Recognition: unknown
Defeasible Conditional Obligation in a Two-tiered Preference-based Semantics (Extended Version)
Pith reviewed 2026-05-07 14:34 UTC · model grok-4.3
The pith
A two-tiered preference semantics with separate ideality and normality orderings models defeasible conditional obligations that withdraw with new conflicting information.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
By employing two separate orderings on worlds—one for ideality and one for normality—each equipped with its own ranking method, a preference-based semantics for dyadic deontic logic can incorporate nonmonotonic reasoning. This enables the withdrawal of previously derived conditional obligations when new, potentially conflicting information is added, while preserving overall consistency and meeting a number of nonmonotonic postulates.
What carries the argument
The bi-preferential structure of two orderings (ideality and normality) on worlds, each with a dedicated ranking method, that together support nonmonotonic withdrawal of conditional obligations.
Load-bearing premise
That the two separate orderings and their ranking methods together implement nonmonotonic withdrawal of obligations correctly and without producing inconsistencies.
What would settle it
A concrete example consisting of a set of worlds, an initial conditional obligation, and a new fact where the semantics either retains the obligation that should be withdrawn or produces an inconsistency in the deontic statements.
Figures
read the original abstract
In response to a concern raised by Horty, this paper develops a two-tiered, preference-based semantic framework for modeling defeasible conditional obligations. The paper extends a Hansson-Lewis style preference semantics for dyadic deontic logic by incorporating a nonmonotonic reasoning mechanism that enables previously derived obligations to be withdrawn when new, potentially conflicting information comes in. The account is bi-preferential: two orderings--ideality and normality--on worlds are employed to address shortcomings in earlier approaches, with a separate ranking method for each. At the nonmonotonic layer, a number of postulates are considered, including antecedent strengthening, inclusion and no-drowning. A connection is established with so-called constrained input/output (I/O) logic--an existing standard for normative reasoning based on a different methodology.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper develops a two-tiered preference-based semantic framework for modeling defeasible conditional obligations by extending Hansson-Lewis dyadic deontic logic. It introduces a bi-preferential account employing independent ideality and normality orderings on worlds (with separate ranking methods for each) to support nonmonotonic reasoning, allowing previously derived obligations to be withdrawn upon new conflicting information. The framework considers postulates including antecedent strengthening, inclusion, and no-drowning at the nonmonotonic layer and establishes a connection to constrained input/output logic.
Significance. If the formal details and proofs hold, the work offers a semantic solution to limitations in prior preference-based deontic logics for handling defeasibility, directly responding to concerns such as those raised by Horty. The explicit link to constrained I/O logic unifies two distinct methodologies in normative reasoning, and the use of two independent orderings provides a clean mechanism for nonmonotonic withdrawal without introducing inconsistencies.
minor comments (3)
- [Abstract] The abstract refers to 'a number of postulates' (antecedent strengthening, inclusion, no-drowning) but does not enumerate them fully or indicate where they are formally stated; add an explicit list or reference to the relevant section for clarity.
- [Semantics] Notation for the two orderings (ideality and normality) and their ranking methods should be introduced with a single consistent symbol table or definition block early in the semantics section to avoid reader confusion when comparing to standard Hansson-Lewis models.
- [Introduction] The connection to constrained I/O logic is claimed but the precise translation or embedding theorem is not previewed; include a brief statement of the correspondence result (e.g., which I/O rules are captured) in the introduction or conclusion.
Simulated Author's Rebuttal
We thank the referee for the positive summary and significance assessment of our manuscript, as well as the recommendation for minor revision. We are pleased that the bi-preferential framework and its connection to constrained input/output logic were viewed favorably as addressing limitations in prior preference-based deontic logics.
Circularity Check
No significant circularity identified in the derivation chain
full rationale
The paper defines a bi-preferential semantics by extending the Hansson-Lewis preference-based account for dyadic deontic logic, introducing separate ideality and normality orderings on worlds together with their respective ranking methods to support nonmonotonic withdrawal of obligations. It then considers a set of postulates (antecedent strengthening, inclusion, no-drowning) and proves a connection to the independent constrained input/output logic framework. These steps consist of explicit semantic definitions, postulate verification, and cross-framework theorems rather than any self-definitional reductions, fitted inputs renamed as predictions, or load-bearing self-citations that collapse the central claims back onto the paper's own inputs by construction. The account remains self-contained against external benchmarks.
