Primary Goods
Primary goods are Rawls's answer to a problem that plagued earlier contractarian theories: how can parties in the original position rank alternative institutional arrangements without knowing their particular conceptions of the good? Rawls's solution was to identify a class of goods that any rational person would want more of, regardless of what she happens to want out of life. Whatever your life plan, you will be better able to pursue it with more income than less, more opportunities than fewer, more rights than fewer. Primary goods are the currency in terms of which distributive claims are evaluated — not because they are the only things that matter, but because they are the things that matter to everyone independently of what else she happens to value. The list Rawls settled on includes basic rights and liberties, freedom of movement and occupation, powers and prerogatives of office, income and wealth, and — most importantly — the social bases of self-respect.
The Abstraction Hides the Substrate — Contrarian ^ Opus
There is a parallel reading that begins not with what rational persons want but with what the production of primary goods actually requires. Rawls's framework treats income, wealth, and opportunities as distributable resources—things that exist to be allocated. But these goods do not fall from the sky. They are produced through specific material processes: extraction, manufacturing, logistics, computation. The difference principle asks how to distribute the social product, but it systematically obscures the question of what substrate conditions make that product possible in the first place.
The AI transition makes this obscurity dangerous. When we ask how to distribute the social bases of self-respect in an AI economy, we are not asking an abstract institutional question. We are asking what happens when the material substrate shifts from human labor to computational infrastructure—when meaning-making moves from craft and expertise to prompt engineering and model supervision. The primary goods framework treats this as a distribution problem: some people will have less income, fewer opportunities, diminished recognition. But the deeper change is substrate: the social bases of self-respect under artisanal production were grounded in accumulated mastery that took years to develop and could be seen in the object itself. The social bases of self-respect under AI mediation are grounded in... what, exactly? Access to compute? Skill at model interrogation? The recognition that comes from effective prompt craft? These are not just different quantities of the same good. They are different kinds of substrate entirely, and no amount of careful distribution can recover what the substrate shift itself destroys.
In the AI Story
The function of primary goods in Rawls's theory is to make interpersonal comparisons possible without requiring agreement on controversial questions about the good life. Under the difference principle, the least advantaged are those with the smallest share of primary goods — a specification that does not require resolving whether their life plans are the best plans or whether they would be happier with different aspirations. The framework is deliberately neutral among comprehensive conceptions of the good, while still providing a determinate metric for evaluating distributions.
The primacy of the social bases of self-respect is particularly important for the AI transition. Self-respect, in Rawls's framework, is the most important primary good — the good without which all other goods lose their value. The social bases of self-respect are the institutional conditions that support a person's sense that her life plan is worth pursuing and that she possesses the capabilities to pursue it. These conditions include meaningful work, social recognition, and the sense that one's skills and contributions are valued by others. The AI transition threatens the social bases of self-respect for a significant portion of the population — not necessarily by making them unemployed but by eroding the social recognition of their expertise, their craft, their accumulated mastery.
Amartya Sen argued that primary goods were the wrong metric — that what matters is not what resources people have but what they are able to do with those resources, their capabilities. Two people with identical bundles of primary goods can face radically different prospects depending on their conversion capacities. A disabled person and an able-bodied person with the same income do not have the same opportunities. The Sen critique has reshaped development economics and influenced the design of human development metrics, but it has not displaced primary goods from Rawlsian theory; many philosophers treat the two approaches as complementary rather than competing, using primary goods as the metric at the level of institutional design and capabilities as the metric at the level of individual evaluation.
Origin
Rawls introduced primary goods in A Theory of Justice (§§15, 60–67) and refined the specification in his 1982 essay "Social Unity and Primary Goods" in response to Sen's and other critics' arguments. The list of primary goods evolved across his career, with the social bases of self-respect taking on increasing prominence as Rawls came to see it as the most important good and the one most directly threatened by institutional failures.
Key Ideas
Goods every rational person wants. Primary goods are defined by their relationship to rational agency as such, not to any particular conception of the good life.
Currency of distributive justice. The difference principle evaluates distributions in terms of primary goods, enabling interpersonal comparisons without requiring agreement on controversial questions.
Priority of self-respect. The social bases of self-respect are the most important primary good because without them all other goods lose their value to their possessor.
Not fungible. Primary goods cannot be substituted freely for one another; a person cannot be compensated for the loss of liberty by additional income, or for the loss of self-respect by additional opportunities.
Metric challenge from capabilities approach. Sen and Nussbaum argued that what matters is what people can do, not what resources they have; the debate between primary goods and capabilities remains central to contemporary theories of justice.
Debates & Critiques
The primary goods framework has been contested from multiple directions. Sen's capabilities critique argued that resource-based metrics systematically mismeasure advantage because conversion capacities vary widely. Libertarians argued that the list itself embeds substantive views about what makes a life good. Communitarians argued that the framework abstracts from the particular communal contexts in which goods actually have meaning for persons. Each critique has shaped the subsequent development of the theory. Primary goods remain central to Rawlsian analysis because no alternative metric has demonstrated equivalent usefulness at the level of institutional design.
Appears in the Orange Pill Cycle
Substrate and Distribution Both Constitute Justice — Arbitrator ^ Opus
The right weighting depends on which justice question we are answering. For questions of *institutional design*—how to structure tax systems, educational access, labor protections—the primary goods framework is close to 100% correct. These are distribution problems, and the metric Rawls provides (goods every rational person wants, evaluated without requiring agreement on the good life) remains the most useful tool we have. The contrarian reading offers no better answer to how we should allocate income or opportunities across persons with different life plans.
But for questions of *what is being distributed*—what income and opportunities actually consist of after the AI transition—the substrate reading dominates at perhaps 70%. The social bases of self-respect under craft production included the felt experience of accumulated mastery, the social recognition of skill visible in objects, the dense web of interdependencies that made expertise meaningful to others. These were not just quantities you could have more or less of; they were qualities that emerged from specific material and social conditions. When the substrate shifts, you cannot simply redistribute the new goods to recover the old social bases. A person with high income from supervising AI systems and a person with high income from pottery production do not have the same relationship to their work's meaning, even if the primary goods metric scores them identically.
The synthetic frame the concept itself benefits from: primary goods should be understood as a *two-layer* framework. At the distribution layer, the Rawlsian metric remains correct—we evaluate institutional arrangements by how they allocate income, opportunities, rights. But there is a prior substrate layer: *what forms can these goods take*, and what does the substrate shift itself enable or foreclose? Justice requires both getting the distribution right and attending to what the substrate makes possible.
Further reading
- John Rawls, A Theory of Justice, §§15, 60–67
- John Rawls, "Social Unity and Primary Goods," in Utilitarianism and Beyond, ed. Sen and Williams (Cambridge, 1982)
- Amartya Sen, "Equality of What?" in Tanner Lectures (1980)
- Norman Daniels, Just Health (Cambridge, 2008)