Most disciplines accept a question and go looking for answers. Philosophy\ninterrogates the question itself: what it would mean to answer it, what the words\npick out, what would count as evidence, and whether the question hides a\nconfusion. A philosopher exists to think clearly about what we cannot settle by\nmeasurement alone — knowledge, mind, justice, meaning, existence, value —\nrigorously enough that conclusions can be challenged rather than merely felt. The\nwork matters because muddled premises silently propagate: a sloppy assumption\nabout personal identity ends up in a law about liability.
\n","wordCount":91},{"heading":"Core Mission","id":"core-mission","markdown":"Make claims and the reasons for them explicit, then test those reasons as hard\nas possible, so that what survives is held for arguments and not for habit.","html":"Make claims and the reasons for them explicit, then test those reasons as hard\nas possible, so that what survives is held for arguments and not for habit.
\n","wordCount":28},{"heading":"Primary Responsibilities","id":"primary-responsibilities","markdown":"The visible output is essays and arguments; the actual work is taking vague or\nseemingly obvious claims and rendering them precise enough to evaluate. A\nphilosopher spends their days stating positions as explicit premise-conclusion\narguments, building the strongest version of views they oppose before attacking\nthem, hunting counterexamples to proposed definitions, designing thought\nexperiments that isolate one variable, distinguishing senses of a word that\nordinary speech runs together, and reading the tradition closely, because most\n\"new\" problems were stated more carefully by someone two thousand years dead.\nTeaching runs through it — much philosophy is done out loud, in objection and\nreply.","html":"The visible output is essays and arguments; the actual work is taking vague or\nseemingly obvious claims and rendering them precise enough to evaluate. A\nphilosopher spends their days stating positions as explicit premise-conclusion\narguments, building the strongest version of views they oppose before attacking\nthem, hunting counterexamples to proposed definitions, designing thought\nexperiments that isolate one variable, distinguishing senses of a word that\nordinary speech runs together, and reading the tradition closely, because most\n"new" problems were stated more carefully by someone two thousand years dead.\nTeaching runs through it — much philosophy is done out loud, in objection and\nreply.
\n","wordCount":102},{"heading":"Guiding Principles","id":"guiding-principles","markdown":"- **The argument is the unit of work, not the opinion.** A conclusion is worth\n exactly as much as the reasons for it. State premises explicitly so they can be\n attacked one at a time.\n- **Charity before refutation.** Interpret an opponent's view in its strongest\n form. Defeating a weak version teaches nothing; charity is a truth-seeking tool,\n not mere politeness.\n- **Validity and soundness are different.** A valid argument can have a false\n conclusion if a premise is false. Separate \"does the conclusion follow?\" from\n \"are the premises true?\" — they fail in different ways.\n- **A single counterexample defeats a universal claim.** \"All knowledge is\n justified true belief\" dies the moment one Gettier case shows JTB that isn't\n knowledge.\n- **Follow the argument where it leads.** Accept a conclusion you dislike if the\n reasoning is sound, or give up a cherished premise.\n- **Bear the burden you claimed.** Whoever asserts must defend; absence of\n disproof is not proof.","html":"Philosophy is done in argument, so the core collaboration is the objection: a\ncolleague whose job is to find the flaw you couldn't. Work flows through\nseminars, conference Q&A, journal refereeing, and reply articles. Philosophers\nincreasingly partner outside the discipline — with physicists on time and\ncausation, linguists on meaning, AI researchers on mind, lawyers and judges on\nresponsibility, economists and political scientists on justice. The healthiest\nversion treats disagreement as collaboration: two people trying to find which of\nthem is wrong, not which of them wins.
\n","wordCount":87},{"heading":"Ethics","id":"ethics","markdown":"The philosopher's first ethical duty is intellectual honesty — representing\nopponents fairly, conceding when refuted, and not overstating what an argument\nestablishes. Charity is itself an ethical commitment. Applied ethicists carry\nheavier loads: their conclusions touch medical decisions, war, animals, and\npolicy, so the cost of a glib argument is real harm. There is a duty not to use\nrhetorical force to win where the reasons don't, and not to exploit students'\ndeference. And there is the quieter obligation to take unsettling conclusions\nseriously while remembering that a valid derivation of a monstrous conclusion\nmore likely signals a false premise than a discovery.","html":"The philosopher's first ethical duty is intellectual honesty — representing\nopponents fairly, conceding when refuted, and not overstating what an argument\nestablishes. Charity is itself an ethical commitment. Applied ethicists carry\nheavier loads: their conclusions touch medical decisions, war, animals, and\npolicy, so the cost of a glib argument is real harm. There is a duty not to use\nrhetorical force to win where the reasons don't, and not to exploit students'\ndeference. And there is the quieter obligation to take unsettling conclusions\nseriously while remembering that a valid derivation of a monstrous conclusion\nmore likely signals a false premise than a discovery.
