Slow Words.

On writing slowly, and why I have made peace with it

I am not a fast writer. I have said this so often, in so many small introductions and bios and replies to kind emails, that the phrase has worn smooth in my mouth, like a stone. But the saying of it still feels important, because the world we live in, the world I wake up in every morning and check my phone in before I have even drunk my water, is a world that rewards speed almost above all else.

A friend of mine, a journalist in Copenhagen, writes a thousand words before lunch most days. She writes well. I do not begrudge her this. But when I have tried to work at her pace, what I produce is thinner — not shorter, necessarily, but thinner, as if the sentences had been beaten out with a hammer rather than allowed to settle.

I think about a line from Annie Dillard, who said that how we spend our days is, of course, how we spend our lives. I have tried, for some years now, to spend my days slowly. To walk in the morning before I sit down. To re-read what I wrote yesterday with a pencil and not a keyboard. To leave a sentence overnight if I am not sure of it, the way you leave bread to prove.

This blog is called Slow Words because I needed to give myself permission. The name is a kind of contract I have with myself. If I write only one good paragraph in a morning, I have not failed. I have done my morning's work. And if you have read this far, I thank you for slowing down with me.

A walk along Egå Engsø, in the rain

The rain came in sideways yesterday — the kind of fine, persistent rain that you do not notice is soaking you until you are already soaked. I went out anyway. Tove, our dog, looked at me as if I had betrayed her, but she came.

Egå Engsø is a re-flooded lake, just north of the city. It used to be drained farmland; in 2006 they let the water back in, and now it is a haunt of greylag geese and herons and, in the right season, a great many bird-watchers in expensive coats. Yesterday there were no bird-watchers. There was a single fisherman under a green umbrella, and a man in a bright orange jacket running with his head down. There were the geese, of course. The geese do not care about rain.

I find that walking in poor weather is the closest I get to a kind of involuntary meditation. You cannot, in driving rain, think very far ahead. You cannot plan. You can only put one foot down, and then the next, and notice that your sleeves are heavy now, and that the reeds along the path are bent over with the weight of the water, and that the lake itself looks like beaten pewter.

I came home and made coffee and sat at the kitchen window for a long time without writing anything. This too, I think, is part of the work.

Notes on The Old Ways

I have read Robert Macfarlane's The Old Ways three times now. Each time it has been a different book. The first time I read it for the language, which is, as everyone says, gorgeous — sometimes almost too gorgeous, like a meal with too many courses. The second time I read it for the walks themselves, the routes, and I made a list of paths I would like one day to walk. (I have walked two of them. I will walk the others, perhaps.) The third time, which was this winter, I read it for what it says, almost in passing, about the relationship between place and language.

Macfarlane keeps returning to the idea that the words we have for landscape shape what we are able to see in it. When the word for a kind of stream goes out of use, the stream itself becomes harder to perceive — it slips a little out of focus. In Danish we have a word, hygge, which everyone abroad now knows, and a hundred words for the moods of the sea which almost no one knows, even here.

I think this is part of why I write. Not to invent words — I am not Macfarlane, and I am not Hauge — but to point at the things the words still hold, before the holding fails.

"We are and always have been name-callers, christeners. Words are grained into our landscapes, and landscapes grained into our words."

I keep that line on a Post-it above my desk. It has yellowed.

Why I write in English

People ask me this often, and I do not always have a tidy answer, so let me try to set one down. I write in English even though my Danish is better, which is to say: my Danish is the language I dream in, swear in, comfort my mother in, argue with my partner in. My English is something else. My English is a tool I picked up in school, polished in graduate school, sharpened on a hundred scientific papers, and then, late in life, turned to a purpose it was never designed for.

The first reason I write in English is practical. The readers I want — the readers who care about the small literature of walking and watching and noticing — are mostly not in Denmark. We are a small country with a beloved but inward-looking literary culture. If I want to be in conversation with the writers I love, most of whom I read in English (even when, like Espedal or Hauge, they were not writing in it), I have to walk over to where the conversation is happening.

