Nominative-Accusative Alignment

The two-case picture

An English speaker already knows nominative-accusative alignment intimately. Compare:

• She sleeps. — intransitive; she is the subject (S).
• She sees him. — transitive; she is the agent (A); him is the patient (P).
• He sees her. — transitive; he is the agent (A); her is the patient (P).

Notice that the form she is used in both the intransitive (she sleeps) and the transitive agent (she sees him) slots. That's the nominative — the case that groups S and A. The form her shows up only as the transitive patient. That's the accusative.

If a language instead grouped her with she in she sleeps — i.e. used the same form for the intransitive subject as for the transitive object — it would be ergative, not accusative. About one in six languages does that. The other roughly half use accusative-style grouping. The rest divide along less common patterns.

The English pronoun paradigm

English is a striking case because the accusative pattern survives only on a closed set of pronouns. Here is the full inventory:

Person/Number	Nominative	Accusative	Genitive (possessive)
1SG	I	me	my, mine
2SG/PL	you	you	your, yours
3SG.M	he	him	his
3SG.F	she	her	her, hers
3SG.N	it	it	its
1PL	we	us	our, ours
3PL	they	them	their, theirs
WH	who	whom	whose

The accusative collapse is uneven. You and it dropped the distinction entirely; whom is rapidly being levelled by who in spoken English. Meanwhile the genitive — historically a third case, dative having merged into accusative around 1200 — survives as a separate column. By the late 22nd century English may well have only two surviving pronoun cases.

Languages that mark every noun

Outside English, full nouns commonly inflect for nominative and accusative. Examples:

• Latin: puer puell-am videt ("the boy sees the girl"). Puer is nominative (no suffix in this declension); puellam takes accusative -am. Reverse it: puella puer-um videt ("the girl sees the boy"), with feminine nominative -a and masculine accusative -um.
• Russian: Ivan vidit Mariyu ("Ivan sees Maria"); Ivana vidit Maria ("Maria sees Ivan"). The -a and -u suffixes carry the alignment; word order can be permuted freely.
• German: case is partially marked, primarily on articles and adjectives. Der Hund beißt den Mann ("the dog bites the man"); den Hund beißt der Mann ("the man bites the dog") — the article switches from der (NOM) to den (ACC).
• Japanese: particle-based marking. Watashi-ga sushi-o tabeta ("I-NOM sushi-ACC ate"). The particles -ga and -o are syntactically equivalent to case suffixes.
• Turkish: agglutinative. Adam kitab-ı okudu ("the man read the book"). The accusative -ı appears only on definite/specific objects, an instance of differential object marking.
• Mandarin: no case morphology at all. Word order alone disambiguates: wǒ kàn tā ("I see him") vs tā kàn wǒ ("he sees me"). Pronouns don't change shape.

Where nominative-accusative sits

Alignment	S grouped with	Marking on	Example language	Sample sentence	Approx. share
Nominative-accusative (full)	A	every noun	Latin / Russian	puer puellam videt	~25% of all languages
Nominative-accusative (pronoun-only)	A	pronouns only	English	he sees her	~25% of all languages
Ergative-absolutive	P	nouns and/or verbs	Basque / Inuktitut	gizonak emakumea ikusi du	~16–25%
Split ergative	changes by tense or person	partial	Hindi-Urdu	laṛke ne kitāb paṛhī	large subset of "ergative" group
Active-stative	S split by volitionality	verbs/pronouns	Lakota	wa-ʔú ("I come") vs ma-tʔá ("I die")	small minority
Tripartite	none — S, A, P all distinct	nouns	Nez Perce	3 distinct case forms	very rare
Direct (neutral)	none — no marking at all	nothing	Mandarin / Vietnamese	wǒ kàn tā	~30%, mostly isolating

The trade-off with word order

The more case marking a language has, the freer its word order tends to be. Nominative-accusative languages illustrate this clearly:

Latin (full case):       any of 6 word orders, all = "the boy sees the girl"
   puer puellam videt    /    puer videt puellam
   puellam puer videt    /    puellam videt puer
   videt puer puellam    /    videt puellam puer

Russian (full case):     scrambling driven by pragmatics, not grammar

German (partial case):   limited scrambling; verb-second main clauses

English (no nominal case): essentially fixed SVO; scrambling only with pronouns
   "Him I saw."           (rare, marked, archaic-feeling)

Mandarin (no case):      strictly SVO; permutation changes meaning

The reason is mechanical. If both arguments are unmarked, only word order can tell who did what to whom — so the order has to be locked down. If both are marked, the listener can recover roles from the morphology and the order is freed up to encode information structure (topic, focus, contrast) instead. English has been drifting toward the Mandarin pole for nearly a millennium.

