Is Logic a Game of Asking and Giving Reasons?

What If Logic Was a Game of Questions and Answers?

Imagine you and a friend are arguing about the school sports day. You claim, “If Amara runs the relay, our team will win.” Your friend fires back, “Prove it! What if she trips?” You reply, “She never trips. Besides, we trained every day.” Your friend presses again, and you defend each part of your claim. Without realizing it, you’re playing a dialogue game — a back-and-forth of reasons and challenges. The philosopher Paul Lorenzen (1915–1994) thought this kind of exchange was the heart of logic itself.

In the 1960s, Lorenzen and his colleague Kuno Lorenz (born 1932) built dialogical logic, a framework that explains the meaning of logical words like “and,” “or,” and “if … then” by treating arguments as a game. Instead of interpreting logic with abstract truth tables and models, they set up two players — the Proponent (who defends a claim) and the Opponent (who attacks it). Every logical move is a turn in that game, and the rules decide whether the Proponent can win.

The Moves Allowed: How Particle Rules Work

To play, you need to know what counts as a legal move. Dialogical logic provides two kinds of rules: particle rules and structural rules. The particle rules tell you how to challenge and defend any statement that contains a logical constant. For example, if you say “A and B,” your opponent can pick either A or B and demand, “Give me that part.” You then must provide the chosen half. If you say “If A then B,” the opponent can challenge by stating A; you defend by stating B.

These rules define the local meaning of logical words — not by telling you what they “really are,” but by showing how to use them in an argument. Lorenzen and Lorenz were inspired by Ludwig Wittgenstein’s idea that meaning is use. So the particle rule for disjunction (“A or B”) is different: if you state A or B, the opponent can demand you pick a side to defend, and you, the speaker, get to choose which disjunct to give. The rule for universal quantification (“all x …”) lets the challenger pick a specific instance; you must then provide that instance.

Here’s a tiny play using the thesis “If (A and B) then A.” The Proponent (P) claims it. The Opponent (O) must find a weakness.

• Move 1: P states, “If (A and B) then A.”
• Move 2: O challenges the implication by stating “A and B.”
• Move 3: P, stuck, challenges O’s “A and B” by asking for the left part.
• Move 4: O defends and says “A.”
• Move 5: P now uses that “A” to answer the original challenge: “A.” O has no reply. P wins.

The play had a strict turn order and followed the particle rules. But what about broader rules that govern who can move when?

Winning Whenever It Matters: Strategies and Validity

Winning a single play doesn’t prove your thesis is logically valid — the Opponent might just have played poorly. For example, with the thesis “A and (if A then A),” the Opponent can choose the left conjunct and win immediately because the Proponent can’t produce the atomic “A” from nowhere. But if the Opponent foolishly picks the right conjunct, the Proponent will win. So a one-off victory isn’t enough.

Dialogical logic separates the play level (individual games) from the strategy level (a plan to cover all possible moves). A winning strategy for the Proponent is a complete conditional plan: no matter what the Opponent does, the Proponent’s replies lead to a win. A thesis is valid if and only if the Proponent has a winning strategy for it. This mirrors the proof-theoretic idea that a statement is provable when a systematic defense always succeeds.

The structural rules shape what strategies are possible. One key structural rule is the repetition rank, a number each player chooses at the start, capping how often they can challenge the same move. That keeps plays finite. But the pivotal structural rule for intuitionistic logic, the Last Duty First rule, says you must answer your opponent’s most recent unanswered challenge before you can go back to an older one. This restricts what you can defend, and it gives logic an intuitionistic flavor.

Consider the claim “A or not‑A” (the law of excluded middle). In an intuitionistic game, after the Opponent challenges the disjunction, the Proponent must defend immediately. Since “A” hasn’t been stated, he can only choose “not‑A.” The Opponent then attacks “not‑A” by stating “A,” and the Proponent, whose last duty is still the fresh challenge, has no legal move — he loses. So “A or not‑A” is not valid in intuitionistic dialogical logic.

