The Age of the Essay — Paul Graham

School forces you to write essays with a rigid structure: topic sentence, introductory paragraph, supporting paragraphs, conclusion. The conclusion might be that Ahab in Moby Dick was a Christ-like figure.

Oy.

Real essays are different. Here's the other side of the story.

Mods

The most obvious difference: real essays are not exclusively about English literature. Schools should teach writing, but due to historical accidents it got mixed with the study of literature. Students write about symbolism in Dickens instead of how a small-budget baseball team might compete with the Yankees, the role of color in fashion, or what makes a good dessert.

Result: writing seems boring and pointless. Who cares about symbolism in Dickens? Dickens would have been more interested in an essay about color or baseball.

How did this happen? Go back almost a thousand years. Around 1100, Europe caught its breath after chaos and rediscovered "the classics." Earlier civilizations were so sophisticated that for centuries European scholars' main work was assimilation.

Classical texts acquired great prestige. By 1350 better science teachers existed than Aristotle, but schools change slower. In the 19th century ancient texts were still the curriculum backbone.

The question arose: if ancient texts are worthy of scholarship, why not modern ones? The archaeological work on classics was mostly done; it implied classics scholars were working on minor problems. But no one wanted to say that.

And so modern literature study began. Resistance at first. Newer colleges (Dartmouth, U Vermont, Amherst, University College London) offered English lit in 1820s. Harvard 1876, Oxford 1885 (Oxford had Chinese chair first).

Professors should do research. Imported from Germany (Johns Hopkins 1876 onward). Writing teachers transformed into English professors. Francis James Child (Boylston Professor of Rhetoric since 1851) became Harvard's first English professor in 1876.

Writing was casualty. Closest to research on composition seemed English literature. (a) Lit expert need not be good writer (art historian not necessarily painter). (b) Subject of writing tends to lit.

High schools imitate. 1892 NEA recommended literature and composition unified. 'Riting morphed into English. Students write imitations of what English profs published decades earlier. Three steps removed from real work.

No wonder it seems pointless.

No Defense

Another difference: a real essay doesn't take a position and defend it.

This principle comes from long-forgotten origins in law schools (medieval universities more law than seminaries). Lawyers trained to argue either side. Rhetoric was 1/3 of trivium. Disputation common.

But defending a position isn't the best way to get at truth. You miss subtleties. You can't change the question.

Yet this is built into school essay structure: topic sentence = thesis chosen in advance, supporting paragraphs = blows, conclusion = restate (concluding remarks to jury).

Good writing is convincing because right answers, not clever arguing. When giving drafts to friends: which parts bore, which unconvincing? Boring: cut. Unconvincing: talk it over, come up with clearer explanation (or change what you were saying).

At least explain badly. More often, change the claim. Aim is truth, not convincing per se. Convincing smart readers ≈ near truth.

Defend-a-position may be valid/inevitable, but historically inaccurate to call it "essay." An essay is something else.

Trying

Reach back to Michel de Montaigne, 1580, "essais." Essayer = French "to try." An essay is an attempt.

You don't know what you're figuring out yet. Can't begin with thesis (may never have one). Essay starts with question, not statement. Notice door ajar, open and walk in.

Why write at all? Why not just think? Montaigne's discovery: expressing ideas helps form them. Most of what ends in essays only thought of when sitting to write.

Not quite just for yourself. Inviting people over forces cleaning apartment; writing for readers forces thinking well. Audience matters. Things written just for self are no good; peter out. Run into difficulties, vague questions, drift to tea.

Published essays (especially newsmagazine staff) often peter out too. Outside writers do defend-a-position (beeline to foreordained conclusion). Staff writers "balanced": start with radioactively controversial, recoil in terror. "The question is complex. (But don't get mad at us.)"

The River

Questions aren't enough. Essay has to come up with answers (interesting ones). Not always (inconclusive experiments not published). Tells reader something new.

What you tell doesn't matter so long as interesting. Accused of meandering: flaw in defend-a-position (you know where going, bluster through). Not in essay: search for truth. Suspicious if it didn't meander.

The Meander (Menderes) river in Turkey winds, but not frivolity. Most economical route to sea. Algorithm: at each step, flow down.

For essayist: flow interesting. Of places to go next, choose most interesting. Not quite as little foresight as river. Know generally what writing about, but not specific conclusions. Paragraph to paragraph let ideas take course.

Doesn't always work. Hit wall: backtrack. (Example in this essay: after following thread, ran out of ideas, went back 7 paragraphs, started over.)

Fundamentally essay is cleaned-up train of thought (like cleaned-up conversation). Real thought full of false starts; exhausting to read. Cut and fill to emphasize central thread (illustrator inking pencil drawing). Don't change so much you lose spontaneity.

Err on side of river. Not reference work. Not read for specific answer (feel cheated if not). Much rather essay that went unexpected but interesting direction than plodded prescribed course.

Surprise

What's interesting? For me: surprise. Interfaces follow least astonishment (button that looks like stop should stop, not speed up). Essays opposite: maximum surprise.

Afraid of flying, traveled vicariously. Asked friends what surprised them about faraway places. Extremely useful question. Extracts info they didn't know they recorded. Surprises = didn't know + contradict what you thought you knew. Most valuable facts. Counteract unhealthy effects of prior knowledge.

How find surprises? Half the work of essay writing (other half: express well). Use yourself as proxy for reader. Write about things thought about a lot. Anything surprising you (who thought a lot) will probably surprise most readers.

Example: in essay on great hackers, pointed out you can only judge programmers by working with them, so no one knows best overall. Didn't realize when began; still kind of weird. That's what looking for.

Two ingredients for essays: few topics thought about a lot + ability to ferret unexpected.

What think about? Anything can be interesting if get deeply enough. (Exception: things variation sucked out, like fast food. But even there: color important to kids (yellow), Pralines 'n' Cream appeal (salt?), fathers vs mothers buying ice cream. At 16 was lump of rock; now see more in memory fragments.)

Observation

Ability to ferret surprises not purely inborn; can learn. Habit of asking questions (esp. Why). Not random like 3yo. Infinite questions; find fruitful ones.

Useful: ask why about things that seem wrong. (Humor and misfortune? Slipping on banana peel? Whole essay there.)

Skepticism helps (achieving only 1% of what could). Counteracts "things are the way they are because have to be."

Surprises make us laugh. Surprises = what want to deliver.

Habit of paying attention to "inappropriate," not important, or not assigned. If curious, trust instincts. Follow threads. (Comb-over example: how doesn't he see how odd? Incremental over 20 years. Gradualness powerful — can use constructively. Start stripped kernel, grows to OS. Same in painting/novel?)

Study lots of different things (unexpected connections: jam/bacon/pickles/cheese as preservation; same for books/paintings).

Include history — social and economic, not just political. All the data we have so far.

Swords: evolved from daggers with separate hilt. Longer swords, hilts broke. Took 500 years before casting hilt and blade as one.

Disobedience

Above all: pay attention to things not supposed to. Follow what attracts. Study history.

Disobedience: write essays at all. Fortunately, rampant now. Web changing: anyone publishes, judged by what said, not who. (Previously only approved writers; magazines judged by who, not what.)

Popular magazines made golden age of short story (literacy to TV). Web may make golden age of essay.

(Not something realized when starting this.)