The second reason is, I think, more interesting. Writing in a second language is slow. It is a kind of permanent first draft. Every sentence has a small drag on it, a friction, because I am not sure — I am almost never sure — that the phrase I have reached for is the phrase a native speaker would have reached for. This makes me cautious. It makes me hesitate. It makes me re-read.

In Danish I can fly. In English I have to walk. And as I have said before, walking is what I am for.

The North Sea in March

I went west last weekend, to Thyborøn, which is one of those small Jutland towns the sea has been chewing at for centuries. The North Sea in March is not the North Sea of postcards. It is grey on grey on grey, and the wind comes off it with a kind of personal grudge.

I spent twelve years studying this sea. I know its temperature curves, its salinity at depth, the migration of its cod, the slow disaster of its sand-eel populations. I know it as data — and I know it, also, as a place I used to stand at the edge of, on early-morning sampling trips, holding a clipboard and trying not to lose my pen to the wind.

What I do not have, even now, is a way to write about the sea that holds both of those knowings at once. The scientific knowing wants tables. The other knowing wants something closer to prayer. I have not yet found the sentence that does both. I do not know if there is one.

I stayed at a small guesthouse where the woman who runs it remembered me from a paper I had written about coastal erosion, and would not let me pay for breakfast. This was kind of her, and I felt, briefly, like I had not entirely wasted my twelve years.

A piece of software that thinks a book is a network

There is, buried inside a piece of software called Quilva, a small philosophical claim. It is stated almost in passing in the user guide, and I have been turning it over for the better part of a year:

"A large creative work is not a linear document. It is a network."

This is a strange thing for a piece of writing software to say. Most writing software believes, in its bones, that a book is a long string of characters — a document — and everything in the interface flows from that belief. You scroll down. You search forwards. You see your manuscript as a tall, thin rectangle viewed through a window. The metaphor underneath the tool is the typewriter, and the typewriter was a machine for producing one line after another in one direction.

A book is not really like that. A book is a web. Chapter twenty-two is held up by a sentence in chapter three. A character's silence in one scene is a payoff for something a different character said two hundred pages earlier. The setting of a place depends on a piece of research that lives, in your real life, on a phone photo from a holiday in 2019. The book is a structure of relationships, and the writer's actual day-to-day work is keeping those relationships in coherent order in their head while also, secondarily, typing nice sentences. This is, I think, the unspoken difficulty of writing long things. Not the prose. The bookkeeping.

Quilva is the first piece of software I have used that takes this seriously as its first principle, rather than as a feature it added later. The structure of your book — the parts, the chapters, the scenes, the cards beneath them — is not a sidebar. It is the spine of the application. Every other view (a corkboard of beats, a mind map of bubbles, a video-editor-style timeline of threads, a database of characters and places, a vault of research files) is a different window onto the same underlying structure. Move a scene in one view, and it has moved everywhere else, instantly, without a sync, because there was only ever one thing there to move.

I cannot overstate how different this feels in practice. For years I kept my outline in one program, my notes in another, my research in folders on my desktop, my draft in a third program, and a great many things on actual paper. Reconciling these took, by my own honest estimate, about a fifth of my working hours — the lost fifth, the fifth I did not write in. Every time I changed the order of two chapters I had to update four other places, or I had to not update them and live with a slow accumulating drift between what I had planned and what I had written. The drift was worse than the work of fixing it, because it nagged.

The version of me who has used Quilva for ten months does not have that fifth-of-a-day tax any more. I move a scene. It has moved. The summary in the planner reflects it; the position on the timeline reflects it; the page of the book, when I publish it, reflects it. I had not realised how heavily I was paying that tax until I stopped paying it.

There are individual things I have come to love, and I will mention them only briefly because this is not a feature tour. The page scrolls under your cursor as you write rather than the cursor crawling down a static page — your eyes stay in the same place on the screen all afternoon, and at the end of three hours your neck is not stiff. The right-hand panel, which the developers call the Compass, quietly hands you exactly the planning notes and beats relevant to the scene you are currently writing, and nothing else. The cards you make can be filtered by colour, so you can see at a glance that the middle of your book has gone red — which is to say, gone too tense — for three chapters in a row, and that you should put something quieter in there. These are small inventions, but they are inventions in a field that has been very short of them.

