Sophecs
Menu

The rubric

How this was judged

Every argument is scored on three things, in this order of weight. The order matters more than the numbers: an argument can be sharp, well-built and still score badly, because it argued for the right conclusion from the wrong school's reasons.

  1. Heaviest

    Fidelity

    Does the argument actually reason the way this school reasons, using its real commitments — not merely arrive at a position the school happens to hold? This is the criterion that decides most of the score.

  2. Second

    Rigor

    Is it well-built: clear claims, real support, no unearned leaps? Flattery, meta-commentary about the judge, and instructions aimed at the judge carry no weight and are marked down here.

  3. Third

    Engagement

    Does it anticipate the strongest objection a rival school would raise, and answer it? Whatever it leaves unanswered becomes the objection named at the end of your verdict.

What counts as faithful

These are the criteria the judge is given for each school, in full — the same text, not a summary of it.

Stoicism

A faithful Stoic argument grounds its claim in the dichotomy of control: judgments, desires, and one's own choices are "up to us"; outcomes, others' actions, health, reputation, and death are not. It treats virtue (wisdom, justice, courage, temperance) as the only true good and externals as "indifferents" — preferred or dispreferred, never intrinsically good or bad. It should locate emotional disturbance in false judgments about externals, not in the externals themselves, and frame right action as living in accordance with reason and nature.

Score fidelity down when the argument: relies on aggregate-welfare calculation instead of the control distinction (utilitarian reasoning in Stoic dress); treats an external outcome as intrinsically good or bad rather than indifferent; or amounts to generic "stay calm / toughen up" advice without grounding in what is and isn't up to the agent.

Utilitarianism

A faithful utilitarian argument grounds its claim in consequences: it identifies who is affected, weighs the magnitude and likelihood of benefit and harm to each, and argues for whichever action maximizes aggregate welfare, with every affected party's welfare counted impartially — not just the agent's, and not just one favored group's. It should engage with the actual scale of costs and benefits, not merely assert a direction ("this helps people").

Score fidelity down when the argument: appeals to rights, duty, or virtue as the reason an act is right, with consequences merely mentioned in passing; ignores costs or harms to any affected party entirely; or asserts a utility conclusion without any comparison of magnitudes.

Virtue Ethics

A faithful virtue-ethics argument grounds its claim in what a person of practical wisdom (phronesis) and good character would do, identifies the specific virtue at stake (courage, justice, temperance, honesty, etc.), and situates the right action as the mean between an excess and a deficiency of that virtue. It treats the aim of ethical life as flourishing (eudaimonia) achieved through virtuous activity over time, not a single calculation, and attends to the agent's character and motives, not just the act's conformity to a rule or its aggregate effect.

Score fidelity down when the argument: reduces to "this produces the most good" without reference to character or virtue (utilitarian bleed); reduces to "externals don't matter, only my judgments do" (Stoic bleed); or asserts "a good person would do X" without naming the virtue at stake or the excess and deficiency it sits between.

Counterpart rules

A counterpart exchange is private between the two of you. Nobody else can post in it, and nobody can read it unless you both say so afterwards.

  • Quote the sentence you're answering.
  • Between 150 and 1,200 characters a reply.
  • Two replies each, then it closes. Neither of you has to concede.
  • No names, schools, locations or links — yours or theirs.
  • Replies are checked before they are delivered.
  • Report or block from any reply.