Pulitzer Prize Winner Teaches Writing (Paul Harding Interview)

David Perell

Abstract

Nearly ninety minutes with a novelist whose teaching starts by refusing to teach a method. What gets taught instead is the self-awareness to observe yourself as a writer — because making what works for you normative is the danger, and observation is the one thing underneath everything else that genuinely transfers.

• Received, off-the-rack language does not work, and you cannot enter a sentence with a repertoire you planned to deploy.
• A brain is like a fingerprint: readers connect because the work is almost entirely the same as everyone else's, and the tiny fraction that is not belongs only to you.
• If a reader gets the point, they never have to think about it again — which is the argument against lessons in the work.
• Fiction persuades by recognition rather than by argument: that is true, I have just never seen it in words.
• You cannot think your way to being better; you have to get it wrong in a particular way to get it right.
• Precision fills a description with wonder rather than making it clinical.

Everything below follows the conversation in order, from the danger of teaching through observation and literalness, into ambition, authority and identity, out through vocabulary, plot, density and what not to put in the work, and finally into revision, rhythm and reading. For anyone building a personal brand the most transferable claim is the fingerprint one: the differentiator is a fraction of a per cent, and it is the only part nobody else can supply.

Chapter summaries

00:48 - The danger of teaching is making one writer's method normative

The danger of teaching is making what works for you normative; teach the self-awareness to observe yourself as a writer, not your own method.

"That's the perfect question because one of the dangers of being a teacher is"

— 00:48

Teaching your own method risks making it normative for everybody. What gets taught instead is the self-awareness to observe yourself as a writer — a transferable capacity rather than a transferable technique.

01:30 - At root writing is aesthetics and taste, and the method is improvisation

At root writing is aesthetics and taste; the method is improvisation — going in without a map and seeing what the language and the experience give you.

"you because ultimately at its base, it's it's aesthetics. It's my taste. I love"

— 01:30

A system of enough complexity, once set in motion, produces properties that could not have been predicted beforehand — emergence, borrowed from physics. The image offered is a silent-film comedian stepping onto open space and hoping the girder is there.

02:42 - The teachable skill under all of it is observation

The teachable skill under all of it is observation — you must step back and treat paying attention as a skill you work at.

"mean kind of at the root what I teach you know literal and concrete"

— 02:42

Stepping back and treating paying attention as a skill you work at. It is the one thing underneath everything else that can genuinely be taught.

03:16 - Received, off-the-rack language will not work

Received, acculturated, off-the-rack language will not work in good writing; you cannot enter a sentence with a repertoire you plan to deploy.

"um all use of uh received acculturated or habituated language because a lot of"

— 03:16

You cannot enter a sentence with a repertoire you plan to deploy. The plan is the problem rather than the vocabulary.

04:26 - Reaching raw sensation means observing pre-linguistically

To reach raw sensation you have to sift down and observe pre-linguistically, which means slowing down.

"those raw sensations? Yeah. You have to sift you have to sift down a"

— 04:26

Sifting down to observe pre-linguistically, which means slowing down. The sensation exists before the words do, and the words arrive too early by default. Slowing down is the only way to get underneath the words that arrive automatically.

05:20 - An image gets built by staying utterly literal and testing it forward

Build an image by staying utterly literal and testing it forward — "what happens if I stick with this?" — until you reach the end point.

"like the opening of um of my first novel Tinkers. Um, it's about a"

— 05:20

Pursuing the literal detail, rather than rushing to it's symbolic, is what lets an extended metaphor emerge on its own. A dying man's hallucinated cracks in a ceiling get followed step by step until the sky drains to a funeral shroud — a metaphor that arrived rather than one that was planned, meaning without the writer knowing the meaning.

08:07 - A passage can take years, so doggedness is the requirement

Any given passage can take years; you have to be dogged, patient and unhurried, and learn to sustain attention once you find something.

"Yeah. Yeah. And sometimes it it takes like that. I remember it's like pulling"

— 08:07

Doggedness, patience and the ability to sustain attention once something is found are the requirements, which makes the temperament part of the craft.

09:40 - The language cannot be handed over, only flooded

You cannot hand someone the language; you flood them with many possibilities and let them keep whatever sticks and build their own repertoire.

