The representation of ethical dilemma is admittedly a quite difficult topic. On the first hand, the very definition of what constitute a dilemma is subject to numerous vivid discussions and debates among philosophers (from Kant denying the existence of such things as ethical dilemma to Sartre and the famous example of the existential dilemma).
Nevertheless, it seems plausible that since ethical dilemma actually exists (at least according to persons subject to these dilemma) and therefore should be representable. Here comes another difficulty, more theoretical and grounded in the very essence of what constitute a dilemma. A dilemma, it would seem, is a contradiction, a conflict between two (or more) obligations that can't be mutually fulfilled.
Modal logic and particularly deontic logic studies the relations between Propositional Calculus and the deontic operators ([] obligation and permission). I have not been able to find an interpretation of these operators in description logic. I would be interested in opinion concerning relations between deontic operators and the usual constructors of description logic (set intersections/existential and universal quantification).
Many so-called ethical dilemmas may arise because general rules specifying obligations are regarded as if absolute when they are really subject to implicit ceteris paribus conditions. Those conditions also raise epistemological problems. W.D. Ross distinguished between prima facie duties and duties proper, to resolve conflicting obligations. Duties proper were arrived at by assigning different weightings, relative to context and circumstances, to the prima facie duties.
- Ross, W.D. The Right and the Good (1930). Philip Stratton-Lake, ed. New York: Oxford University Press, 2002.
- Ross, W.D. Foundations of Ethics: the Gifford Lectures Delivered at the University of Aberdeen, 1935-6. Oxford: Clarendon Press, 1939.
I don't quite understand the issue from my point of view, which is admittedly like that of a boy with his nose pressed against a store window. Aren't (some) description logics just syntactic variants of modal logics? And since deontic logic is just a modal logic with a different interpretation for the modal operators and some different axioms, deontic logic must also have a similar syntactic variant. Anyway, here are a few links that seem to support that idea, but of course I myself am in no position to judge how useful they might be to your project:
Article Unification in modal and description logics
http://www.diag.uniroma1.it//~nardi/dlhb/chapdlhb/dlhb-06.pdf
http://ceur-ws.org/Vol-11/derijke-inv.ps
https://en.wikipedia.org/wiki/Description_logic#Modal_logic
Cheers.
Karl Pfeifer Thank you for these precision that give me the opportunity to clarify the use of the word dilemma in the way I try to use it.
On the first hand, I think we can acknowledge the existence of moral dilemma as reported subjective experiences. There are human being that have reported being in situation of moral dilemma, quite undeniably.
A recent example can be seen in medic that had to perform triage (patient selection) for emergency room due to COVID. There is a quite interesting analysis here : Article Ethical dilemmas due to the Covid-19 pandemic
The paper emphasize that medical practictionners reported durable mental health problem from having been placed in this terrible situation.
A classic example of dilemma is given by Sartre about a young man during WWII that is thorn apart between his (subjective) obligation to join the resistance army and his (subjective) obligation to take care of his sick mother.
From these example, I would be inclined to think that moral dilemma actually exist and that they have material effect on person experiencing them. These material effect seem to be close to remorse, which suggest that people having experienced such dilemma are somewhat blaming themselves.
From examining some properties of a well-crafted deontic logic, I think we might be able to get a better understanding of this curious phenomenon. It might indeed seem surprising that people having experienced such impossible to satisfy obligation are suffering. Those people are rational and knows perfectly that they physically could not satisfy contradictory obligations, and yet they report and experience moral suffering.
Karl Pfeifer Thank you for your second remark concerning the relation between description logics and modal logics. This give me again the opportunity to clarify something that I think deserve some clarifications.
There are indeed quite strong relations between description logics and modal logics but, up to now, I've not found any paper that gives a clear overview of the question. So I'll try to give a quick and dirty clarification of how I view things.
Modal logic have been developed, in modern time, on the ground of Kripke possible world semantic interpretation. The mathematical formalization is mainly due to Von Wright work that is mostly influenced by Kripke semantic of possible worlds.
Description logic have been developed only in modern time and the mathematical formalization of their semantic is the one of Tarski. Alfred Tarski was more of a "pure" mathematician and he proposed a definition of what should be a semantic that is extremely formal (you can look at "model theory").
Ultimately, it seems to me that Tarski definition is congruent with Kripke/Von Wright one but I've not been able to find a paper confirming that.
Karl Pfeifer As I am reading me back I realize that my answers lack references and clarity, please indulge me about that. I was quite excited that someone replied to this discussion and wanted to reply back right away!
I'll provide a more in depth response tomorrow about the relation between modal logic and description logic.
- Badges
- Science topic
- Similar topics
- Mathematics