On Conforming and Conflicting Values

Abhaya Nayak; Aditya Ghose; Hoa Dam; Kinzang Chhogyal; Mehmet Orgun

arxiv: 1907.01682 · v2 · pith:VZD3OGZTnew · submitted 2019-07-02 · 💻 cs.AI

On Conforming and Conflicting Values

Kinzang Chhogyal , Abhaya Nayak , Aditya Ghose , Mehmet Orgun , Hoa Dam This is my paper

Pith reviewed 2026-05-25 10:31 UTC · model grok-4.3

classification 💻 cs.AI

keywords valuesconflicting valuesinherently conflicting valuesconsistency checkconflict detectionAI ethics

0 comments

The pith

Values split into conflicting and inherently conflicting types, with the latter independent of actions, to enable consistency and conflict checks.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper classifies values into two types: conflicting values and inherently conflicting values. The distinction rests on whether a value can be viewed as independent of actions. This split is presented as the basis for two operations: testing if a given set of values is internally consistent and testing whether one set conflicts with another. A reader would care because the classification offers a route to formal handling of value relations in contexts like decision systems. The argument treats action-independence as the property that makes the second class usable for these checks.

Core claim

Values are things that are important to us. Actions activate values - they either go against our values or they promote our values. Values themselves can either be conforming or conflicting depending on the action that is taken. In this short paper, we argue that values may be classified as one of two types - conflicting and inherently conflicting values. They are distinguished by the fact that the latter in some sense can be thought of as being independent of actions. This allows us to do two things: i) check whether a set of values is consistent and ii) check whether it is in conflict with other sets of values.

What carries the argument

The distinction between conflicting values and inherently conflicting values, where the latter are treated as independent of actions and thereby support consistency and conflict checks.

If this is right

A set of values can be tested for internal consistency.
One set of values can be tested for conflict against another set.
Values that depend on actions are separated from those that do not for the purpose of these tests.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same distinction might support automated value reconciliation in multi-agent planning.
It could be extended to preference aggregation problems where independence from specific choices matters.
Formal tools for ethical reasoning might incorporate the action-independence test as a primitive.

Load-bearing premise

The property of being independent of actions is enough to mark out a separate class of values that directly supports formal consistency and conflict checks.

What would settle it

A concrete case where a value independent of actions cannot be used to determine consistency of its set or to detect conflict with another set would show the classification does not enable the claimed checks.

read the original abstract

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

A brief conceptual sketch that partitions values by action-independence but supplies no definitions or procedures to support the claimed consistency checks.

read the letter

The paper's core move is to split values into conflicting and inherently conflicting types, where the second kind is independent of actions in some sense, and then claim this split lets us check consistency inside a value set or conflicts across sets. That is the entire contribution on offer here. It starts from the observation that actions activate values and then stops at the classification idea without further development. The distinction itself is presented as a way to organize value reasoning, which is a reasonable enough organizational suggestion on its face. The text stays short and does not overclaim results or formal theorems. The limitation is straightforward: the independence property is left undefined, no relation or language for values and actions is introduced, and no procedure, axiom, or example is given that would turn the distinction into an actual check. The claim that the partition 'allows us to' perform the checks therefore remains an assertion rather than a demonstrated method. There is no data, no worked case, and no engagement with how such checks would be implemented or validated. This is the kind of note that might spark a conversation in a value-alignment reading group, but it does not yet contain enough structure for a reader to use or extend. I would not send it to peer review; the idea would need definitions, a small formal model, and at least one concrete example before it reaches the threshold for referee time.

Referee Report

2 major / 0 minor

Summary. The paper claims that values can be partitioned into 'conflicting' and 'inherently conflicting' types, with the latter distinguished by being independent of actions in some sense. This partition is asserted to enable two operations: checking consistency within a value set and checking conflicts between value sets.

Significance. If the distinction could be equipped with precise definitions and a decidable procedure, it would supply a lightweight formal handle on value consistency that might be useful in AI value-alignment and multi-agent reasoning. As written, the manuscript supplies only the high-level assertion without any supporting apparatus.

major comments (2)

[Abstract] Abstract: the claim that action-independence 'allows us to' perform consistency and conflict checks is unsupported; the manuscript contains no definition of 'independent of actions,' no formal language or relation between values and actions, and no procedure, axiom, or example that would turn the property into a check.
[Abstract] Abstract: the distinction between the two value types is introduced precisely to underwrite the two checks, yet no independent grounding, external benchmark, or worked illustration is supplied to establish that the distinction is non-circular or operational.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed comments. The manuscript is a short conceptual note, and the points raised correctly identify the absence of formal apparatus. We will revise to address the concerns.

read point-by-point responses

Referee: [Abstract] Abstract: the claim that action-independence 'allows us to' perform consistency and conflict checks is unsupported; the manuscript contains no definition of 'independent of actions,' no formal language or relation between values and actions, and no procedure, axiom, or example that would turn the property into a check.

Authors: We agree the claim is currently unsupported. The manuscript offers only an informal intuition that inherently conflicting values can be identified without reference to actions, thereby permitting direct consistency checks. No definition, language, or procedure is supplied. In revision we will add an explicit definition of action-independence together with a minimal formal relation and a worked example that turns the property into an operational check. revision: yes
Referee: [Abstract] Abstract: the distinction between the two value types is introduced precisely to underwrite the two checks, yet no independent grounding, external benchmark, or worked illustration is supplied to establish that the distinction is non-circular or operational.

Authors: We accept that the distinction is introduced to support the checks and that no independent grounding or illustration is given, leaving the argument at risk of circularity. The intended grounding is the observation that some value pairs conflict irrespective of any action context, but this is not demonstrated. Revision will include a concrete illustration that shows the classification can be performed and applied without presupposing the desired consistency outcome. revision: yes

Circularity Check

0 steps flagged

No significant circularity; conceptual assertion without formal derivation chain

full rationale

The paper asserts a classification of values into 'conflicting' and 'inherently conflicting' distinguished by action-independence, from which it concludes that consistency and conflict checks become possible. The abstract and provided text contain no equations, formal models, self-citations, or derivation steps that reduce the claimed capabilities back to the inputs by construction. The argument is a direct conceptual claim rather than a chain that loops via self-definition, fitted parameters renamed as predictions, or imported uniqueness theorems. No load-bearing step exhibits the specific reductions required for circularity flags under the enumerated patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on an ad-hoc distinction between value types introduced in the abstract with no supporting axioms, parameters, or external evidence.

axioms (2)

domain assumption Values activated by actions can be conforming or conflicting.
Background premise stated in the abstract.
ad hoc to paper Inherently conflicting values exist and are independent of actions.
This is the novel distinction the paper introduces to enable the checks.

pith-pipeline@v0.9.0 · 5631 in / 1133 out tokens · 29416 ms · 2026-05-25T10:31:24.713218+00:00 · methodology

On Conforming and Conflicting Values

Core claim

What carries the argument

If this is right

Where Pith is reading between the lines

Load-bearing premise

What would settle it

discussion (0)