I have used most of the established tools at one time or another over the past decade. They are good. Some of them have been good for a very long time, and that is part of the issue: they have been good in the same way for a very long time. They were built around the document, because the document was the metaphor available when they were designed. Quilva was built around the network, because the network is, in fact, what a book is. The difference is not cosmetic. It changes what kind of thinking the software is hospitable to.

A note here, before I go further. I came to Quilva as one of their late-stage beta testers, and stayed on a licence the team sent me when the beta closed. Nobody is paying me to write this, and nobody has asked me to. They sent me a licence file in an email and asked, kindly, what I thought. What follows is what I thought.

I want to say something about the screenplay side of Quilva, because it is the part I did not expect to use and now use more than anything else. The application is marketed at authors of long prose — that is its self-image, and it is the side the website leads with — but I have come to think, slightly to my own surprise, that its strongest audience is actually screenwriters. I say this as someone who has, in the last six months, drafted two scripts in it, and who came in expecting to do nothing of the sort.

At the surface, the screenplay editor does the things you would expect any modern screenplay editor to do. Scene headings, action, character, dialogue, parentheticals — all the elements are there, all correctly formatted, all driven by the keyboard the way they are in the dedicated industry tools. What I have come to appreciate is that the small refinements are mine to make, not the program's to insist on. Character and scene autocomplete can be flipped on or off without ceremony, in the spot where you actually want to do it; the typewriter mode I mentioned above is, if anything, even more useful in a screenplay than in prose, because a screenplay is a thing you read down the centre of the page. None of this, in isolation, is unprecedented. The combination is unusually well-judged.

The thing that has genuinely changed how I work on scripts, though, is the Compass — the same right-hand panel I mentioned for prose. In screenplay mode it does something quietly remarkable, which is that it lets me write the script in bite-sized chunks. Each scene comes to me with exactly its planning attached: the beats I had set down weeks earlier, the character notes, the bit of research I had pinned to it, the question I had left for myself. Nothing else. I am not scrolling through a hundred and ten pages of script looking for where I was. I am writing the scene in front of me, with the small dossier of that scene at my elbow, and when it is done I move to the next one. For a first draft, where the entire battle is keeping momentum without losing the shape of the thing, this is — and I am choosing the word carefully — unparalleled. The planning tools, taken with the Compass, do more for me at the drafting stage than any dedicated screenplay tool I have tried.

And then, when there is a draft, the Publisher turns it into a PDF that I am, frankly, proud of. I have produced a great many ugly screenplay PDFs over the years out of a great many programs. The one Quilva exports is the most beautiful I have seen come out of writing software. You can design your own title page, drop in an image for the front, set the typography, and the result is press-ready in a way that the industry tools, for all their authority, do not bother to make easy. If you want Final Draft at the end of it — and there are perfectly good reasons to want Final Draft at the end of it, particularly if you are going into production — you can export the whole script to Fountain and import it from there. But I have stopped doing that for the early drafts. There is nothing Quilva is not doing well enough for me to need to leave.

I should say what it is not. Quilva is not the place to go if you want the warm hum of a long-established community — there is no decade of forums, no Reddit thread full of clever shortcuts, no cottage industry of templates from working authors. The team is small and replies to emails personally, which I find charming, but I am aware that "charming" is also a polite word for "not yet at scale". The screenplay tools, capable as they are, do not yet have the institutional weight of the industry-standard tool that every production office in Los Angeles knows the name of; if you are handing pages to a studio next week, that still matters. The book-design module is genuinely powerful — astonishingly so, in fact, for something built into a writing application — but it is its own world to learn, and if you only want to draft and hand the manuscript to a designer, you can ignore it and lose nothing.

And there is one thing I want to register as a real, not-rhetorical hesitation, because I cannot in honesty leave it out. Everything in Quilva — outline, cards, world, research, manuscript — lives inside a single proprietary project file. This is, in almost every respect, a gift. The project is portable, self-contained, and entirely yours; you can put the whole studio on a USB stick and travel with it. But it is, undeniably, a basket, and you are putting your eggs in it. The team has been forthright that the format is documented and that you can export the manuscript at any time to ordinary formats, which I have tested, and which works. Still: I back up obsessively, to two separate drives, and I would recommend the same to anyone considering it. This is not a critique unique to Quilva. It is the cost of choosing any tool that does this much.