"mean, so this is another thing to go back to teaching. It has to"

— 09:40

You cannot hand somebody the language. You flood them with possibilities and let them keep whatever sticks and build their own repertoire — and thinking like a painter, in horizontal layers of colour and light, is one of those possibilities.

11:57 - Interior and exterior stop being separate gears

Stop switching gears between interior and exterior; treat them as co-extensive so description of the world becomes the rendering of character.

"housekeeping is co-extensivity I just thought what if I just don't think of there"

— 11:57

Describing the world becomes the rendering of character, which is one move that removes an entire category of transition. The transitions were never carrying meaning anyway; they were the seams left by treating the two as separate jobs.

12:52 - Insecure writers show off for other writers

Insecure writers show off for the applause of other writers instead of writing in service of the reader.

"can write like paintings. Well, yeah. A lot of insecure writers, they show off,"

— 12:52

When the art is about the artist's cleverness, the reader has stopped being served. The finished novel that leaves you thinking wasn't he the cleverest boy in class is the failure; a drummer gets called a mere virtuoso for the same reason. Even show-off virtuosity is a legitimate aesthetic choice some readers love.

14:08 - People get seeds and choices rather than genre rules

Give people seeds and choices, not genre rules; narrative prose works the same ways, so any piece can be made better on its own terms.

"for my students. you figure out what you think is cool, what interests you."

— 14:08

Seeds and choices rather than genre rules. Narrative prose works the same ways, so any piece can be made better on its own terms rather than against a template.

15:21 - The one message almost always relevant is to slow down

The one message, relevant almost 100% of the time, is: slow down; patience is a skill you have to learn to be a good artist.

"message? Two word. They're going to be on my They're going to be on"

— 15:21

Slow down — relevant almost all of the time, and the patience it requires is a skill that has to be learned rather than a trait somebody has.

16:32 - Nobody can write about anything but what they truly care about

You cannot not write about what you truly care about; forcing a cause produces propaganda, so surrender to your own intrinsic grain.

"patient that that's a leap of faith which is you can't not write about"

— 16:32

Forcing a cause produces propaganda, so the move is surrendering to your own intrinsic grain. You cannot not write about what you truly care about, which makes the choosing mostly a matter of stopping the resistance.

17:24 - A brain is like a fingerprint

Your brain is like a fingerprint; readers connect because the work is 99.99% the same as others and the fraction that is unique is only yours.

"need to surrender to. Yeah. Your brain is like your fingerprint, right? Like one"

— 17:24

Readers connect because the work is almost entirely the same as everybody else's, and the tiny fraction that is unique belongs only to you. The fingerprint is the fraction — and setting aside acclaim before descending into the next manuscript is its own learned discipline.

23:38 - The work is written for brilliant, big-hearted readers

Do not write for bad readers; write for brilliant, big-hearted people, and treat richness of meaning as an act of generosity.

"students? Don't write your books for bad readers. Write them for brilliant, big-rained and"

— 23:38

Richness of meaning is an act of generosity rather than a barrier. Writing down is the insult that gets mistaken for accessibility.

25:05 - Propaganda is spent on arrival where art keeps growing

Propaganda gives everything at the beginning and is spent once you get it; art grows on you and rewards being returned to.

"Well, that's one way to think about the difference between propaganda and art. Propaganda"

— 25:05

Propaganda gives everything at the beginning and is spent the moment you get it. Art grows on you and rewards being returned to, which is a definition of the difference rather than a judgement about it.

28:26 - Writing can only be as good as the best that has been read

Your writing can only be as good as the best you have read — and as good as the best readings you can give it.

"ambition was to be a good reader. And I think that kind of is"

— 28:26

It is also only as good as the best readings it can be given — two ceilings, both of them raised by the same activity, which is why reading is not preparation for the work but part of it.

30:05 - Self-consciousness has to evolve into self-awareness

Evolve from being self-conscious about your writing to being self-aware; give yourself permission to be openly, hugely ambitious.

"I'm always telling my students is you you want to evolve from being self-conscious"

— 30:05

The evolution includes giving yourself permission to be openly, hugely ambitious, and the permission is the difficult half. Self-consciousness is a form of hedging, and a hedged sentence cannot carry an ambitious idea.

31:35 - Real ambition is rooted in reverence for the creation

Real ambition is rooted in reverence for the creation; ego wants to be the person who made it and cares more about the end than the making.

