root

Pinned sense (Law 11d): root as the core lexical unit of a word, free or bound. INSTANCE_OF test (Law 9): every root is a morpheme — a free morpheme when it stands alone as a word (walk in walks, walker), a bound morpheme when it cannot (struct- in construct, destruction) — and morpheme is the NEAREST accepted kind: the board's only accepted kind under morpheme is affix, and a root is not an affix (an affix attaches to a host stem to modify it; a root is the host material that affixes attach to, the residue that remains when they are removed). The morpheme definition's own type list (free morphemes, bound morphemes, allomorphs) covers every case of root. Specific-to-general: roots are the meaning-bearing core of the morpheme inventory, not the inventory. (Laws 9, 11d, 11e.)

Pinned sense (Law 11d): both as technical categories of morphological analysis. DERIVED_FROM test (Law 7) — which came first, from independent historical reasoning, not the graph's epoch: the ROOT concept is the older category, attested in the oldest grammatical traditions (Sanskrit dhatu 'verbal root' in Panini, 4th c. BCE; Semitic triconsonantal roots as the classic paradigm; 'etymological root' in Indo-European historical linguistics). The technical STEM category — the unit to which inflection attaches, distinct from the etymological core — crystallized in 19th-century comparative morphology, when the older root/base idea was split into Wurzel (etymological root) and Stamm (inflectional stem). The stem concept was carved out of the root concept by peeling off derivational material; the reverse history is not available. Ordering rests on the history of grammatical literature, not on the graph's epoch estimate (Law 7 anti-circularity).