Switching Logics by Changing the Rules

If we change the structural rule, we get a different logic. In the classical version, there’s no Last Duty First requirement. A player may answer any unanswered challenge, as long as they stay within their repetition rank. Replay the same “A or not‑A” game: the Proponent starts by defending with “not‑A,” the Opponent attacks with “A.” Now the Proponent can go back to the original challenge on the disjunction and, using the Opponent’s “A” as permission, defend with “A.” The Proponent wins. So classical logic appears simply by allowing information from later moves to be used earlier in the dialogue.

This idea — that different logics come from tweaking structural rules, not from changing the meaning of the logical constants — is called dialogical pluralism. It also works for modal logic (necessity and possibility). Each move happens in a labeled dialogical context, like a “possible world.” Particle rules for “necessarily” and “possibly” force context switches. Structural rules then regulate which contexts the Proponent is allowed to reach. For instance, the system K allows access only to contexts the Opponent has already opened; adding different permissions yields the systems T, B, S4, and S5. So a whole family of modal logics blossoms from the same dialogical seeds.

Why the Proponent Can’t Just Make Things Up

Perhaps the most famous dialogical rule is the Formal Rule (also called the Copy‑Cat Rule). In any play, the Proponent may state an atomic sentence — like “Amara is fast” — only if the Opponent has already stated it earlier. The Opponent faces no such restriction. This ensures that the Proponent can’t win by smuggling in unsupported facts. His defense must rely entirely on what the Opponent herself has granted during the debate.

This rule captures the idea of analyticity: a valid argument shouldn’t need outside help; its conclusion should follow from the terms of the debate itself. It also reflects an ancient insight from Plato’s dialogues: the best grounding for a claim is that your opponent has already accepted it.

Critics have complained that the standard framework is “too formal” — the dialogues never examine the actual content of atomic statements, only the logical scaffolding. To answer this, the Immanent Reasoning framework (developed by Shahid Rahman and others) enriches the language: every statement comes with an explicit local reason. Instead of just saying “Amara is fast,” the Proponent might say “The stopwatch shows 11.2 seconds: Amara is fast.” The opponent can then challenge that reason, leading to material dialogues where the content of real-world claims — about natural numbers, citizenship, or passport evidence — can be argued.

Immanent Reasoning also adds formation dialogues that allow players to first check whether a statement is even well-formed before the reason‑giving game begins. This makes the entire process, from syntax to deep content, happen inside the dialogue itself, without any outside metametal tools.

From Formal Games to Real Arguments: Why It Matters Today

You’ve probably never played a formal dialogical game, but you use its moves every day. When you defend an opinion by pointing to something your friend already admitted, you’re following the Formal Rule. When you demand, “Prove it,” and reject your friend’s attempt to change the subject, you’re enforcing the Last Duty First constraint. The dialogical approach shows that logical reasoning isn’t a mysterious black box — it’s a social practice of giving and asking for reasons.

This matters in law, science, and everyday life. A courtroom trial is a dialogue between prosecution and defense, each testing the other’s claims and drawing only on admitted evidence. Scientific peer review works similarly: a new finding must survive the challenges of skeptical colleagues. Recognizing logic as a dialogue game helps us see why these practices are reliable — they don’t just check facts; they check whether the argument can withstand any opponent’s best moves.

Lorenzen and Lorenz’s project also challenges the idea that meaning is something fixed by an external world. In dialogical logic, meaning emerges from the rules of interaction. That shift opened the door to comparing different logics within one framework — and even to seeing how logic itself might be negotiated in real debates. You may never build a formal winning strategy, but the next time you argue with a friend about the best movie or the fastest runner, you’ll know: every good defense is a move in an ancient and precise game.

Think about it

1. If you’re in a debate and your opponent accidentally says something that supports your side, should you be allowed to use it to win? Why might that be fair or unfair?
2. Suppose you could change one rule in a dialogue — for example, letting the Proponent introduce a new fact or requiring the Opponent to answer in a fixed time. Would the game still test logical validity, or would it test something else?
3. If someone always wins a certain kind of argument, does that mean their claims are true, or just that they are very good at the rules of that particular debate? What’s the difference?