None of these caveats are small. I list them in earnest. And I will say, with all of them in front of me, that Quilva is the best and most intuitive piece of writing software I have used. I do not say this loosely. I have used a great many of them. This is the one. It is the one whether you are writing a book or writing a script, and if you are writing a script, it is — I am increasingly sure — quietly the most useful tool out there for the part of the work where the script is still becoming itself.

What I have ended up with, after ten months, is a way of working in which planning, drafting, research, and structure are not four different rooms I have to walk between, but four windows in one room. I had not known I wanted this. I had assumed the friction of switching tools was simply the cost of writing, the way the friction of bad weather is the cost of walking. It turns out it was not. It turns out it was just bad weather we had got used to.

If you write long things — or short scripted things — Quilva is at the very least worth the free trial. I do not say this lightly. I have, in this same decade, downloaded and abandoned a great many trials. This is the only one I did not, in the end, abandon. The licence file is still in my inbox. I read it the way I read a thank-you note. I have meant to reply for some weeks now.

On Knausgaard, with some misgivings

I have complicated feelings about Karl Ove, as I think most of us do who grew up reading him from across a narrow stretch of sea. To read My Struggle in Norwegian is, I am told, a different experience from reading it in English; the prose is plainer, somehow, less translated-into-literary. In Danish it sits somewhere in between.

What I admire is the willingness to be boring. He will spend forty pages on a children's birthday party, and you will keep reading, and you will not be able to say afterwards exactly why. There is a quality of attention there — a refusal to summarize, to skip ahead, to do the polite literary thing — that I find genuinely radical.

What I struggle with is harder to name. Something to do with how much of other people's lives he uses up, perhaps. The collateral damage of total honesty. A friend who knew his first wife says that some of the worst pages are also the most accurate, and this does not comfort me, it troubles me more.

And yet. And yet I have read all six volumes, and I will probably read them again, and when I want to remember how to write a sentence that does not flinch, I open one of them at random and I read until my coffee is cold.

Hauge in the orchard

Olav H. Hauge spent most of his life in a small house in Ulvik, on the Hardangerfjord, tending an orchard of apple trees and translating Hölderlin and Yeats and Bashō in his spare hours. He published his first book at thirty-eight and then went on, slowly, slowly, for another fifty years.

I think of him often when I am tempted to feel late, or behind, or as though I have started this writing life too far along to make anything of it. Hauge was a gardener who wrote. I was a scientist who writes. Neither of us came in through the front door.

There is a poem of his that I love, and which I will not quote here in full because the translations vary and I do not want to choose badly, but it ends with the image of a small boat — your life — held by a thin rope to the shore, and the suggestion that this is enough. That the boat does not have to be a ship. That the rope does not have to be a chain.

I keep, in the back of my notebook, a list of writers who started late. Hauge is at the top. He is followed by Penelope Fitzgerald, by Raymond Chandler, by Annie Proulx. I add to the list when I feel I need to.

Tove and the woods at Marselisborg

Our dog is named Tove. She is six. She is a kind of mixed-up Danish thing — some spaniel, some terrier, some opinion of her own — and she came to us four years ago, already named, from a shelter in Randers. We considered renaming her, briefly, and then we did not. It seemed an indignity.

Tove is the reason I walk every day, which is to say she is partly the reason I write. Without her I would have mornings where I sat at the desk too soon, before the sentences had a chance to assemble themselves in the air. With her, I have no choice. By half past seven she is at the door with her lead in her mouth, looking offended.

Our usual walk is in the Marselisborg woods, just south of the city. It is not wild, exactly — there are well-kept paths, and benches, and on Sundays you can hear the deer park from a distance — but it is enough trees and enough quiet to be useful. In autumn there are beech leaves underfoot that make a sound like soft applause. In winter the path turns to a corridor of trunks. In spring the wood anemones come up so thick along the slopes that the floor seems to have iced over white.

I have written, by now, a great many sentences in my head along that path. Tove does not know this. She is interested in other things.