Differential object marking

Many nominative-accusative languages don't mark every object the same way. They mark only the objects that are most likely to be confused for subjects — typically animate, definite, or specific ones. This is differential object marking (DOM), and it shows up across families:

• Spanish: the preposition a precedes human or specific direct objects. Veo a Juan ("I see Juan") but veo la película ("I see the movie"). It's the same accusative slot — the marking depends on who fills it.
• Turkish: the accusative suffix -(y)ı/i/u/ü appears only on specific objects. Bir kitap okudum ("I read a book", non-specific) vs kitabı okudum ("I read the book", specific).
• Hindi-Urdu: -ko marks animate or specific objects. Maine bachhe-ko dekhā ("I saw the child") vs maine kitāb dekhī ("I saw a book").
• Romanian: the preposition pe marks human direct objects, optionally for inanimates.

The pattern recurs because the function is the same: differentiate object from subject most strongly when the object is animate enough to plausibly be the subject.

How English became case-poor

Old English, before 1100, looked very much like modern German. Nouns inflected for four cases (nom/acc/gen/dat), three genders (m/f/n), and two numbers (sg/pl). Adjectives agreed in all three feature sets. The system collapsed across the 12th–14th centuries:

1. Phonological erosion. Unstressed final syllables — exactly where case markers lived — eroded systematically. -an, -um, -es, -e all eventually merged toward schwa and then silence.
2. Norse contact. The Danelaw region (eastern and northern England) had centuries of intense Old English / Old Norse bilingualism. The two languages shared vocabulary and inflection, but different specific endings — speakers may have simplified inflection to ease cross-comprehension.
3. French superstrate. After 1066, French had no productive case on full nouns and contributed no pressure to retain it.
4. Word-order rigidification. As inflection eroded, SVO order tightened to compensate — the cause and effect ran in both directions.

By Chaucer (late 1300s) the case system was already mostly gone for full nouns. The pronouns and the genitive 's are the survivors — they were either too frequent to erode (pronouns) or sufficiently distinctive (the genitive often had a clear consonant, not just a vowel).

Common pitfalls

• Confusing "subject" with "agent". The subject is the syntactic role; the agent is the semantic role. In "the door opened", the door is the syntactic subject (and nominative if marked), but it is not an agent. Alignment cares about syntactic roles.
• Reading dative as accusative. English collapsed the two — him is used for both direct objects (I see him) and indirect ones (I gave him the book). German, Russian, and Latin keep them distinct, and learners must too.
• Assuming SVO order is universal. SVO is dominant in nominative-accusative languages worldwide, but SOV (Japanese, Turkish, Korean, Hindi) is just as common a pattern within the alignment.
• Misreading prepositional accusatives. Spanish "a" looks like a preposition but is the accusative marker for human/specific objects, not a separate case. Beginners sometimes interpret it as English to.
• Equating English's word order with grammatical poverty. English compensates rich morphology with rigid syntax. The total information bandwidth is comparable; the storage strategy differs.

Why nominative-accusative alignment matters

• Default for typology. The most common alignment, baseline for cross-linguistic comparison.
• Grammar pedagogy. Teaching most major languages requires explicit attention to subject vs object marking.
• Historical linguistics. Drift in or out of accusative alignment marks deep changes in a language's grammar.
• NLP. Parsing pipelines for English, Russian, Turkish and the Romance languages all rely on the alignment as a structural backbone.
• Theoretical syntax. Case theory developed in the 1980s and 1990s was built on accusative-language data; ergative facts later forced significant revision.
• Translation and L2 learning. Mapping English's pronoun-only system to a fully marked language (or vice versa) is the first hurdle every learner crosses.