Counting Words, Drawing a Line — What Writers Actually Measure

In the previous article, I argued that writing regularity is not discipline but an inhabitation device: what prevents the work from becoming foreign to the one who writes it. That article ended on a peculiarity of this practice: a text's progress cannot be read in the moment. The writer can see their file filling, the pages accumulating, the counter rising. But knowing whether what they have just written advances the project — whether the structure holds, whether the passage finds its place, whether the rhythm is taking shape — requires something else: reading, judging, sometimes letting it rest for several days. At the scale of the session, the work itself sends no signal.

In this silence of the work, measurement takes hold.

Routine and Trace

One must first agree on what we are talking about. Having a routine — being at one's desk from 9 to noon each morning — and keeping a record of it are two distinct things. One can hold the first without the second. Flannery O'Connor sat at her desk three hours every morning for decades, without noting anything. Joan Didion sustained her work without a writing journal, and Gabriel García Márquez wrote from 8 to 1 with no quantitative notation either. The routine, for them, was sufficient in itself.

Others — and this is the case for the majority of writers whose daily practice is documented — add to the routine an additional gesture. A number noted, a cross drawn, a line in a journal: something that records the fact of having been there. Trollope kept a numerical log. Hemingway pinned a card on the wall. Virginia Woolf commented in a journal on her relationship to what she was writing, and Seinfeld, later, kept his chain of crosses¹. This gesture — the trace rather than the routine it testifies to — is the subject of this article.

Why, when one already holds the routine, feel the need to leave proof of it? The answer, following the previous article, is that the routine itself can become blurry to its own eyes. The writer has no attendance sheet, no employer, no time clock. No one checks whether they are there. To the invisibility of the manuscript's progress is added, for those who doubt, the invisibility of their own regularity — unless it is recorded.

What Measurement Actually Measures

The traces we have just described have something strange about them, when looked at closely. None provides information on the quality of what was written. The word count does not say whether the scene holds. The cross on the calendar does not say whether today's 200 words were the right ones. What these traces record is something else: that one was there. They inscribe presence in the practice, not the manuscript's progress.

This shift is not merely a nuance. It changes the role we assign to measurement. In most human activities, measuring means checking a result: a runner records their time to know if they are running faster, a salesperson looks at their figures to know if they are selling more. In writing, this result cannot be read in the moment. The trace of presence becomes, by default, the only reliable daily signal available.

Psychologists Teresa Amabile and Steven Kramer analysed 11,637 daily journal entries from 238 creative workers². Their main finding is clear: of all the factors influencing motivation over the course of a day, the most powerful is the perception of progress in meaningful work. Progress was cited in 76% of participants' best days. Not recognition, not remuneration: the perception of progress.

Yet in writing, real progress — that of the manuscript — cannot be read on a counter. What form of progress do these traces make visible? Albert Bandura, in his work on self-efficacy³, offers a precise answer. When results are delayed or ambiguous — the very definition of creative work — one cannot derive from them the feeling of having done one's job. A substitute is needed. Bandura speaks of proximal mastery markers: countable acts, attributable to oneself, that serve as concrete proof that one has acted as someone who does this work.

If one takes seriously what the previous article argued, this proof is not merely a psychological artifact. It is indirectly linked to what is really at stake. Presence, as we saw, is what keeps the text alive in the practitioner's mind. The more regular it is, the more forcefully the text lives there. The more forcefully it lives there, the more it evolves — in the mind, before evolving on the page. The trace of presence is therefore not just a substitute for missing progress. It is, in its own way, the trace of real progress: not that of the manuscript, but of the inhabitation that is deepening.

Duration and Density

Among the traces writers leave of their practice, two main forms stand out.

The most universal is the temporal trace: the time slot. Thomas Mann wrote from 9 to noon behind a closed door, while Toni Morrison worked before dawn to carve out space before the household woke. Murakami, for his part, kept five to six hours every morning, and Graham Greene a fixed schedule he never exceeded. All impose a block of time, and the block itself serves as a trace: I was there, from this hour to that one.

This trace has a long history. Benedictine monks divided their day into fixed hours — the horarium — and practiced lectio divina at set time slots. Historians Jean Leclercq and Mary Carruthers read into this not simply an organization of time, but a psychotechnique against acedia, the particular anxiety of the monk facing the uncertainty of salvation⁴. The temporal ritual produced nothing measurable; it did something else. It cut ordinary time — entropic, given to dissolution — into ritualized segments renewed daily through repetition. In the vocabulary of Mircea Eliade, who formalized this opposition in The Sacred and the Profane⁵, the rite introduces a rupture in profane time: it restores, through repetition, a form of presence that would otherwise dissolve. Contemporary sociology has taken up the idea. Anthony Giddens, for whom routine protects against existential anxiety, is perhaps its clearest heir⁶. The monastic schedule, Mann's time slot, Murakami's daily session: same gesture, same function — holding presence against what, without it, would crumble.

Some writers stop at this temporal trace, and it suffices. Three hours, five hours, an entire day, with no number attached: if the slot has been held, presence occurred, and that is enough.

