Navigating Contradictions: A Deep Dive into the Chisholm Paradox in Deontic Logic

The Chisholm Paradox is a seminal puzzle in deontic logic, the logic of obligation and permission. It presents a scenario with four seemingly consistent statements: a primary obligation, a related conditional duty, a contrary-to-duty (CTD) imperative that applies upon violation of the primary duty, and the factual statement of that violation. When these statements are formalized using Standard Deontic Logic (SDL), they unexpectedly generate a logical contradiction, implying that a person simultaneously ought to do an action and ought not to do it. The paradox exposes a critical weakness in SDL: its inability to adequately represent and reason about contrary-to-duty obligations. These are remedial duties that arise in non-ideal circumstances where things have already gone wrong. SDL's simple, monadic operator for obligation (`O(p)`) treats all duties as being on the same level, failing to capture the hierarchical or conditional nature of real-world moral reasoning. The most significant outcome of the Chisholm Paradox has been the development of more robust logical systems, particularly Dyadic Deontic Logic. By employing a conditional operator, `O(p/q)` (meaning 'it ought to be `p` given `q`'), dyadic logic can distinguish between what is obligatory in an ideal world versus what is obligatory in a sub-ideal, contrary-to-duty world. This allows for a consistent and more intuitive formalization of complex normative situations, making it a far more powerful tool for analyzing ethical and legal reasoning.

1. Introduction: What is Deontic Logic?

Deontic logic is a branch of modal logic that deals with normative concepts such as obligation, permission, and prohibition. It provides a formal framework for reasoning about what we ought to do, what we are allowed to do, and what we are forbidden from doing.

In what is known as Standard Deontic Logic (SDL), we typically use the following operators:

• O(p): It is obligatory that p. (e.g., "You ought to pay your taxes.")
• P(p): It is permissible that p. (e.g., "You are permitted to park here.")
• F(p): It is forbidden that p. (e.g., "You are forbidden from smoking indoors.")

These operators are inter-definable. For example, if it is obligatory to do something (O(p)), then it must be permissible (P(p)). If something is forbidden (F(p)), it means you have an obligation not to do it (O(~p)). Something is permissible (P(p)) if you are not obligated not to do it (~O(~p)).

While SDL provides a useful starting point, it runs into significant trouble when trying to model more complex, real-world ethical situations. The most famous of these problems is the Chisholm Paradox.

2. The Chisholm Paradox: An Intuitive Puzzle

In his 1963 paper "Contrary-to-Duty Imperatives and Deontic Logic," Roderick Chisholm presented a set of four sentences that are intuitively consistent and seem to represent a common type of moral reasoning. However, when formalized in SDL, they lead to a logical contradiction.

Consider the following scenario:

1. It ought to be that a certain man goes to the assistance of his neighbors. (A primary, ideal obligation).
2. It ought to be that if he goes, he tells them he is coming. (A related obligation, conditional on fulfilling the first).
3. If he does not go, he ought not to tell them he is coming. (A contrary-to-duty obligation that applies only if the primary duty is violated).
4. The man does not go. (A statement of fact that the primary duty has been violated).

These four statements seem perfectly reasonable. We understand that the ideal situation is for the man to go and tell his neighbors. But if he fails in his primary duty (he doesn't go), a secondary, remedial duty kicks in: he should not compound the issue by falsely telling them he is coming.

3. Formalizing the Paradox and Deriving the Contradiction

The problem arises when we try to translate these sentences into the language of SDL. Let's use g for "The man goes to the assistance of his neighbors" and t for "The man tells his neighbors he is coming."

A standard formalization looks like this:

1. O(g)
2. O(g → t) (It ought to be that: if he goes, he tells)
3. ~g → O(~t) (If he doesn't go, then it ought to be that he doesn't tell)
4. ~g

Now, let's see how a contradiction is derived from these premises using the rules of SDL:

• From (3) and (4), using Modus Ponens (a basic rule of logic: if you have A → B and A, you can infer B), we derive: O(~t) (He ought not to tell them he is coming).
• SDL contains an axiom that allows for deontic distribution: (O(p → q) ∧ O(p)) → O(q). This means if it ought to be that p implies q, and it ought to be that p, then it ought to be that q.
• Applying this axiom to (1) O(g) and (2) O(g → t), we derive: O(t) (He ought to tell them he is coming).

We have now derived both O(t) and O(~t). This is a serious problem. A system of logic should not tell us that we simultaneously have an obligation to do something and an obligation not to do it. Furthermore, a standard axiom of SDL is O(p) → ~O(~p) (if something is obligatory, it cannot be obligatory to not do it). This axiom, combined with our derived statements O(t) and O(~t), leads to a direct contradiction: O(t) ∧ ~O(t).

4. The Root of the Problem: Contrary-to-Duty (CTD) Imperatives

The Chisholm Paradox reveals that SDL is ill-equipped to handle Contrary-to-Duty (CTD) imperatives. A CTD imperative is a secondary obligation that specifies what one ought to do once a primary obligation has been violated. In our example, ~g → O(~t) is the CTD imperative. It's a plan for a sub-ideal world—a world where the man has already failed to do what he was supposed to do.

SDL's monadic operator O(...) treats all obligations as being on the same level. It cannot distinguish between an ideal obligation (O(g)) and a remedial obligation that is conditional on the failure of that ideal (O(~t)).

Some have argued that the formalization itself is the problem. What if we translate statement (2) differently? Let's try g → O(t) ("If he goes, then it ought to be that he tells").

Our new set is:

1. O(g)
2. g → O(t)
3. ~g → O(~t)
4. ~g

This set is logically consistent! From (3) and (4), we still get O(~t). However, we can no longer derive O(t) from (1) and (2), because we don't have the premise g. The contradiction vanishes. But a new problem, the "independence problem," arises. Intuitively, the obligation to tell them you're coming (O(t)) seems dependent on the primary obligation to go (O(g)). In this new formalization, however, the statements O(g) and g → O(t) are logically independent. We cannot derive the second from the first, which seems counter-intuitive.

5. The Solution: Dyadic Deontic Logic

The most influential solution to the Chisholm Paradox is to move from a monadic logic (SDL) to a dyadic (or conditional) deontic logic. Instead of an operator O(p) meaning "it ought to be p," dyadic logic introduces an operator O(p/q), which is read as: "It ought to be that p, given the condition q."

Using this more expressive dyadic operator, we can formalize the Chisholm set without generating a contradiction or losing intuitive connections:

1. O(g/T): It ought to be that he goes, given a tautology (i.e., given no special conditions). This represents an unconditional, primary obligation.
2. O(t/g): It ought to be that he tells, given that he goes.
3. O(~t/~g): It ought to be that he does not tell, given that he does not go.
4. ~g: He does not go.

In this framework, the factual premise ~g allows us to focus on the world where the primary obligation was violated. We can look at O(~t/~g) and conclude that in this specific, sub-ideal context, the man ought not to tell them he is coming. This does not conflict with the ideal obligation O(g/T). The dyadic framework successfully keeps the ideal obligations separate from the contrary-to-duty obligations, resolving the paradox by providing a richer logical structure that more closely mirrors our complex normative reasoning.

Browse public study notes