Axiom & Free-Parameter Ledger
axioms (1)
- domain assumption Hansson-Lewis style preference semantics for dyadic deontic logic
invented entities (1)
-
Bi-preferential semantics with separate ideality and normality orderings
no independent evidence
Reference graph
Works this paper leans on
-
[1]
Defeasible normative reasoning: A proof-theoretic integration of logi- cal argumentation. In Wooldridge, M.; Dy, J.; and Natarajan, S., eds.,Thirty-Eighth AAAI Conference on Artificial Intel- ligence, AAAI 2024, Thirty-Sixth Conference on Innovative Applications of Artificial Intelligence, IAAI 2024, 10450– 10458. AAAI Press. Benferhat, S.; Cayrol, C.; Du...
2024
-
[2]
Intell.287:103348
De- signing normative theories for ethical and legal reasoning: Logikey framework, methodology, and tool support.Artif. Intell.287:103348. Extended abstract presented atKR 2020 (Recently Published Research track). Bochman, A. 2005.Explanatory Nonmonotonic Reasoning. Word Scientific. Boutilier, C
2020
-
[3]
In Calimeri, F.; Leone, N.; and Manna, M., eds.,JELIA, volume 11468 ofLNCS, 182–197
Taking defeasible entailment beyond rational closure. In Calimeri, F.; Leone, N.; and Manna, M., eds.,JELIA, volume 11468 ofLNCS, 182–197. Springer. Chellas, B. 1980.Modal Logic. Cambridge: Cambridge University Press. Chisholm, R
1980
-
[4]
In Marquis, P.; Son, T
Streamlining in- put/output logics with sequent calculi. In Marquis, P.; Son, T. C.; and Kern-Isberner, G., eds.,Proceedings of KR 2023, 146–155. Ciabattoni, A.; Olivetti, N.; and Parent, X
2023
-
[5]
Dyadic obligations: Proofs and countermodels via hypersequents. In Aydogan, R.; Criado, N.; Lang, J.; S ´anchez-Anguix, V .; and Serramia, M., eds.,PRIMA 2022: Principles and Prac- tice of Multi-Agent Systems - 24th International Conference, 54–71. Springer. Delgrande, J. P
2022
-
[6]
In Calvanese, D.; Erdem, E.; and Thielscher, M., eds.,Proceedings of KR 2020, 326–335
A preference-based approach to de- feasible deontic inference. In Calvanese, D.; Erdem, E.; and Thielscher, M., eds.,Proceedings of KR 2020, 326–335. Forrester, J
2020
-
[7]
Intell.98(1):209–235
Preferential reasoning in the perspective of Poole default logic.Artif. Intell.98(1):209–235. Gabbay, D.; Horty, J.; Parent, X.; van der Meyden, R.; and van der Torre, L., eds. 2013.Handbook of Deontic Logic and Normative Systems, volume
2013
-
[8]
In Gabbay et al
Prima facie norms, normative conflicts and dilemmas. In Gabbay et al. (2013). 241–35. Goble, L
2013
-
[9]
Defaults with priorities.J. Philos. Log. 36(4):367–413. Horty, J. 2012.Reasons as Defaults. Oxford University Press. Horty, J
2012
-
[10]
InProceedings of KR 1991, 357–367
S4 as the conditional logic of nonmono- tonicity. InProceedings of KR 1991, 357–367. Morgan Kaufmann Publishers Inc. Lehmann, D., and Magidor, M
1991
-
[11]
Another perspective on default reason- ing.Ann. Math. Artif. Intell.15(1):61–82. Lewis, D. 1973.Counterfactuals. Oxford: Blackwell. Lewis, D
1973
-
[12]
In Gabbay et al
Input/output logic. In Gabbay et al. (2013). 499–544. Parent, X., and van der Torre, L
2013
-
[13]
In Broersen, J
I/O logics with a con- sistency check. In Broersen, J. M.; Condoravdi, C.; Shyam, N.; and Pigozzi, G., eds.,Deontic Logic and Normative Sys- tems - DEON 2018, 285–299. College Publications. Parent, X
2018
-
[14]
1st Polish Congress of Logic, Torun, September 2025 (Invited lecture)
Defeasible conditional imperatives. 1st Polish Congress of Logic, Torun, September 2025 (Invited lecture). Stolpe, A
2025
-
[15]
van der Torre, L
Priority structures in deontic logic.Theoria80(2):116–152. van der Torre, L. 1997.Reasoning about Obligations. Ph.D. Dissertation, University of Erasmus. van Fraassen, B
1997
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.