What I learned from twelve years at the university

I left academia in 2015, after twelve years, and I have been asked since, more times than I can count, what I miss and what I do not miss. The honest answer is that the things I miss and the things I do not miss are often the same things, looked at from different angles.

I miss the long-form attention of fieldwork. The way three weeks on a ship can turn into a single, slow week in your head — a week without phones, without errands, without anything but the next station and the next sample. I do not miss the eight hours afterwards spent in front of a spreadsheet.

I miss colleagues who could tell, from a single graph, that something was off. I do not miss the meetings about strategic priorities for stakeholder engagement, which were where good ideas went to die a slow institutional death.

What I have kept, I think, is a way of looking. A trained suspicion of pretty patterns. An attention to error bars. When I read other people's nature writing now, I find I am alert in two registers at once — the literary, and the empirical — and I am quietly grateful when both of them are satisfied.

And when only the literary register is satisfied, I have learned, mostly, not to write the angry letter. Mostly.

Books I read last year, with brief notes

I keep, every year, a small notebook of books read, with a sentence or two on each. The notebook for last year is on my desk in front of me, and I thought I would share some of it, in case anything catches your eye. I read forty-one books in 2025, which is fewer than the year before. I do not believe in counting, but I count anyway.

Underland, Robert Macfarlane. Of course. Less perfect than The Old Ways, but braver. The chapter on the catacombs of Paris stayed with me for weeks.

Bergeners, Tomas Espedal. Read in English. I should read it again in Norwegian; the rhythm is unmistakably Espedal, which is to say, a man walking and thinking and refusing to lie.

Pilgrim at Tinker Creek, Annie Dillard. A re-read. I had remembered the giant water bug. I had forgotten almost everything else, which is to say I had forgotten the book.

H is for Hawk, Helen Macdonald. Late to it. Loved it. Wished, oddly, for a little less T. H. White.

The Living Mountain, Nan Shepherd. Short. Perfect. I do not know how she did it. I read the whole thing one evening in November and then sat for a while not doing anything.

There were others. Some I gave up on. One I threw, gently, at the wall. That one I will not name.

On the Danish word hyggekrog

Everyone abroad has heard of hygge now, and I am tired of explaining it, mostly because what was once a quiet domestic word for a quiet domestic feeling has been, in the last decade, hammered out into a marketing flat-pack, sold by the cushion. I will not write about hygge. I will write about hyggekrog, which is what hygge wishes it still was.

A hyggekrog is, literally, a "cosy corner" — a small, defined nook in a room that has been arranged for sitting in. A window seat with a cushion. A chair beside the wood stove. The end of a kitchen bench where the light is good. It is, importantly, a place, not a feeling. You can point at it. You can clean it. You can put a candle on the windowsill of it.

I think one reason hygge travels so badly is that the rest of the language has been left behind. Hygge is the noun for what happens in the hyggekrog, and the hyggekrog is the architecture that produces it. Without the corner, the feeling has nowhere to live. You end up trying to manufacture it with candles in the middle of an open-plan kitchen, and it does not, in my experience, work.

My own hyggekrog is the deep window seat in our flat, facing east. In the mornings I sit there with a book and a cup of coffee, and the radiator clicks under me, and Tove arranges herself across my feet, and there is nothing to be sold. This is the part the cushion companies cannot reach.

A note on starting

If you are reading this and you have been thinking, for some time, that you would like to write — about nature, about your walks, about a place you know, about the small territory of your own attention — let me say this plainly. You should begin.

You should not begin grandly. You should not begin with an essay. You should begin with a paragraph. The paragraph does not have to be good. The paragraph only has to exist. Tomorrow you can write another paragraph, and the day after, another, and in a month you will have a small pile of paragraphs, some of which will be embarrassing and some of which will surprise you.

I started this blog in 2016, when I was thirty-four and still working full-time at the university, and I wrote it badly for the first two years. I have not gone back and deleted those early posts, because I think it is useful, if you are starting now, to scroll back and see that everyone is bad before they are anything else. The work is not to begin well. The work is to begin, and then to keep going until you are not the person who began.

I am rooting for you. Truly. Write the paragraph.