Has Morality Without Principles Been Discovered? — Moral Particularism, Ontology, and Epistemology
1. Separate two claims
Moral particularism can contain at least two different theses.
P1: True morality or reason structure may be irreducibly context-sensitive.
P2: Humans with sufficient maturity or sensitivity can reliably recognize that structure in particular cases.
P1 ⇏ P2
This project treats P1 as a serious live possibility, while requiring an independent epistemology for P2. The central claim is:
Particularist ontology may be right, while particularist epistemology may be premature.
2. Ross — general pro tanto reasons, particular overall judgment
W. D. Ross treats promises, non-maleficence, gratitude, justice, and related considerations as general prima facie duties, usefully read today as pro tanto reasons.
promise → +R
harm → −R
Yet what one ought to do all things considered still requires judgment across the concrete circumstances. Ross can therefore be read as:
generalist about contributory reasons + particularist about overall judgment.
3. Dancy — context can change polarity, not merely weight
Jonathan Dancy's reasons holism goes further. A feature F may count in favor in one context, be irrelevant in another, and count against in a third.
F + C₁ → +R
F + C₂ → 0
F + C₃ → −R
Context therefore need not merely alter weight. It can alter whether a feature is a reason at all and which way it counts. Dancy thus radicalizes Rossian pluralism into holism.
As in the companion exploration “Are Reasons the Foundation of Normativity?”, this project does not assume that everything called a normative reason is already robustly normative. Here the focus is narrower: if robust normative reasons exist, could their structure itself be holistic?
4. Is particularism a modern intuitionism?
Classical intuitionism often moves from judgments about cases toward general self-evident prima facie principles.
case judgments → general prima facie principles
Dancy rejects that move. That F is a reason here does not establish that F is the same kind of reason elsewhere.
case → particular reason relation
In that sense, particularism can resemble “intuitionism without principles.” Historically, however, the Aristotelian and McDowellian lineage of phronesis, salience, and moral perception is at least as important as Ross.
5. McDowell — moral judgment as perception
On a McDowellian picture, the virtuous agent does not first register value-neutral facts and then apply a general rule. What matters is already salient within the situation.
situation → moral salience / perception → judgment
“Perception” need not mean literal sensory perception. It can include attention, sensitivity, and evaluative understanding. The central idea is that the right features show up with the right practical importance.
6. Skill-based particularism — not only seeing, but doing
A skill model treats moral judgment as expertise. Novices rely more heavily on explicit rules; experts acquire a capacity to classify cases, identify salient dimensions, integrate conflicting considerations, and handle novel situations.
novice → experience / training → expert competence
The perceptual question is:
What does the agent correctly see?
The skill question is:
What can the agent reliably do?
7. Perceptual particularism and perceptual intuitionism are different axes
Perceptual particularism concerns whether general principles are necessary for moral judgment. Perceptual intuitionism concerns whether moral perceptual experience can non-inferentially justify moral beliefs.
Particularism: Are general principles necessary?
Perceptual intuitionism: Can perceptual experience non-inferentially justify moral belief?
Thus perceptual intuitionism can coexist with generalism. One might non-inferentially see an act as cruel while also endorsing a general principle that cruelty normally counts against an act.
8. Do particularists assume that we can already judge correctly?
Dancy and McDowell do not treat ordinary humans as morally infallible. Bad education, vice, self-deception, inattention, and lack of experience can distort judgment. Yet a substantive assumption remains:
moral maturity / sensitivity → reliable particular judgment ?
The question is how that reliability has been established.
9. The calibration problem
Human moral sensitivity may be shaped by evolution, culture, education, emotion, social reinforcement, and power.
evolution + culture + learning + emotion + social reinforcement → 𝒱̂human
Using the distinction introduced in the companion reasons exploration:
𝒱*: the true normative structure.
𝒱̂human: the normative model represented by humans.
The question is:
Why suppose 𝒱̂human ≈ 𝒱* ?
The possibility that 𝒱* is particularist and the possibility that human sensitivity tracks it are distinct hypotheses.
10. Skill models face a bootstrapping problem
Skills normally improve through attempt, feedback, correction, and expertise. Chess has wins and losses; medicine has at least some external patient outcomes.
moral judgment → ?
What supplies the correctness signal for moral expertise? If the answer is “learn from virtuous exemplars,” another question appears: who identifies the exemplars? If existing social values do the selecting, then:
𝒱̂society → exemplar selection → training → 𝒱̂′
may describe cultural reproduction rather than convergence on 𝒱*.
11. Perceptual models face an appearance / reality gap
The same issue appears in perceptual language.
it appears normatively salient ⇏ it is normatively salient
That a feature appears important to a virtuous or sensitive agent does not by itself show that the appearance accurately represents an external normative structure. The possibility of moral perception must be separated from its calibrated reliability.
12. Returning to generalism does not solve the problem
This objection does not automatically favor Rossian generalism. Generalism also begins from human judgments and compresses them into principles.
human judgments → general principles
Systematization does not itself establish tracking of 𝒱*. The symmetry is:
Ross: risk of compressing human intuitions into general principles.
Dancy: risk of trusting human context-sensitive sensitivity too directly.
13. Dancy's ontological insight can still survive
The true normative structure may be high-dimensional and nonlinear, depending jointly on agents, relations, histories, intentions, alternatives, and context.
Rn = f(agent, relation, history, intention, alternatives, context, ...)
If so, a fixed-weight model such as “promise-breaking = −3” or “pain = −5” may be a very crude approximation. Dancyan holism can remain a serious hypothesis about 𝒱* even if Dancyan epistemology is too optimistic.
14. Particular judgments are observation data, not ground truth
Human judgments that one lie is wrong, another permissible, and another required may provide evidence about 𝒱*. But they should not be identified with the ground truth.
human particular judgments = observational evidence about 𝒱*
The instrument producing those observations — human moral cognition — may itself be poorly calibrated. Both general principles and case intuitions should therefore be treated as fallible ways of compressing or observing an unknown normative structure.
15. AI and posthuman stress test
Advanced AI or posthuman agents may possess unfamiliar affective repertoires, salience structures, much larger context windows, simultaneous multi-perspectival integration, or self-modifying value-recognition systems.
𝒱̂AI ≠ 𝒱̂human
Human phronesis therefore has no automatic privilege as the final model of normative cognition. Conversely, agents with richer representational and comparative capacities might be better able to infer a genuinely holistic 𝒱*.
16. Provisional conclusion
Moral particularism preserves an important possibility: true normative structure may not be a set of invariant general principles but a high-dimensional context-sensitive field. That ontological possibility fits this project well.
But it does not follow that present human moral sensitivity already reads that structure correctly.
Particularist ontology may be right, while particularist epistemology may be premature.
The project therefore gives neither Rossian general principles nor Dancyan case judgment final epistemic privilege. Both remain fallible evidence about 𝒱*.
Related discussions
Important background includes Aristotle on phronesis, W. D. Ross on prima facie duties, John McDowell's “Virtue and Reason,” David McNaughton's Moral Vision, Jonathan Dancy's Moral Reasons and Ethics Without Principles, Brad Hooker and Margaret Little's edited volume Moral Particularism, Sean McKeever and Michael Ridge's Principled Ethics, David Bakhurst on McDowell and Dancy, and Stephen Darwall's criticism of reason particularism. The focus here is not merely whether principles exist, but on separating ontological holism from epistemic tracking.