PART G — GREEK VS LATIN: TELLING THE FAMILY APART BY SPELLING

English vocabulary is really two overlapping vocabularies wearing one alphabet: a Latin-derived layer (often arriving via French) and a Greek-derived layer (often arriving via scientific and philosophical writing). You do not need to know etymology as a subject to use this — you only need to recognise a handful of spelling signatures, because each family tends to combine its own roots with its own affixes, and knowing which family a word belongs to narrows down what its pieces are likely to mean.

Signatures that usually mean Greek origin:

• ph for the "f" sound → philosophy, phobia, graph, phase, morphology
• ch pronounced as "k" → chronology, chaos, character, chorus, architect
• y in the middle of a word (not at the start/end) → physics, gymnasium, symbol, mythology, hypothesis
• ps or pn at the start of a word → psychology, pseudonym, pneumonia
• Combining forms joined with a connecting -o- → thermometer, chronograph, biology, philology
• Compound-feeling words built from two full concept-roots stuck together → telephone (far + sound), photograph (light + write), democracy (people + rule)

Signatures that usually mean Latin origin:

• -tion/-sion, -ance/-ence, -ity, -ous, -ive endings → creation, tolerance, civility, gracious, decisive
• Prefixes like in-/im-/il-/ir-, ex-, sub-, trans-, ad- attached to a verb-like root → inaccurate, exclaim, submerge, transmit, adhere
• Words that feel like they could plausibly appear in French or Spanish with only minor spelling changes → nation/nación, action/acción, animal/animal
• Roots that are recognisable Latin verb-stems (dict, duc, scrib, spec, vert, ject) rather than Greek concept-nouns (phon, phil, path, log, chron)

Why this matters in the exam hall: the two families rarely mix their affixes with each other's roots — you will almost never see a genuine Greek root take a Latin -tion suffix, or a Latin root take a Greek -ology suffix, in standard formal vocabulary. So if you can identify the family from the spelling, you can often predict which kind of suffix is even grammatically likely to appear on that word, which narrows an "insert the correct form" question dramatically. It also explains apparent irregularities: biography uses graph (Greek, "to write") not scrib (Latin, "to write") — both roots mean the same thing, but they never substitute for each other, because English inherited two entire parallel vocabularies for "writing," one from each donor language, and kept them separate. The same doubling happens for "hearing" (aud-Latin vs phon-Greek: audible vs telephone), for "people" (popul-Latin vs dem-Greek: population vs democracy), and for "love" (am-Latin vs phil-Greek: amiable vs philosophy). Recognising a doublet like this instantly doubles your working vocabulary for that concept, because now two roots, not one, unlock the same idea.

A working table of Latin/Greek doublets. English frequently kept both the Latin and the Greek word for the same underlying concept, and uses them in slightly different registers — the Latin version often feels more everyday, the Greek version more technical or scientific. Recognising a doublet lets one root-family unlock the other.

Once you notice this pattern, you stop being surprised that English has two words that feel like they should mean the same thing but sit in different registers — audible (Latin, ordinary) versus a hypothetical Greek coinage using phon- for the same idea, which instead surfaces in telephone, phonetic, symphony. The two families didn't compete so much as divide the territory: Latin supplied everyday adjectives and legal/administrative vocabulary; Greek supplied the vocabulary of science, medicine, and philosophy. That division is itself a clue — if a word sounds like it belongs in a science textbook, bet on Greek; if it sounds like it belongs in a court document or an everyday formal letter, bet on Latin.