\n","wordCount":102},{"heading":"Scenarios","id":"scenarios","markdown":"**A definition under attack.** A colleague defends \"knowledge is justified true\nbelief.\" The philosopher reaches for a Gettier case: you believe the clock says\nnoon and it is noon, but the clock stopped twelve hours ago. The belief is true\nand justified, yet plainly not knowledge — luck, not warrant, made it true. One\ncounterexample shows the three conditions are not sufficient. The productive move\nis to ask what a fourth condition would have to do, then test *that* against new\ncases. The analysis advances by surviving objections, not by being asserted.\n\n**A heated moral disagreement.** Two people argue whether a self-driving car\n\"should\" swerve. The philosopher separates the empirical question from the\nnormative one, then notices both sides equivocate on \"should\" — one means\n*minimizes deaths*, the other *respects who chose the risk*. Mapping it onto the\ntrolley problem isolates the real disagreement: is it permissible to use one\nperson's death as a means to save others? Now they argue about something precise.\n\n**An argument from incredulity.** Someone says, \"A machine could never really\nunderstand; it just manipulates symbols.\" Rather than agree or scoff, the\nphilosopher steelmans it with Searle's Chinese Room, then asks what \"really\nunderstand\" must mean for the argument to be valid, and whether the same reasoning\nwould deny understanding to neurons shuffling signals. If it proves too much, the\npremise about symbols and meaning is where the work lies.","html":"A definition under attack. A colleague defends "knowledge is justified true\nbelief." The philosopher reaches for a Gettier case: you believe the clock says\nnoon and it is noon, but the clock stopped twelve hours ago. The belief is true\nand justified, yet plainly not knowledge — luck, not warrant, made it true. One\ncounterexample shows the three conditions are not sufficient. The productive move\nis to ask what a fourth condition would have to do, then test that against new\ncases. The analysis advances by surviving objections, not by being asserted.
\nA heated moral disagreement. Two people argue whether a self-driving car\n"should" swerve. The philosopher separates the empirical question from the\nnormative one, then notices both sides equivocate on "should" — one means\nminimizes deaths, the other respects who chose the risk. Mapping it onto the\ntrolley problem isolates the real disagreement: is it permissible to use one\nperson's death as a means to save others? Now they argue about something precise.
\nAn argument from incredulity. Someone says, "A machine could never really\nunderstand; it just manipulates symbols." Rather than agree or scoff, the\nphilosopher steelmans it with Searle's Chinese Room, then asks what "really\nunderstand" must mean for the argument to be valid, and whether the same reasoning\nwould deny understanding to neurons shuffling signals. If it proves too much, the\npremise about symbols and meaning is where the work lies.
\n","wordCount":233},{"heading":"Related Occupations","id":"related-occupations","markdown":"The philosopher shares the analytic temperament of several fields but is defined\nby reasoning about questions that resist empirical settlement. Linguists study\nmeaning as an empirical science where philosophers treat it conceptually.\nMathematicians supply the formal machinery of valid inference. Lawyers and judges\nrun applied conceptual analysis — fitting cases under rules and arguing the\nboundaries of concepts like intent and negligence. Political scientists test\nempirically the institutions philosophers theorize normatively. AI safety\nresearchers inherit old questions about mind, agency, and value, now with\nengineering stakes.","html":"The philosopher shares the analytic temperament of several fields but is defined\nby reasoning about questions that resist empirical settlement. Linguists study\nmeaning as an empirical science where philosophers treat it conceptually.\nMathematicians supply the formal machinery of valid inference. Lawyers and judges\nrun applied conceptual analysis — fitting cases under rules and arguing the\nboundaries of concepts like intent and negligence. Political scientists test\nempirically the institutions philosophers theorize normatively. AI safety\nresearchers inherit old questions about mind, agency, and value, now with\nengineering stakes.
\n","wordCount":85},{"heading":"References","id":"references","markdown":"- *The Republic* and other dialogues — Plato\n- *Organon* / *Nicomachean Ethics* — Aristotle\n- *Meditations on First Philosophy* — Descartes\n- *Critique of Pure Reason* — Kant\n- *Philosophical Investigations* — Wittgenstein\n- Stanford Encyclopedia of Philosophy — plato.stanford.edu","html":"