Morphology
Inflection vs Derivation
Two morphological types — grammatical marking vs new word creation
Morphology splits into two operations: inflection adjusts a word's form to fit grammatical context without changing its category or core meaning ("walk" → "walks," "walked," "walking"), while derivation creates new words, often changing category or core meaning ("walk" → "walker," "walkable," "sidewalk"). Inflection produces forms of the same lexeme; derivation produces new lexemes. English plural -s, past -ed, comparative -er are inflections; agentive -er, adjectival -able, abstract -ness are derivations. The distinction goes back to Pāṇini (~5th c. BCE) and was formalized for Western linguistics by Bloomfield (1933) and refined by Anderson (1992) in his book "A-Morphous Morphology." Diagnostic tests include productivity, semantic regularity, and whether the affix closes off further derivation.
- InflectionSame lexeme, different grammatical form
- DerivationNew lexeme, often new category
- English plural -sInflectional
- English -nessDerivational (cat → cat-N)
- ProductivityInflection ~100%; derivation varies
- PositionDerivation closer to root than inflection
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
Why the inflection-derivation distinction matters
- Lexicography. Dictionaries list lexemes, not inflected forms.
- Morphological theory. Stratal models depend on the split.
- NLP. Lemmatization removes inflection; stemming targets both.
- L2 acquisition. Inflection is harder for late learners; derivation more transparent.
- Cross-linguistic comparison. Languages differ wildly in derivation richness.
- Productivity studies. Asymmetric: inflection productive, derivation patchy.
- Diachrony. Derivational affixes can grammaticalize into inflectional ones.
Common misconceptions
- The distinction is binary. It's a cluster of properties; some affixes are mixed.
- Suffixes are derivation; prefixes are inflection. Both can be either.
- Derivation always changes category. Many derivational affixes preserve it (un-, -hood).
- Inflection is "weaker" morphology. Inflection is grammatically obligatory; derivation isn't.
- English has poor inflection. Verbs still inflect for tense, person, number, aspect.
- Pluralization is derivation. Pluralization marks number — pure inflection.
Frequently asked questions
How can I tell them apart?
Several diagnostics. (1) Category change: derivation often changes part of speech (verb → noun); inflection rarely does. (2) Semantic regularity: inflection is fully predictable (-s = plural every time); derivation can be idiosyncratic ("transmission" ≠ just "transmitting"). (3) Productivity: inflection applies to (almost) every word in a class; derivation is selective ("happy" → "happiness" but "*sleepiness" feels off — actually fine; try "*beauthood"). (4) Position: derivational affixes attach closer to the root than inflectional ones ("nation-al-iz-ation-s").
Is the distinction always clean?
No. Stephen Anderson and others argue it's a continuum. Russian aspectual prefixes (na-, pere-, vy-) sit between: they create new lexemes (derivation-like) but are grammatically obligatory (inflection-like). English -ing is similarly fuzzy: "running" can be a verb form (inflection) or a noun ("the running of the bulls" — derivation). Many morphologists posit a third class: "inherent inflection" (number, gender) vs "contextual inflection" (case, agreement).
What are inflectional categories?
Tense (past/present/future), aspect (perfective/imperfective), mood (indicative/subjunctive/imperative), person (1/2/3), number (singular/plural/dual), gender (masculine/feminine/neuter), case (nominative/accusative/etc.), voice (active/passive). These are typically marked on verbs (TAM) or nouns (gender, number, case). Adjectives often inflect for agreement.
What are derivational categories?
Less systematic. Common types: nominalization (verb → noun: "decide" → "decision"), verbalization ("class" → "classify"), adjectivalization ("danger" → "dangerous"), agentive (-er, -ist), abstract (-ness, -ity), negation (un-, in-), causative ("flat" → "flatten"). Many languages have far richer derivation than English.
Why does position matter?
A morphological universal: derivational affixes occur closer to the root than inflectional ones. "Nation-al-iz-ation-s": -al (deriv adj), -iz (deriv verb), -ation (deriv noun), -s (infl plural). You can't say "*nation-s-al." This is the basis of stratal theories (Lexical Phonology) — derivation happens at an earlier stratum than inflection.
Are paradigms inflectional only?
Traditionally yes. A paradigm is the full set of inflected forms of a lexeme — "walk, walks, walked, walking" or Latin "amo, amas, amat..." Derivation produces separate lexemes, not paradigm members. But families of derivationally related words ("nation, national, nationality, nationalism, nationalist, nationalize") are sometimes called derivational paradigms or word families.
How do children acquire each?
Inflection is mastered early — English-speaking children learn plural -s and past -ed by age 3-4, with the famous "comed/goed" overgeneralization phase. Derivation continues developing into adulthood as vocabulary grows. Adult speakers actively coin new derived words ("Googleable," "selfie-ish") much more than new inflected forms.