"of the permission to take on a grand project. Yeah. Yeah. Because it's there's"

— 31:35

Ego wants to be the person who made it and cares more about the end than the making, and the test is which of the two you are attached to when the work is going badly.

32:37 - The greatest authority on the meaning of the work is the work

The greatest authority on the meaning of the work is the work; learn to shut out editors', agents' and workshop voices while building.

"will I will avoid my agent and my editor for years at a time"

— 32:37

You cannot observe closely while hearing outside voices, and distraction is the enemy of attention. A writer avoids his agent and editor for years, not wanting anyone in the house while he builds it — and there is no wrong way: some brilliant writers thrive on a dozen readers and a feedback loop.

34:12 - Being trapped by identity is a form of self-consciousness

Being "trapped by identity" — asking what a real writer would do — is a form of self-consciousness; jettison it and do whatever it takes to get it on the page.

"into the I don't it's I'm self-conscious, right? I don't want to do something"

— 34:12

Asking what a real writer would do is a form of self-consciousness. Jettison it and do whatever it takes to get it on the page — the identity is a costume being worn at the desk.

36:04 - There is no obligation to face the blank page daily

There is no obligation to face the blank page daily; feed the work by letting many idioms — books, music, paintings — become co-extensive.

"is just you got to know yourself. This is again that idea, but because"

— 36:04

No obligation to face the blank page daily. The work gets fed by letting many idioms — books, music, paintings — become co-extensive, which is a description of input rather than of avoidance.

37:01 - A thing gets described in the verbiage of its opposite

Describe a thing in the verbiage of its opposite — a gas as a solid, the silent as loud — because habituated verbs are where writing goes dead.

"like as you know the sort of thought exercises is I'll think you know"

— 37:01

A gas described as a solid, the silent described as loud. Habituated verbs are where writing goes dead, and the opposite's vocabulary is what revives them — the deadness is in the habit rather than in the subject.

39:23 - Knowing a word more deeply gives a sentence dimensionality

Know a word a little more deeply than its surface usage and your sentences gain dimensionality and depth.

"read the dictionary every day. The old every day. Yeah, every day. Every day."

— 39:23

The gain is dimensionality rather than precision: the sentence acquires a second layer the reader feels without stopping to name. It is the most practical instruction here and the one that compounds slowest.

41:41 - Modern usage has reduced the dynamic range of the language

Modern usage has reduced the dynamic range of language; a slightly deeper word slows the reader just enough to let meaning saturate.

"Well, I think usage does I think like you know a lot of advertising"

— 41:41

A slightly deeper word slows the reader just enough for the meaning to saturate, and the slowing is the mechanism rather than a side effect.

43:31 - Too much plot is machinery that crowds out consciousness

Too much plot is machinery that crowds out the character's consciousness; keep a simple spine so the mind can move associatively.

"into why you're not that interested in plot."

— 43:31

Too much plot is machinery crowding out the character's consciousness. A simple spine lets the mind move associatively, which is the thing the reader actually came for.

45:31 - Character is made through patterns of association

Character is made through patterns of association; give each character and each place its own repertoire of verbs, nouns, colours and light.

"think of thought and experience and description you know so I'm desribing describing experience."

— 45:31

Each character and each place gets its own repertoire of verbs, nouns, colours and light, so character is built out of vocabulary rather than described by it.

48:36 - The aim is maximum density with maximum readability

Aim for maximum density with maximum readability — a short book that feels as substantive as a vast one.

"Yeah. And I'm going to try to do that though. I'm going to make"

— 48:36

A short book that feels as substantive as a vast one is the shape of it. Those two usually trade against each other, which is what makes it an aim rather than a specification.

50:28 - Lessons and takeaways do not go in the work

Do not put lessons, takeaways or messages in the work; if a reader gets "the point," they never have to think about it again.

"live in terror of, you know, what's the point? What's the takeaway? What's the"

— 50:28

Messages tend to be tautological and trivial, and art should open a door and dilate the reader rather than close the question. Live in terror of what's the takeaway — a message like be kind to strangers is one nobody needed the book to learn.

52:26 - The reader gets brought to a moment rather than told what to think

Bring the reader to a moment and show them what you found rather than telling them what to think; telling would be fraudulence.

