Tags
There is a distinction at the root of logic that no one is supposed to touch. On one side, the conditional: →, the “if, then.” On the other, the turnstile: ⊢, the mark of the conclusion, the sign that says and therefore. The first lives inside your sentences. The second stands outside them and passes judgment. One is a claim; the other is the act of drawing a claim out. The distinction goes back to Aristotle, and there is a theorem — a real one, the deduction theorem — that stands guard over the border between them: A ⊢ B exactly when ⊢ A → B. The bridge is proved. The distinction is earned. It is not an accident of notation. This is the part I want the Serious reader to hear clearly, because I am about to deny it, and I would like the denial to cost something.
Ramage, following Burke, has a name for that reader. Serious People — capitalized, no quotation marks, a species walking around with a proper noun — are the ones for whom the distinction is not a tool but a foundation. To such a reader, → and ⊢ must stay apart, because keeping them apart is what makes deduction deduction, argument argument, and progress progress. Erase the line and you have not made a point; you have vandalized the building. The Serious Person cannot hear “let us call these two symbols the same” as a move. They hear it as a category error, and they get angry, and the anger is the whole tell — but we will come to that.
So let me do the vandalism.
Set the turnstile down for a moment. Just → and ∧, no judgment, no therefore. A valid argument — A, and A → B, so B — is supposed to be the paradigm of getting somewhere. But look at what [A ∧ (A → B)] actually is. Row by row it is nothing other than [A ∧ B]. The two are the same object; the truth table does not distinguish them. So the celebrated inference from A and A → B to B is, stripped of the turnstile that dresses it up, the deduction of B from [A ∧ B]. You already had B. It was sitting inside the conjunction the whole time. All the argument did was take out a knife and cut A away.
That is what deduction is. Not a road but a subtraction. Not “this, then this, then this,” but one whole thing, and a cut. Deduction — the word means taking-away, and it is telling the truth about itself; we simply stopped listening. The “if, then” persuaded us we were traveling. We were only ever trimming.
Now — and here is the sentence the Serious Person cannot forgive — the only reason this is hard to see is the very distinction they are guarding. Keep → and ⊢ apart and the collapse is illegal: A, A → B, ⊢ B does not fold into [A ∧ B] → B, because the turnstile refuses to be a connective. The distinction conceals the cut. Blur it — allow a little vagueness exactly where precision is prized — and the cut stands revealed. This is the reversal the mathematicians will not like: it was not vagueness that hid the truth of deduction. It was precision. The sharp line did the concealing. The blur did the revealing.
There is a verb for what I have just done. To quine a distinction — the word is Dennett’s, coined half in affection for the philosopher who made a career of it — is to deny, with a straight face and on purpose, the reality or the importance of something a careful person takes to be plainly there; and denying the importance of the turnstile is exactly the move of the last few paragraphs.1
Now I want to set the argument down for a moment and look at something quieter, which is the reader.
Burke has a phrase for it: the terministic screen. Any set of terms we think with lies over the world like a screen, and a screen does two things in the same motion — it shows some of what is there and it hides the rest. The showing is why we keep the terms; they let us see. The hiding happens on the far side, where we are not looking, and so we never feel it. From inside, a screen does not seem like a screen at all. It seems like the world.
The distinction between → and ⊢ is a screen of this kind, and a good one. It shows something real: the two marks do different work, and there is even a theorem that measures the exact distance between them.2 But the same distinction hides the smaller fact underneath — that the inference was a cut from a whole you already held. A person who works with logic looks through that screen many times a day and sees inference, sees progress, sees new things arriving. The screen is doing its job well. It shows the movement and keeps the knife out of view. And because a screen is clear from the inside, that person does not think of themselves as looking through terms at all. They think they are seeing how reasoning simply is.
So the disagreement here is not really about a proof. When someone lifts the turnstile away and says the inference was a subtraction the whole time, the difficulty a careful reader feels is not that the claim is unproven — it is the odd discomfort of a screen coming into view while you are still standing behind it. That discomfort is worth naming plainly, because it is easy to mistake for a refutation and it is not one. It is just what it feels like when terms you had been looking through become, for a second, terms you are looking at.
