Chapter 12 Key Takeaways: Applied Ethics

The Central Argument

Applied ethics is not a simplified version of theoretical ethics. It is where theoretical ethics does its most important work. The frameworks developed in Chapters 4–11 — consequentialism, Kantian ethics, virtue ethics, care ethics, Rawlsian justice — are not academic exercises. They are tools for thinking about exactly the kinds of questions that arise every week in technology, medicine, business, and the natural world.

The chapter demonstrates a core methodological principle: when frameworks converge, you have stronger grounds for action; when they diverge, the divergence tells you precisely what is at stake. Divergence is not a failure of ethics; it is ethics doing its best work by clarifying the real choice.

Technology Ethics

Algorithmic bias illustrates how systems without explicit discriminatory intent can produce discriminatory outcomes by learning from historical data that encoded historical discrimination. The problem is not primarily a technical one; it is an ethical one about responsibility, design choices, and whose interests are centered.

  • Consequentialism condemns algorithmic discrimination when it produces disparate impact with long-run, self-reinforcing harms that compound over time — not just one-time costs.
  • Kantian ethics condemns it as treating persons as data points correlated with group membership rather than as individuals with dignity. Group membership is not a morally relevant fact about a person.
  • Virtue ethics asks about institutional character: what kind of organization fails to audit its systems for bias, and what kind of organization addresses it?
  • Care ethics reveals that the people most harmed by algorithmic bias are precisely those least likely to have relationships with the institutions making decisions about them.
  • Rawlsian justice demands that the rules governing AI systems be ones you would accept without knowing which side of the algorithm you would be on.

Responsibility for AI harms is distributed: programmers, companies, regulators, and users each have genuine obligations. Distributed responsibility does not mean diluted responsibility.

Bioethics

The four principles of biomedical ethics — autonomy, beneficence, non-maleficence, and justice — map onto the theoretical frameworks and organize clinical ethical reasoning.

  • Autonomy reflects Kantian respect for persons: patients are not passive recipients of medical expertise; they are persons with their own values and goals.
  • Beneficence and non-maleficence reflect consequentialist concern for welfare.
  • Justice reflects Rawlsian distributive concern.
  • All four require virtue — practical wisdom — to apply in specific situations where they conflict.

Key bioethical cases from this chapter:

  • End-of-life decisions: Patient autonomy is central; the Kantian argument is that we must try to reconstruct what this particular person would have wanted. Virtue ethics requires honesty about prognosis and humanity in the dying process. Justice requires attending to the allocation of scarce ICU resources.

  • Genetic engineering: Consequentialist arguments for preventing suffering are complicated by precautionary concerns about irreversible, heritable effects and by distributional questions about who will have access. The Kantian objection to "designing" children centers on treating future persons as products rather than ends. The "playing God" argument, translated into secular terms, is a substantive argument about the dangers of wielding power that exceeds our wisdom.

  • Resource allocation in medicine: No formula resolves tragic allocation decisions, but careful reasoning supports transparent, consistently-applied criteria, genuine attention to each patient as a person, and appropriate humility.

Business Ethics

The fundamental debate about corporate purpose divides between:

  • Shareholder theory (Friedman): corporations owe obligations only to shareholders; social goods should be achieved through markets and law, not managerial discretion.
  • Stakeholder theory (Freeman): corporations exist within a web of relationships with employees, customers, communities, and the natural environment, and have genuine obligations to all of them.

These are not just competing policy preferences; they rest on different theories of what a corporation is.

Pharmaceutical pricing illustrates the tension between legitimate claims (return on investment, incentives for innovation) and ethical demands (access for patients who need treatment). The Kantian analysis is especially pointed: pricing according to "value to patients" exploits the vulnerability of people who need treatment rather than serving their needs.

Complicity in organizational wrongdoing requires a threshold analysis: moral agency cannot be fully outsourced to an employer, but the appropriate threshold for resistance rises with the seriousness of harm, the directness of personal participation, and the availability of alternatives. Practical wisdom means finding the appropriate middle ground — not refusing every instruction you disagree with, and not deferring to authority in all but extreme cases.

Environmental Ethics

The expanding circle is the historical process by which moral consideration has widened. Environmental ethics asks whether it should expand beyond all humans to animals, species, ecosystems, and the natural world.

  • Singer's utilitarian argument: if suffering is bad regardless of who experiences it, animal suffering must count. Speciesism — arbitrary privileging of human interests over equivalent animal interests — is a moral error analogous to racism or sexism.

  • Legal and moral standing for nature: Christopher Stone's argument that natural entities should have standing — like corporations, which are not persons but have legal rights — reflects a philosophical position that nature has value independent of its utility to humans.

  • The non-identity problem (Parfit): people harmed by climate inaction are different people from those who would have existed had we acted. This complicates harm-based arguments but does not eliminate obligations. Impersonal consequentialist arguments and stewardship arguments both ground environmental obligations without requiring that we identify specific harmed individuals.

  • Indigenous environmental frameworks offer relational models of human-nature obligation that predate Western environmental ethics and differ fundamentally from it — not by providing different answers to the same questions, but by framing the questions differently. Land as a relational partner, not a resource.

  • Rawlsian justice across generations: from behind a veil of ignorance about which generation we'd be born into, we would choose significant caution about irreversible harms to the natural world and significant weight given to the interests of people who will bear the costs of today's decisions without having any voice in them.

Methodological Takeaways

  1. Multiple frameworks are better than one. Each framework illuminates something the others miss. The full moral imagination requires all of them.

  2. Convergence strengthens conclusions. When consequentialism, Kantian ethics, and justice frameworks all condemn the same practice, the convergence is evidence.

  3. Divergence clarifies the real question. When frameworks disagree, the point of divergence identifies the underlying value conflict. This is not a failure; it is clarity.

  4. The personal is philosophical. The frameworks are not just for analyzing corporate behavior or government policy. They apply to the ethical questions in your own life: at work, in relationships, as a citizen.

  5. Responsibility is distributed but real. When AI systems, corporate structures, and social institutions produce harm, responsibility is spread across a network of actors. Distributed responsibility does not mean no one is responsible; it means everyone in the network has genuine obligations.

Questions to Carry Forward

  • Do AI systems have the same ethical status as the humans who design and deploy them, or does responsibility always trace back to human decision-makers?
  • What does it mean to take seriously the interests of future generations when those generations have no voice in current decisions?
  • Is there a principled limit to the expanding circle — a point where the extension of moral consideration loses coherence?
  • When is it enough to follow the rules, and when does acting ethically require going beyond them?