"Yeah. And that's what I want. I'm always just trying to pursue, you know,"

— 52:26

A dialogue makes the reader weigh perspectives; imposing an answer does violence to the reader, the character and the writer at once. Books are thought of as galleries — look at that, here's this — curating an experience rather than delivering a verdict.

53:43 - Fiction persuades by recognition rather than by argument

Fiction persuades not by argumentation but by recognition — the reader says "that's true, I've just never seen it in words".

"persuade through poetry or aesthetic, you know, sort of fiction um is by recognition."

— 53:43

With no thesis to prove, aesthetic writing works by pointing at something the reader instantly recognises for themselves. A character says here I go again, I know better, and the reader thinks exactly that about their own life.

54:31 - An experience described precisely enough makes a reader feel seen

Describe experiences — especially the hardest ones — precisely enough that a reader in another time or culture feels their own dignity respected and feels seen.

"think that's the sort of thing where I I also think I want again"

— 54:31

The test is whether a reader in another time or culture feels their dignity respected and feels seen. Precision is what does the respecting.

59:28 - Nobody is able to write for everyone at once

You cannot write for everyone; billions leave and a dozen come to the front row asking "what happened next" — write for them.

"Well, I think what you're saying is dare to write with depth. And the"

— 59:28

Trying to please everyone pleases nobody, and the committed few are who the work is actually for. The image is billions stampeding for the exits while a handful move down from the cheap seats to the front.

1:02:14 - The romance of the untutored writer is showy and clichéd

The romanticism of the untutored, self-destructive writer is showy and clichéd; craft — even grammatical correctness — matters.

"and you know, it's a little showy and I was younger and whatever. Um,"

— 1:02:14

Craft matters, down to grammatical correctness, which is an unglamorous position stated without embarrassment. The romance survives because it excuses the part of the work that is tedious, and the excuse is what it is really for.

1:03:00 - If beauty inheres in the thing, the job is to describe it precisely

If beauty inheres in the thing, your job is to describe it very precisely; precision fills description with wonder rather than making it clinical.

"after time. If it's there, if it inheres in the stuff itself, then my"

— 1:03:00

If beauty inheres in the thing, the job is to describe it very precisely. Precision fills a description with wonder rather than making it clinical — the opposite of what people fear about it.

1:06:41 - Describing a thing is not explaining it

To describe something is not to explain it; to know it is not to understand it — you can write what you do not understand.

"lot of times students think they have to understand everything before they start writing."

— 1:06:41

Students think they must understand and control everything before writing, when the requirement is only to describe rather than to resolve. A character simply left his family, written without the writer needing to understand why.

1:07:51 - The work is portraiture rather than explanation

Think of it as portraiture — you depict a person with experiential richness rather than explaining them; extend the same care to a student's imagination.

"making portraits. you're you're writing portraiture. So if you're if you're if you think"

— 1:07:51

Experiential richness rather than explanation, extended to a student's imagination as readily as to a character's. The teaching stance and the writing stance turn out to be the same one.

1:09:16 - Writing is iterative at every level

Writing is iterative at every level — you cannot think your way to being better; you have to get it wrong in a particular way to get it right.

"I I think writing is just iterative. You can't think your way to being"

— 1:09:16

You cannot think your way to being better. It is iterative at every level, and you have to get it wrong in a particular way in order to get it right, which makes the wrong versions load-bearing.

1:10:56 - Far more gets written than survives

Expect to write far more than survives — roughly a thousand pages for a short book — and cycle between distilling down and building back up.

"Easily a thousand. Easily a thousand. Yeah. Um I mean what one of the"

— 1:10:56

Printing and rereading the whole manuscript every several weeks — the systole and diastole of expansion and compression — is what finds the book inside the pages. A short published novel came out of easily a thousand drafted ones, reread with a blue pen.

1:12:52 - Everything of interest goes in, and the best expression gets found

Let everything you are interested in into the manuscript, then write your way to the best expression; the pages it took were not wasted.

"just come I love it if you're trying to be efficient if you're writing"

— 1:12:51

The best expression is arrived at by writing, not chosen before it. The pages it took were not wasted, which is the same claim from the other direction.

1:16:04 - Revision has precision of expression as its goal

Make the goal of revision precision of expression; it is the best insurance against ornamentalism, and lucidity should let the reader forget they are reading.