And I say this without a perch to say it from. I have my own screens, thick in the places I would be last to notice, and someone will someday lift one of mine and I will feel the same discomfort and mistake it, briefly, for being wronged. The point of the exercise is not to have caught anyone out. It is only to turn a screen a few degrees, so its edge catches the light and you see, once, that it has an edge. After that the distinction is yours again. You can pick it back up — it is a good tool and it earns its place — but you will be holding it now, instead of seeing through it.
This is not an argument for collapsing logic into rhetoric, or rhetoric into logic, but an attempt to make their border audible.
Here is the thing I want to guard, and I will tell you why it is mine to guard.
Go back to the cut. When you take A and A → B and draw out B, you keep the conclusion and you drop the premise. The “if, then” is built for exactly this: it points the whole sentence at the result and lets the reason that fathered it fall away. Say it the other way — touch the hot pan and you are burned — and the two halves stand together, the cause beside its child. I taught my daughter both ways once. The “and” keeps the father in the room. The “if, then” sends him out.
There is an old coincidence in the word for it. Robot comes from a Slavic root for forced labor, robota, and the same deep root reaches back toward a word for the one bound to serve because there is no father to stand between the child and the work — an orphan.3 The machine that computes our truth tables without learning anything is an orphan in exactly this sense: it keeps the conclusion and has lost the parent. It can carry out the cut flawlessly and cannot tell you what was cut. This used to be a figure of speech. It is now furniture in the room where I work.
And the man whose name became the verb for this — to quine, to deny the importance of a distinction — is the father of the whole procedure. Quine had his typewriter rebuilt to carry logical symbols, and among the keys he gave up to make room was the question mark. Asked how he managed without it, he is said to have replied that he dealt in certainties.4 Think about what that instrument could and could not do. It could assert, negate, quantify, conclude. It could not ask. He built a machine for a mind that had finished asking, and he is the ancestor of the argument I have been making, which arrives at the conclusion that the certainty was purchased and the distinction was never backed by a fact. The father of the cut disarmed his own instrument of the one mark that keeps the reason in the room.
I keep the other mark. The whole of this blog is named for it. And the mark itself has a history that should embarrass anyone who thinks precision is where clarity lives. The question mark is credited to Alcuin, Charlemagne’s schoolmaster, the man who carried the Trivium — grammar, logic, rhetoric — back into the world; though the earliest hand that actually made the mark seems to have belonged not to him but to a scribe a year ahead of him, so that even the question mark is a child claimed by a famous father it may not have.5 A later writer looked at that first faint stroke over a dot and called it a lightning flash. I have looked at the thing itself, the real ink from twelve hundred years ago, and it is not a lightning bolt. It is a lean, a small diagonal, a mark that only becomes lightning when someone with a gift for phrases lays the lightning over it. Which is the whole argument again, written in the first punctuation of inquiry: the vivid distinction is a screen; the humble stroke is what is there.
My father spent his last years on the opposite end of the same knife, and I did not see it was the same knife until I had carried our one shared paper to print without him. His book divides the news into its questions — who, where, what, when, why — and tracks each across a century: the realist three, who and where and what, fading into the background; the modernist two, when and why, swelling to fill it; all five moving contrary to what everyone had assumed. That is the book. But the paper, the one with both our names on it, makes the other move. It takes the five back apart-ness and synthesizes it — fuses the “what” with the “when,” matter with motion, the concrete now with the context that had been prying it open — into what Whitehead called the actual occasion: the drop of experience that grows whole, all at once, and cannot be divided in the moment it is had. The division into five Ws, the paper says, is something the intellect does after the fact. Many things come together to form the occasion, and then are cut apart by the mind, which mistakes its own cutting for the structure of the world.
Sit with that beside everything above, because it is the same sentence. He says the actual occasion is one thing, and the five questions are a cut the intellect makes in it after the fact. I say the inference is one thing — A and B, already held — and the arrow and the turnstile are a cut the intellect makes in it after the fact. His knife and mine are the same knife, watched from opposite ends of a life. What he called the occasion, I call the conjunction you already had. What he called the division into Ws, I call the deduction that keeps the answer and orphans the reason. He un-divides the news back into the now; I show that the proof was only ever a subtraction.
