). Against: cannot obtain the subject’s consent; unpredictable long-term consequences; risk of "genetic stratification" of society; where is the boundary between ...

CRISPR-Cas9 enables genome editing with unprecedented precision. In 2018, Chinese scientist He Jiankui announced the first genetically edited children — twins Lulu and Nana. International scientific reaction: condemnation. Why?

Editing somatic cells (adult human) — therapy. Editing the germline (embryo) — alteration of the inheritable genome. This affects future people who could not give consent. Is this "designer children"? Genetic advantage for the wealthy?

For: elimination of hereditary diseases (Huntington’s, cystic fibrosis, sickle cell anemia). Against: cannot obtain the subject’s consent; unpredictable long-term consequences; risk of "genetic stratification" of society; where is the boundary between treatment and "enhancement"?

Data are not neutral. They are collected for specific purposes, stored by specific entities, and used in specific contexts. Three key ethical questions: consent, transparency, fairness.

Predictive policing algorithms (PredPol): trained on historical arrest data—which reflects existing biases. Result: increased patrols in areas with historically more arrests → more arrests → more “confirmation” for the algorithm. A self-reinforcing biased cycle.

What does a “fair algorithm” mean? Mathematically incompatible definitions of fairness: equal accuracy for all groups vs equal false positive rates vs equal false negative rates. It is impossible to optimize all at the same time.

Peter Singer (1972): if I can prevent something bad without sacrificing anything of comparable importance — I am morally obliged to do so. Distance is irrelevant: a drowning child in front of you and a child dying of a preventable disease on the other side of the world require your intervention e...

Consequence: people in wealthy countries have a moral obligation to give significantly more in aid to the poor. Effective altruism is built on this foundation: give where the maximum effect per unit of money is achieved.

"This is too demanding": Singer replies — morality is sometimes demanding. "It destroys market incentives": Singer — this is about personal morality, not about policy. "Structural changes are more important than charity": possibly, but that does not mean charity is unimportant.

John Rawls' "A Theory of Justice" (1971) is a book that changed academic political philosophy and had a tremendous influence on real politics. After decades of utilitarian dominance ("maximize total happiness"), Rawls proposed an alternative based on rights and justice.

Rawls poses the question: on what principles should a just society be built? His method: a thought experiment with the "veil of ignorance". Imagine you have to choose the principles for structuring society, not knowing your place in it — whether you are rich or poor, male or female, a member of t...

Rawls asserts: rational agents behind the veil of ignorance will choose two principles. The first is the principle of equal liberty: everyone is entitled to maximum freedom compatible with the same freedom for others. The second is the difference principle: social and economic inequalities are pe...

This is not egalitarianism in the sense that "everyone should have equally". Inequality is permissible — if it improves the position of the poorest. Progressive taxation, compulsory education, a system of social insurance — all of these are justified by the difference principle.

On December 10, 1948, the UN adopted the Universal Declaration of Human Rights. After the Holocaust and World War II, the international community decided: there exist rights that cannot be violated under any circumstances — regardless of the state, culture, or religion. These include the right to...

Philosophical foundation: human rights are not a gift from the state or religion, but an attribute of human dignity as such. "All human beings are born free and equal in dignity and rights" — Article 1. This is the Kantian argument: every person is an end in themselves, not merely a means.

Can we declare rights "universal" if they are formulated predominantly by Western authors? In 1948, most countries in the world were colonies — and their peoples did not participate in the drafting of the Declaration. Is it universal or cultural imperialism?

Concrete tensions. The right to freedom of religion vs. blasphemy laws in many countries. The right to equal treatment regardless of gender vs. religious practices discriminating against women. The right to property vs. communal forms of ownership among indigenous peoples. The right to freedom of...

Socrates refused to escape from prison, although the law was unjust — because he believed it was a duty to obey the laws of the state. Martin Luther King sat in Birmingham Jail for violating the segregation law — because he considered it unjust. Both positions deserve serious consideration.

Henry David Thoreau (“Civil Disobedience”, 1849) formulated the classic justification: when the law is unjust, the duty of a decent person is to break it. Thoreau refused to pay tax in support of slavery and the Mexican War and spent a night in jail (his aunt paid, to his disappointment). “The be...