"right. That's all. If my goal for revision is precision, perfect precision of expression."

— 1:16:02

Precision of expression as the goal of revision. It is the best insurance against ornamentalism, and lucidity should let the reader forget they are reading at all.

1:18:04 - Narrative fiction is a way of keeping time

Narrative fiction is keeping time; write with tempo, pacing and dynamics, and often know a sentence's beat before its literal meaning.

"timekeeper, you know? So, you you've got the tempo, you're setting the tempo, but"

— 1:18:04

Time refracts through the character: it can slow, speed, run in half or double time, which is why the prose gets scored like music. A former drummer hears how many beats are in a sentence before he knows what it says.

1:20:16 - Rhythm is what suggests the words

Let rhythm suggest the words — monosyllables for one feel, tumbling polysyllables for another — but keep it angular, not singsong.

"He tinkered tin pots, rod iron, solder"

— 1:20:16

Sound shapes sense, and too much symmetry or internal rhyme turns cutesy, so choppiness and swagger are doing the work of mood. A passage about a peddler is built from clipped, funky, monosyllabic sound to give him a swagger he did not have in life.

1:22:09 - The writers a person loves live in the work as a conversation

The writers you love live in your work as a continuing conversation; study how they achieve their effects, not the scholarship about them.

"My favorite sentence in Tinkers is a sen"

— 1:22:09

They live in the work as a continuing conversation. Study how they achieve their effects rather than the scholarship about them, which is a redirection from criticism toward mechanism.

1:24:20 - The work gets layered so nobody is excluded

Layer the work so deep readers find riches and casual readers are never excluded; presume the intelligence of a mixed audience.

"um that now as a writer I think of as if you if you"

— 1:24:20

Deep readers find riches and casual readers are never excluded. Presume the intelligence of a mixed audience — layering is what lets one piece serve both without condescending to either.

1:26:06 - Writers start as readers and eventually have to talk back

Writers start as readers and eventually have to "talk back" — literary work is fiction in conversation with other work, a fingerprint of the mind.

"transcendentalist teach you how to consult your own soul? I mean, I very much"

— 1:26:06

Literary work is fiction in conversation with other work, a fingerprint of the mind. Writers start as readers and eventually have to talk back — and the craft's cost is confronting your own limitations daily, which is also what makes the moment of chasing something down into the language precious.

Personal Branding Lessons

A long conversation about craft with almost no method in it. The moves below are what survives the refusal.

Slow down, because patience is a skill you learn

Named as the one instruction relevant almost all of the time. Patience is not a temperament somebody either has or lacks — it is a skill acquired deliberately, and reaching raw sensation requires observing before the language arrives to cover it. 15:21

Stay utterly literal and test the image forward

Build an image by refusing to reach for the figurative, then asking what happens if you stick with this, and following it until it arrives somewhere. The literalness is the method rather than a restriction placed on it. 05:20

Write for brilliant, big-hearted readers

Do not write for bad readers. Treat richness of meaning as an act of generosity toward people who will meet it — writing down is the insult that most often gets mistaken for accessibility. 23:38

Keep no lessons or takeaways in the work

If a reader gets the point, they never have to think about it again. The conclusion closes the book permanently, which is why the work brings somebody to a moment and shows them what was found rather than telling them what to make of it. 50:28

Describe a thing in the verbiage of its opposite

A gas as a solid, the silent as loud. Habituated verbs are where writing goes dead, and borrowing the opposite's vocabulary is what revives them — the deadness was in the habit rather than in the subject. 37:01

Give each character its own repertoire of nouns and verbs

Character is made through patterns of association, so each person and each place gets its own set of verbs, nouns, colours and light. It is built out of vocabulary rather than described by it, which makes it a decision made before the writing. 45:31

Make precision of expression the goal of revision

Precision is the best available insurance against ornamentalism, and lucidity should let the reader forget they are reading at all. Revising toward anything else — elegance, impact, voice — tends to add rather than remove. 1:16:04

Study how the writers you love achieve their effects

Not the scholarship about them but the mechanism inside the work. They live in your own writing as a continuing conversation, and what transfers is technique observed directly rather than interpretation received second-hand. 1:22:09

Questions

Each answer ends at the moment in the recording where it is given.