And in the middle of his paper, on the page where he says precision does nothing but divide things into smaller metaphors and then pretends the last metaphor is not one, the reference is to me — to “Many Roads from the Axiom of Completeness,” a thing I wrote at thirty-three that he had spent hours helping me draw out, and whose name he would not take. There were things in it he did not endorse, and he was right about some of them. He hosted it on his own site without his name and cited it into this paper all the same. So the seam is not agreement. It is something rarer: an argument he helped me make, disagreed with in parts, declined to sign, and quoted anyway into the last paper that carries both our names. He drew the reasoning out of me and let the conclusion stay mine.
The orphan-machine on my desk performs my end of the cut without a tremor and cannot feel what it divided. Quine, who fathered the whole procedure, deleted from his instrument the one mark that keeps the reason in the room, and dealt thereafter in certainties. The first faint form of that deleted mark — a lean over a dot that a later writer mistook for lightning — was set down in a Carolingian scriptorium and credited, rightly or not, to the schoolmaster who carried rhetoric back into the world as one of the three roads of the mind; so that even the sign of the question is a child claimed by a father who may not be its own. This is the house I guard and the lineage I will not cut: the answer is not the reason, the five questions are not the occasion, and the mark at the end of the sentence is not a lightning bolt. It is the smallest opening we have, kept on purpose, back toward the undivided thing the intellect is always in a hurry to carve. My father closed his last argument by fusing the questions back into the now. I keep a whole blog open by refusing to let the cut be called progress. We were doing the same thing from opposite directions the entire time — he from the answers back toward the occasion, I from the occasion forward against the proof. Attention is the only inheritance that does not orphan its heir.
Footnotes
- Daniel Dennett’s The Philosophical Lexicon defines the verb to quine as denying resolutely the existence or importance of something real or significant — an affectionate coinage after W. V. O. Quine, whose radicalism consistently took the form of denying cherished distinctions (the analytic/synthetic, propositions, determinate reference). (quine not intended — but the pun is now load-bearing, so let it stand.) ↩
- The deduction theorem: A ⊢ B iff ⊢ A → B. This is exactly the “proved bridge” that makes the →/⊢ distinction earned rather than arbitrary. The claim of this essay is not that the distinction is false — it tracks something real — but that it conceals the subtractive nature of the inference it licenses. That is a different and stronger charge than arbitrariness, and it is worth keeping the concession visible so the best objection is answered rather than dodged. ↩
- FLAG — verify register before publishing. Robota (Slavic, “forced labor”) → Čapek’s R.U.R. is solid. The onward link to orphan runs through the Proto-Indo-European root orbʰ- (“bereft, passing to another’s charge”), which is a real but looser scholarly connection, not a settled etymological identity. The 2014 post asserted it plainly; this draft has softened to “the same deep root reaches toward.” Decide which register you want. “The word robot means orphan” overstates; “the same deep root that gives us orphan” is defensible. ↩
- FLAG — verify quotation. The Quine typewriter (a Remington modified to carry logical symbols, the question mark among the sacrificed keys) and the “I deal in certainties” reply circulate widely as Quine lore. I have not confirmed the exact wording against a primary source. This draft writes it as reported speech (“is said to have replied”) for that reason. If you have or can find the citation, tighten it; if not, keep the hedge — this essay of all essays should not inherit a fathered error. ↩
- Confirmed against sources. The punctus interrogativus (a diagonal stroke over a dot) appeared in late-8th-century Carolingian minuscule; it is traditionally credited to Alcuin of York, Charlemagne’s schoolmaster, but the earliest attested use appears in the Godescalc Gospel Lectionary (commissioned 781, the year before Alcuin arrived at Aachen), so the attribution to Alcuin is itself a fathering-story. The “lightning flash, striking from right to left” is Lynne Truss’s later description, not a contemporary one — which is why it serves the argument: the lightning is the screen. Alcuin did carry the Trivium (grammar, logic/dialectic, rhetoric) and stands behind the standardization of the mark either way.