Universals

Whatever it is that many different things share when the same word is true of all of them.

Theme: Being and reality
Tradition: Greek
Term fixed: c. 380 BC

In plain terms

Two apples are different objects. Nothing about one of them is literally part of the other. Yet the same word is true of both, and not by accident - it goes on being true of apples nobody has seen yet.

So what is shared? One answer is that redness or applehood is a real thing existing over and above the individual apples. Another is that there is nothing shared at all: there are just apples, resembling each other, and a word we find convenient.

The reason this is not a quibble is that our classifications work. They support predictions about cases we have never met. Either the world comes divided in a way our words track, or it does not and the success needs some other explanation.

As a philosopher would put it

Realism holds that universals exist. Plato places them apart from particulars; Aristotle places them in the particulars that instantiate them, existing only as instantiated. Nominalism denies them: only individuals exist, and generality belongs to language. Conceptualism puts it in the mind, as an abstraction we form.

The medieval debate is the sharpest. Roscelin's position - that a universal is nothing but a flatus vocis, a puff of the voice - provoked William of Champeaux's realism, and Abelard's reply to both is the pivot: a universal is a word predicable of many, and what things share is not a thing but a status, a way of being, which is not itself an object. Ockham completes the nominalist line with a theory of natural signs.

Scotus offers a middle: a common nature that is genuinely in things and neither universal nor individual until contracted by a haecceitas, a thisness. In India the parallel debate runs between Nyāya realism about sāmānya and the Buddhist apoha theory, which analyses a general term as an exclusion of what it is not.

Where you meet it

Russian has no single word for blue. Light blue is goluboy and dark blue is siniy, and they are not shades of one word in the way that crimson and scarlet are shades of red in English - they are separate basic terms, learned separately.

In 2007 a Boston University group ran a discrimination task with Russian and English speakers, asking them to identify which of two squares matched a third. Russian speakers were reliably faster when the two squares fell on opposite sides of the goluboy-siniy line, and lost that advantage under a verbal load. English speakers showed no effect at that boundary at all.

So the boundary is doing real perceptual work, and it is not in the light. But nor is it invented freely: no language draws a boundary between two randomly chosen wavelengths, and the ones that recur across unrelated languages cluster. Whatever universals are, this is the shape of the evidence about them - neither simply out there nor simply made up.

What is actually disputed

Whether anything general exists.

The realist argument is that similarity has to be grounded in something, and that predication and successful induction are unintelligible if there is nothing the world shares. The nominalist reply is that primitive resemblance between particulars does all the work at a fraction of the cost, and that positing an entity present entirely in each of many places at once is a stranger commitment than the one it was meant to avoid.

The argument is genuinely open, and it recurs under other names: whether mathematical objects exist, whether natural kinds are natural, whether social categories are real or are the boundaries an interest drew. It is the oldest live question on this list.

Commonly confused with

universal in the sense of worldwide Nothing to do with each other. A universal here can have two instances.
Concept 1781 A concept is in a mind. Whether a universal is too is precisely what conceptualists claim and realists deny.

Defined against

Particulars The pair the whole dispute is about, and neither side denies that particulars exist

Philosophers

Plato Universals as separate and fully real
Aristotle Real, but only in the particulars
Roscellinus of Compiègne A universal is a puff of the voice
William of Champeaux The realism Abelard was reacting against
Peter Abelard Neither a thing nor a mere word, but a status
William of Ockham Only individuals exist; generality is in signs

Works

Republic Where the Forms carry the weight
Metaphysics Universals as immanent rather than separate
Sic et Non Abelard, and the method that made the dispute sharp
Summa Logicae Ockham's completed nominalism
Ordinatio The common nature, and thisness
Nyaya Sutras The Indian realist account of sāmānya

Schools

Old Academy Platonic realism
Peripatetic school (Aristotelianism) Immanent realism
Nominalism The denial
Ockhamism Its most worked-out form
Scotism The middle position
Nyaya Realism about universals, argued independently

Related concepts

Particulars c. 330 BC · Form c. 380 BC · Essence c. 330 BC · Concept 1781 · Category c. 330 BC

Plato poses it; Porphyry's three questions in a short introduction to Aristotle, left deliberately unanswered, are what turn it into the central problem of medieval philosophy.