Others add to the temporal trace a second layer: a word count. Trollope was the most methodical of all, holding himself to 250 words per quarter-hour and checking his watch against the page. Stephen King aims for 2,000 words a day, Graham Greene noted 300 to 500 words in the margin of his manuscript, and Murakami, again, fixes his daily goal at ten Japanese pages (roughly 1,600 words), echoed by Jack London's 1,000 words a day. This number does not replace the time slot — it sits within it. And what it captures is not a quantity of text produced for its own sake. It is a form of density of presence: in those three hours spent at the desk, how many words actually came out? A morning at 250 words and a morning at 50 words are two different presences, even if the duration is the same. The number, for those who use it, measures this difference.

Seen this way, temporal trace and numerical trace are not opposed. The second refines the first. Duration says that one was there. The word count adds information about the intensity of the presence that occurred. One is universal, the other optional — because duration alone is already a sufficient trace. Density is a useful precision for some; it is indispensable to no one.

Robert Boice, whose experiments were mentioned in the first article of this series⁷, had validated this point experimentally. His protocol for blocked academics set no word target: it scheduled brief daily sessions. The constraint was temporal. His subjects produced up to nine times more than those who waited for the urge. The first layer — duration, without prescribed density — was sufficient to produce the effect.

When the Trace Stops Being One

Everything then depends on the relationship the practitioner maintains with the number. As long as it remains a trace — a record of what took place — it serves its purpose. As soon as it becomes a target — a threshold to cross — its behavior changes.

Psychologist Jordan Etkin documented this in a series of six experiments published in 2016⁸. She asked participants to count their steps, their pages read, or their time spent on an enjoyable activity. Result: people who counted did more, recording more steps and more pages. But they enjoyed the activity less. Worse: when the measurement was removed, the appeal did not return. Pleasure had been displaced toward the number; measurement had consumed it.

Put simply: measuring an activity one enjoys tends to transform it into something resembling work. Attention shifts from I am writing to have I written enough?. Once this shift begins, it does not repair easily.

Economists know a neighboring phenomenon. Anthropologist Marilyn Strathern reformulated it in a sentence that became famous: "When a measure becomes a target, it ceases to be a good measure."⁹ Once a number counts for something beyond itself, it tends to distort. A writer aiming for 1,000 words may end up padding their sentences. They write for the counter, not for the text. The number then tells less about the density of presence than about the ability to produce under quantitative constraint — two distinct things.

The problem is therefore not measurement itself. It is what one asks it to be.

Descriptive, Not Prescriptive

If these traps were mechanical, no writer who counted their output would have lasted. Trollope should have broken down within ten years. He lasted thirty-nine. King continues. Greene counted until the end. There is therefore a how that allows the number to coexist with the practice.

The most useful distinction comes from Edward Deci and Richard Ryan, in their meta-analysis of 128 studies published in 1999¹⁰. They distinguish two kinds of feedback: informational feedback, which reports on progress without prescribing an outcome, and controlling feedback, which exerts pressure toward a specific target. Both can take the form of the same gesture. Writing down "today: 300 words" as one notes the temperature is the first. Writing "300 words out of 500 expected, failed" is the second. The number does not change. The relationship one maintains with it does.

Trollope himself illustrates this. His notebook did not set a target — it recorded. He writes in his autobiography¹¹: "If I had slipped into idleness for a day or two, the record was there, urging me on to redoubled exertions." The notebook was a mirror, not a judge. And crucially, Trollope admits, in the next sentence: "though of late I have become somewhat more indulgent to myself." Leniency on weak days was part of the system. A numerical target can coexist with a healthy practice, provided the number remains descriptive rather than prescriptive, and variations are accepted.

Phillippa Lally, who tracked habit formation over 84 days¹², gives this flexibility an empirical foundation. In her study, missing a single opportunity to perform the behaviour had no significant effect on the habituation process. The brain does not reset the counter. When rigid, the mental representation of the target transforms a missed day into a structural failure — whereas, behaviourally, that day broke nothing.

There remains a subtlety that makes temporal measurement, in practice, the more robust of the two forms. First, it is more natural to set a time slot ("I'll be at my desk from 9 to noon") than to set an effective production duration ("I need to have been typing for 180 minutes today"). Time measurement is used spontaneously as a container, rarely as a threshold to reach. Second, what is measured — presence at the desk — roughly corresponds to what the practitioner actually controls. The density of a session depends on too many variables: the difficulty of the passage, the state of the manuscript, one's form on the day. Presence, on the other hand, depends mainly on the decision to sit down.

What This Changes

Measuring one's writing is therefore not measuring one's progress. It is making visible, in a practice that returns nothing immediately, the fact of having stayed in it. And, by extension, the fact of having sustained the inhabitation of the text that the previous article identified as the real stake.

Seen this way, the order emerging from the research and documented examples is not the one most often heard in writing advice. The foundation is not the number. It is the temporal trace: a time slot, a place, the fact of having returned. The number, if one insists on it, comes only afterwards, and preferably in descriptive mode: a mirror of the practice, not a judge of production.

The question the writer has often learned to ask — how much should I write today? — also changes in nature. A more accurate question precedes it: does what I measure help me see that I am here, or does it accuse me of falling short?

There remains a form of trace that has barely been touched on here, and that goes beyond the scope of this article. Woolf kept a journal to stay in the presence of her books outside the manuscript. Dorothea Brande and Peter Elbow, mentioned in the second article, proposed writing each morning without a target: no words, no duration, no subject. This form measures nothing. It is itself the trace. And it does something very particular to those who practice it. That is the subject of the next article.

Notes