Mahatma Gandhi developed the concept of nonviolent resistance — satyagraha (“the force of truth”). Principles: break an unjust law openly, accept the consequences (arrest, violence), do not respond to violence with violence. This is not passivity — it is active pressure through moral superiority.

Salt March (1930): Gandhi walked 380 km to the sea to break the British monopoly on salt. Thousands followed him. The British arrested participants — and each arrest was a public defeat for colonial authority. This is strategically brilliant: provoke the system to violate its own principles.

Hannah Arendt (1906–1975) went to the trial of Adolf Eichmann — one of the organizers of the Holocaust — as a correspondent for the New Yorker. She expected to see a monster. She saw something even more terrifying: an ordinary person. An official who thought in terms of "task completed", "instruc...

This observation became one of the most discussed philosophical theses of the 20th century: "the banality of evil". Eichmann was not demonically evil — he was frighteningly ordinary. He did not think. The greatest sin, in Arendt's view, is the refusal to think about the consequences of one's acti...

Arendt did not justify Eichmann. She revealed something more important: the idea that evil is committed only by "evil people" is a dangerous illusion. Most Nazi crimes were committed by ordinary people, following orders, performing their "duties", not thinking about the consequences.

Milgram's experiment (1961, precisely during Eichmann's trial) confirmed this empirically: 65% of participants were willing to apply dangerous electric shocks to another person — simply because an authoritative experimenter said "continue". Authority, form, routine — powerful forces, switching of...

Should Germany be held accountable for the Holocaust? Should the USA be held accountable for slavery? Should Russia— for Soviet crimes? Should Belgium— for Congo? These are questions of the ethics of historical responsibility— one of the most controversial topics in modern political philosophy.

The key distinction: collective guilt and collective responsibility are different things. Karl Jaspers (“The Question of German Guilt”, 1946) distinguished: criminal guilt (of specific individuals, for specific actions), political guilt (citizens of the state which allowed crimes), moral guilt (p...

Germany has paid reparations to Israel and Holocaust survivors since 1952— around $80 billion. This is the only example of large-scale reparations for genocide in history. Psychologically and politically, this mattered: recognition does not absolve guilt, but creates a basis for normalization.

“Transitional justice”— mechanisms that allow societies to emerge from periods of mass atrocities. Nuremberg— criminal prosecution. The Truth and Reconciliation Commission in South Africa (after apartheid)— recognition through testimony. Lustration in Poland and Czechia— restriction of access of ...

Traditional ethics are interpersonal and synchronous: I am responsible to people who live now and whom I know. The ecological crisis calls these boundaries into question: our actions today determine the quality of life for people who have not yet been born. Can we have obligations to those who do...

Edmund Burke (an eighteenth-century conservative) formulated: society is a partnership of the dead, the living, and the yet unborn. The modern state is only a temporary steward of resources belonging to the chain of generations. This is a conservative argument in favor of sustainability.

Hans Jonas ("The Principle of Responsibility", 1979): a new categorical imperative for the technological age — "Act so that the effects of your actions are compatible with the continued genuine human life on Earth." This is responsibility without reciprocity: future generations cannot demand from...

Peter Singer ("Animal Liberation", 1975) expanded the circle of moral concern: the capacity to suffer is a sufficient criterion for moral consideration. If animals suffer, this is morally significant, regardless of their rational status. This is utilitarianism applied to nonhuman beings.

Medicine has always been an ethical space — but the twentieth century created fundamentally new dilemmas. Life-support machines made death controllable. Transplantology — the body as a source of resources. Genetics — traits as mutable. Reproductive technologies — birth as planned and constructed.

Bioethics as a discipline took shape in the 1960–70s: Van Potter introduced the term, Georgetown University developed the principles. Four principles (Beauchamp and Childress): autonomy (the patient’s right to make decisions concerning their own body), non-maleficence (primum non nocere), benefic...

Euthanasia — the deliberate ending of a person’s life to ease suffering — is one of the most ethically charged topics. Distinguish: passive (cessation of treatment), active (administration of a lethal substance), voluntary (at the patient’s request), involuntary (without consent).

Arguments for: autonomy — a fundamental principle, the person has the right to die with dignity; suffering that cannot be relieved itself violates the principle of non-maleficence. Arguments against: risk of the “slippery slope” (weakening protection for the vulnerable); pressure on the elderly a...

