Charles Sanders Peirce's Trichotomies of Signs
Charles Sanders Peirce (1839–1914) held that a sign is never a simple pairing of a mark with a meaning. Every sign involves three things: the representamen, the sign-vehicle itself; the object it stands for; and the interpretant, the understanding it produces in a mind, which is itself a further sign, so that interpretation never quite comes to rest. Beneath this triad lie his three categories: Firstness, quality or possibility; Secondness, brute fact, one thing pushing against another; and Thirdness, law, habit and mediation. In his 1903 essay “Nomenclature and Divisions of Triadic Relations, as Far as They Are Determined”, Peirce ran each part of the sign through each category, and the result is the grid on the board above. Across the top row, a sign considered in itself is a qualisign, a sinsign or a legisign. Across the middle, it relates to its object as an icon, by resemblance, an index, by real connection, or a symbol, by convention. Across the bottom, it presents itself to its interpreter as a rheme, a mere possibility, a dicent, a claim that can be true or false, or an argument, a conclusion drawn from premises. Every sign takes one cell from each row, under one constraint: no row may be of a higher category than the row above it. A mere quality can only resemble, and what only resembles can assert nothing. That rule reduces twenty-seven possible paths to exactly ten classes, and it is why the board refuses any path that climbs back up. A valid path can only hold its column or drift towards Firstness, so it is fully described by the two rows at which it steps down; counting those choices gives (n+1)(n+2)/2 classes for n rows, which is how Peirce's later ten-row scheme arrived at sixty-six. The grid matters to science because science is, to a large degree, the disciplined reading of indices—fumes, flame colours, a diffraction photograph—through symbols, by way of Peirce's three forms of inference: abduction to propose a hypothesis, deduction to draw out its consequences, and induction to test them.
What Is the Meaning of Wet Paint?
No sign has taught me more about the grid than a painter's notice reading “Wet Paint”. Stacked in the back of the van, it is meaningful but claims nothing: it says “___ has wet paint” with the blank unfilled, which is exactly Peirce's rheme, and close to what Aristotle said of a verb spoken on its own. Fixed to a bench, it becomes a dicent. Nothing is added to the words; the placement supplies the index, and with it an unspoken “is”, much as an Egyptian adverbial sentence asserts “he (is) in the house” with no copula at all. Peirce's nearest parallel is the street cry, a conventional type whose every utterance is anchored to the seller who cries it. Yet the same sign on the same bench can carry different news. By a lone bench, the news is that the paint is still wet. At the end of a row that the painters are working along, the news is which bench: the rheme, in the linguists' sense of theme and rheme, has moved onto the location, and in speech we would reach for a cleft—it's this bench that's wet. On a single bench touched up in a row of dry ones, the news is that any painting has happened at all. Leave the sign up for a week after the paint has dried and it has not lost its meaning; it has become false, because its present tense keeps renewing itself while its content does not. And a sign propped against a bench that was never painted, a feint, shows that a lie needs an assertion: the rheme in the van cannot deceive anyone, but the same words, given an index, can. Even the effort of putting a sign up works like stress in speech, marking the whole message as worth attention, though only context tells the reader where inside it the weight falls. Peirce's rheme and the linguists' rheme descend from the same Greek rhēma, “what is said”, yet they measure different things, and these benches show how far apart they can drift.
Tracking the Meaning through a Text
The wheel above is one I drew when I first began reading Latin, as a way of keeping my footing inside a sentence. Its outer ring carries the drift of an argument, forward to build a case or back to concede; within it sit the subordinate clauses, temporal, causal, concessive and of result; then a ring I labelled ontology, for claims of cause and consequence in the world; and at the centre, the future remote and the hypothetical. Read against Peirce, the wheel moves inward from Thirdness to Firstness, with a mixed ring between, since a temporal clause places a claim much as an index does, while a Latin causal clause usually gives a reason rather than a push. It diverges too: Peirce assigned the conditional “would-be” to Thirdness, so some of my remote conditionals belong at the rim rather than the centre. Across the wheel runs a strip I called the problem, and its central station is metaphor. Many Latin verbs split between a literal sense that explains the world and a figurative one that carries the argument: venio, I come, and I come to my next point; fero, I carry, and also ferunt, they say, and I endure; puto, I prune or clear, then I settle an account, then I think. The upward arrows trace this lift from the ontological ring to the argumentative one, the pattern Eve Sweetser maps in From Etymology to Pragmatics (1990), where the same words work in a content domain of real events and an epistemic domain of reasoning. The arrows on the right record the other half of the work: when a first reading will not fit, they are the paths along which a second is sought, which is Peirce's abduction drawn as a diagram. The hardest case I met was linquo. To leave a place empties it; to leave something behind fills it (reliquus, remaining). In Book 1 of De Rerum Natura, Lucretius reports the view that fish can swim through a world without void because they leave places behind them, post loca pisces linquant, into which the yielding water flows. Here one verb does both jobs at once: the fish empties the space ahead and fills the space behind in a single motion. Lucretius's refutation turns on showing that this “at once” is impossible. The fish cannot advance until the water gives way, and the water has nowhere to go until the fish advances; without void, nothing moves. I read this passage as the keystone of the poem, a metaphysical argument about the incompressibility of substance and of time set among empirical ones, and it keeps my linquo problem open, because the grammar holds together in one word a double event that the physics says cannot happen simultaneously. It is also striking that the water is said to yield, cedere, the root of the concession that sits on my outermost ring.
Weaving Together through Context
This investigation has been a sequence of splits. Peirce divides the sign into three relations and each relation into three categories; my wheel divides a sentence into four rings; the words themselves split, venio into a journey and a step in reasoning, linquo into emptying and filling, and even “rheme” divides between Peirce's unasserted predicate and the linguists' news. What makes any of it readable is the thing that turned the painter's sign from a rheme into a dicent: the index. Context works as an index, and the word itself is one more of the metaphors this essay has been tracing: Latin contextus, from contexere, to weave together. It ties a sign not only to its object but to one branch of its meaning. Sometimes it lies in the world, in the bench, the row, the dried paint, the tube of fuming acid; sometimes it lies in the text, in word order, in what has already been said, in whether a clause belongs to an argument about atoms or a story about a fish. A symbol offers a range of possible meanings, and indices select among them. The selection is never certain. Every reading is a hypothesis, formed by abduction when a first reading surprises us and tested by working out what would follow and checking it against the rest of the passage, which is why the board above accepts two answers for some signs and why my wheel has arrows rather than fixed destinations. Peirce held that no belief is beyond revision, and that the first rule of reason is not to block the way of inquiry; reading, whether in a laboratory or in Lucretius, is that rule applied one sign at a time. In the end we make sense the way the sign on the bench does: by being placed.