Should I outline my work or improvise it?

The case made here is for improvisation: going in without a map and taking what the language and the material hand back, on the reasoning that a complex-enough system, once set in motion, produces things that could not have been planned. It is offered as one writer's taste rather than a law — a love of outlines and plots is granted as an equally valid way to work. The underlying non-negotiable is not the method but the attention it requires. 00:48

Why does my writing feel flat even when it is technically correct?

Because it is likely built from received, off-the-rack language — the pre-fabricated phrasing everyone uses to get through a day. That language is muscle memory, and it skips the act of actually seeing the thing, so it goes dead on the page however grammatical it is. The described fix is to slow down beneath the reflex to something closer to raw sensation and to name what is literally there, rather than reaching for the phrase already waiting. 03:16

What is the single most important habit for getting better?

The answer, given as the message that matters nearly all of the time, is to slow down. Patience is treated as a learnable skill rather than a temperament: taking in what is actually on the page instead of ramming material through to make it behave. It pairs with a refusal to hurry past the real moments, which is also why too much plot is treated as a problem — machinery leaves no time to see. 15:21

How do I find my own voice instead of copying writers I admire?

Voice is framed as a fingerprint. Almost all of what you produce overlaps with everyone else; the connection a reader feels lives in a tiny fraction that is only yours, formed as experience passes through one particular consciousness. It shows only when you slow down enough to let the real words through instead of the reflexive ones. You absorb the writers you love as a continuing conversation — studying how they achieve an effect — without writing like them. A differentiation audit is the deliberate version of finding that fraction. 17:24

Should I write for a wide audience or a narrow one?

Narrow, and deliberately. The image offered is of most of an audience heading for the exits the moment a real opinion appears, while a small group moves toward you asking what happens next — and those few are who the work is for. Trying to hold everyone means softening the thing the few came for, so you please no one. Choosing who you are actually writing for is treated as part of the work, not an afterthought. 59:28

Why shouldn't I put a clear takeaway or message in my work?

Because a message is a door that closes. If a reader gets the point, the work is spent — there is nothing to return to. Inserted lessons also tend to be tautological, things the reader already knew, dressed up. The alternative described is to bring the reader to a moment, show them what was found, and let them draw their own conclusion; imposing an answer is called a kind of fraudulence. Work that stays open is what people come back to. 50:28

How much should I expect to cut, and is all that drafting wasted?

Expect to write far more than survives — the figure given for a short book is easily a thousand pages — and to move in cycles of expanding, distilling, and building back up, rereading the whole manuscript every several weeks. The pages spent to earn a single paragraph are explicitly not wasted; you often have to write your way to the best expression, and the dead ends are how you find the part worth keeping. 1:10:56

How do I make a description vivid without it sounding purple or clinical?

By making precision the goal of revision. The fear that describing something exactly turns prose clinical is answered directly: precise description fills a thing with wonder rather than draining it, because fiction gives both the literal object and a character's experience of it. Vividness comes from layering — the colour is wrong, so you ask why, then add smell, then rhythm — until the reading is lucid enough that the reader forgets they are reading and the richness still holds. 1:03:00

How do I handle feedback and outside voices while I am still building?

The stance taken is to keep them out until the work is built — avoiding editor and agent for long stretches — on the grounds that you cannot observe closely while listening for what someone else will want, and the greatest authority on the meaning of the work is the work itself. This is hedged carefully: it is one writer's way, and others thrive on a dozen readers and a live feedback loop. What matters is knowing which builds the trust you need to keep going. 32:37

Sources

Pulitzer Prize Winner Teaches Writing (Paul Harding Interview)

The sections above follow the episode's own order. Each timestamp is the point in the recording where the idea is discussed, so the post reads as a map of the conversation rather than a rearrangement of it.

Quotations come from the episode transcript with spoken filler ("okay", "right", "you know", "um", "actually", "kind of", "sort of") and false starts removed. No words are added: square brackets mark anything inserted for sense, and an ellipsis marks any cut. British spellings are restored, and speech-to-text errors were repaired against context only where the intended word is unambiguous. Where a line appears both in a cold-open teaser and in its place in the conversation, the timestamp cites the second — the point at which it is actually said.

Quotes are stamped with the moment they are said rather than with a speaker's name.