“Data is the new oil” is a popular metaphor that deserves critical scrutiny. Oil extraction destroys ecosystems. Data “extraction” breaches privacy. Oil is a scarce resource. Data is excessive (there is ever more of it). Oil belongs to the one on whose land it is extracted. Data about you—who own...

Data about people is a special type of resource. It carries consequences for the lives of those about whom it is collected. Data on health, finances, behavior, movements, preferences—this is power over a person. Whoever controls this data—controls the possibilities.

The European General Data Protection Regulation (GDPR, 2018) is the most comprehensive legal response to questions of data ethics. Key rights: access to one’s own data, correction, deletion (“the right to be forgotten”), portability, objection to automated decisions.

GDPR is based on principles: legality (there is a legal ground for processing), purpose (data is collected for a specific, declared purpose), minimization (only necessary data), accuracy, storage (no longer than necessary), security, accountability.

Power is an integral part of leadership. But power changes psychology. Research by Dacher Keltner (“The Power Paradox”): people with power become less empathetic, less able to take another’s point of view, more impulsive. This is the paradox: the very qualities that help one gain power (empathy, ...

Lord Acton: “Power corrupts, and absolute power corrupts absolutely.” This is not a metaphor—it is a psychological reality, experimentally confirmed. How can leaders protect themselves?

Practical mechanisms: structural accountability (boards of directors, independent directors, 360-degree reviews), a culture that encourages people to “speak truth to power,” personal discipline of reflection (keeping a journal, coaching, meditation).

Aristotelian definition of courage: the golden mean between cowardice and recklessness when faced with frightening things. Moral courage is the same applied to moral challenges: resisting group pressure, admitting a mistake, exposing a violation, making the right but unpopular decision.

AI systems make decisions that affect people: which loan to approve, which candidate to call, which sentence to propose, which advertisement to show, whom to connect with in a chat. Previously, all these decisions were made by humans—with understandable motives, responsibility, and the possibilit...

The European "Ethics of AI" (EU AI Ethics Guidelines, 2019) proposed seven principles: human oversight, reliability, privacy, transparency, non-discrimination, societal well-being, accountability. This is not law—it is a framework for evaluating AI systems.

The problem of accountability: if an algorithm incorrectly assesses credit risk, who bears responsibility? The programmer? The company? The user of the algorithm? "Multilevel responsibility"—in distributed systems, it is difficult to assign.

Systems like GPT-4 are narrow AI: capable of doing specific things well. Hypothetical AGI (artificial general intelligence) is a system comparable to a human in the breadth of its capabilities. Superintelligence—surpassing human abilities.

Traditional ethics are concentric: first—the family, then—the community, then—the nation. Globalization and communications make suffering on the other side of the planet possible—and visible. This raises an ethical question: do I have obligations to people I will never meet?

Peter Singer (“Famine, Affluence, and Morality,” 1972): the classic argument. If you see a drowning child—most people would agree that you should save them, even if it ruins an expensive suit. But a child dying from a preventable disease in Bangladesh is morally analogous. Distance does not creat...

It is a demanding conclusion. Singer does not deny this. He proposes a pragmatic minimum: 10% of income (“Giving What We Can”—the effective altruism movement).

Climate change is a global issue with uneven distribution. Rich countries have created most of the problem historically (the USA, Europe—the leaders in cumulative CO₂ emissions). Poor countries, which have hardly contributed to the problem, bear the greatest consequences (Bangladesh—floods, the S...

Medicine has always strived to return the sick to "normal". New technologies open up the possibility to go beyond the norm: cognitive enhancers (nootropics, neurointerfaces), genetic enhancement, radical life extension, merging with AI. Transhumanism is a philosophical movement that asserts: tech...

Nick Bostrom, Ray Kurzweil—the main voices of transhumanism. Kurzweil predicts the "Singularity"—a moment around 2045 when AI will surpass humans, and the merger of humans and machines will change the very nature of existence.

Michael Sandel ("The Case Against Perfection"): technological enhancements undermine the values of acceptance, humility, and solidarity. If we can engineer a "better" child—everything about them becomes our choice and our responsibility. This is unbearably heavy.

Francis Fukuyama ("Our Posthuman Future"): "human dignity" is founded on human nature. If we radically alter this nature, we lose the basis for rights and morality.