The Anatomy of Dialogue: Tarantino vs. Sorkin
What happens when you feed two of cinema's greatest screenplays to a parser and measure everything.
Both writers are famous for dialogue. But their approaches are fundamentally different. We parsed every line of Pulp Fiction (Tarantino, 1994) and The Social Network (Sorkin, 2010) to find out exactly how.
Section 1: The Tale of the Tape
| Metric | Pulp Fiction | The Social Network |
|---|---|---|
| Pages | 126 | 164 |
| Dialogue blocks | 1,192 | 1,677 |
| Characters | 52 | ~25 |
| Total words | ~26,000 | ~26,700 |
Nearly identical word counts, radically different structures. Same raw material, opposite architectures.
Section 2: Who's Talking?
Dialogue vs. action line distribution.
Both scripts are dialogue-dominant, but Tarantino edges ahead. His action lines are sparse and functional — stage directions to get characters from one speech to the next. Sorkin gives more room to scene description, environment, character movement.
Section 3: The Monologue Gap
Average words per speech — the biggest finding.
This is where the fundamental difference lives. Tarantino lets his characters ramble. He gives them room to digress, to tell stories within stories, to build monologues that function like jazz solos. Sorkin's characters speak in bursts — rapid, overlapping volleys where no one holds the floor for long.
JulesThere's a passage I got memorized. Ezekiel 25:17. "The path of the righteous man is beset on all sides by the inequities of the selfish and the tyranny of evil men. Blessed is he who, in the name of charity and good will, shepherds the weak through the valley of darkness, for he is truly his brother's keeper and the finder of lost children. And I will strike down upon thee with great vengeance and furious anger those who attempt to poison and destroy my brothers. And you will know my name is the Lord when I lay my vengeance upon thee."
Contrast that with Sorkin's rapid volley — six speakers in the same space, no one holding the floor:
Mark ZuckerbergHow do you distinguish yourself in a population of people who all got 1600 on their SAT's?
EricaI didn't know they take SAT's in China.
Mark ZuckerbergThey don't. I wasn't talking about China anymore, I was talking about me.
Section 4: The Vocabulary
Word richness and diversity.
| Metric | Pulp Fiction | The Social Network |
|---|---|---|
| Unique words | 3,130 | 3,005 |
| Hapax legomena | 1,614 | 1,625 |
| Yule's K | 71.12 | 78.87 |
Despite Sorkin's reputation as the more "literary" writer, Tarantino's dialogue has measurably greater vocabulary diversity. His characters draw from a wider word pool. This makes sense — Tarantino's characters span social classes, ethnicities, and subcultures (hitmen, boxers, drug dealers, crime bosses). Sorkin's characters in The Social Network largely occupy one world: Harvard-educated tech entrepreneurs. Similar vocabularies, similar reference pools.
Section 5: Talk Speed
Sentence structure and conversational rhythm.
Sentence-by-sentence, they're surprisingly similar — both averaging around 7 words, both with a median of 6. The difference isn't sentence length, it's how sentences combine. Tarantino strings many sentences into long unbroken speeches. Sorkin distributes short sentences across rapid character alternation. Same bricks, different buildings.
Sorkin asks more questions (519 vs. 451) — his dialogue is adversarial, deposition-like, characters constantly challenging each other.
Section 6: The F-Word and the Dropped G
Voice, register, and how people actually talk.
Tarantino's characters swear 7.7 times more per word than Sorkin's. In the time it takes Sorkin's characters to drop one F-bomb, Tarantino's have dropped 7.7.
This is the register gap. Tarantino writes speech as people actually speak — messy, contracted, full of slang and profanity. "Fuckin'" appears 90 times in Pulp Fiction. The dropped G alone (doin', goin', talkin') appears 61 times. Sorkin writes speech as people wish they spoke — articulate, complete, precise. Zero dropped G's in The Social Network. Not one.
Section 7: Who Dominates?
Character concentration and ensemble dynamics.
Pulp Fiction — Top 5 by dialogue words
The Social Network — Top 5 by dialogue words
Sorkin's script is dominated by Mark — he speaks 25% of all dialogue words. Tarantino distributes more evenly. Jules and Vincent together account for a similar proportion, but no single voice dominates. Tarantino's ensemble approach vs. Sorkin's protagonist focus.
One standout: The Wolf has the highest avg words/speech at 25.5 — when Harvey Keitel's character talks, he talks.
Section 8: Interruption Rate
Conversational style: collaborative vs. adversarial.
Sorkin's characters interrupt each other nearly 70% more often. His dialogue is competitive — characters talk over each other, finish each other's sentences, redirect constantly. Tarantino's characters are more generous listeners. They let each other talk. Even when they disagree, they wait.
The interruption rate perfectly captures the difference: Tarantino conversations are collaborative; Sorkin conversations are adversarial.
Section 9: The Signature
Top recurring content words in dialogue.
Pulp Fiction
The Social Network
The word "fuckin'" is the 7th most common non-stopword in Pulp Fiction — so frequent it functions as punctuation. The word "yes" appears in The Social Network's top 15 — fitting for a script about a character who can never quite agree with anyone.
What the Numbers Miss
Data can measure rhythm, vocabulary, and structure. It can't measure what makes a line land — the pause before Jules quotes Ezekiel, the way Mark's cruelty curls around a compliment. Both scripts won their writers an Oscar nomination. Both are taught in film schools. The numbers show us how differently you can build a great screenplay. The experience of reading them shows us why both approaches work.
A data essay. All statistics derived from screenplay analysis of Pulp Fiction (1994) and The Social Network (2010).