Again，toseeiftheindignantmanisjealous，askwhoeachofthemis：forthenitwillbeobviouswhetherthestatementistrueorfalse；e。g。ifheis’jealous’whogrievesatthesuccessesofthegood，andheis’indignant’whogrievesatthesuccessesoftheevil，thenclearlytheindignantmanwouldnotbejealous。Amanshouldsubstitutedefinitionsalsoforthetermscontainedinhisdefinitions，andnotstopuntilhecomestoafamiliarterm：foroftenifthedefinitionberenderedwhole，thepointatissueisnotclearedup，whereasifforoneofthetermsusedinthedefinitionadefinitionbestated，itbecomesobvious。
Moreover，amanshouldmaketheproblemintoapropositionforhimself，andthenbringanegativeinstanceagainstit：forthenegativeinstancewillbeagroundofattackupontheassertion。
Thisruleisverynearlythesameastheruletolookintocaseswhereapredicatehasbeenattributedordenieduniversally：butitdiffersintheturnoftheargument。
Moreover，youshoulddefinewhatkindofthingsshouldbecalledasmostmencallthem，andwhatshouldnot。Forthisisusefulbothforestablishingandforoverthrowingaview：e。g。youshouldsaythatweoughttouseourtermstomeanthesamethingsasmostpeoplemeanbythem，butwhenweaskwhatkindofthingsareorarenotofsuchandsuchakind，weshouldnotheregowiththemultitude：e。g。
itisrighttocall’healthy’whatevertendstoproducehealth，asdomostmen：butinsayingwhethertheobjectbeforeustendstoproducehealthornot，weshouldadoptthelanguagenolongerofthemultitudebutofthedoctor。