Almost everyone who tries to do deep work fails the same way, and the similarity should long ago have raised a suspicion: perhaps the fault is not in the people. You set aside three hours on Wednesday morning. At 9.10 someone needs a quick answer, and the answer genuinely is quick. At 9.35 you reread the last paragraph and discover a stranger wrote it. By noon you have produced forty minutes of something and a far more durable feeling of having failed a test you set yourself.
The usual explanation is a lack of discipline. The less comfortable one — and, I think, the more accurate — is that three uninterrupted hours are not the normal standard for an adult with a job, messages and a life; they are a laboratory condition someone converted into a moral yardstick.
What Cal Newport is actually proposing
Deep work, in Cal Newport's 2016 book, is a well-bounded thing: focusing without distraction on a cognitively demanding task. He is a computer science professor at Georgetown, and his central claim is economic rather than motivational: the capacity is becoming rare at the moment it becomes valuable, so whoever cultivates it does well.
The claim is probably right, which is why the book unsettles rather than annoys. The diagnosis is not the problem. What was built on top of it is.
The yardstick was written by someone with an unusual calendar
A university professor has something almost none of his readers have: control over his own diary. That does not knock down the argument, it knocks down the silent assumption that the rest of us are simply being slack.
Look at the average day of an actual knowledge worker. In a field study of 24 professionals at a technology firm, the average time on one working sphere before switching or being interrupted was 11 minutes and 4 seconds, and resuming an interrupted task took an average of 25 minutes and 26 seconds, with two other activities along the way. Gloria Mark, who ran that research and spent two decades on it, later measured something worse: average attention on a single screen has fallen to roughly 47 seconds.
I offer the numbers not as an excuse but as scale. If your day is built of eleven-minute blocks, a target of a hundred and eighty is a sixteenfold jump — and we demand that of no other human capacity without calling it training.
What the evidence supports and what is work philosophy
Supported. Switching tasks is expensive. Sophie Leroy showed, in two experiments published in 2009, that someone who moves to a second task without finishing the first leaves part of their attention stuck on the previous one — attention residue, in her term — and performs worse on the new one. That is a finding about transitions, not duration: it recommends closing before opening, not three-hour stretches.
Partly supported. Csikszentmihalyi's flow, the absorbed state deep work promises. The experience is real, the measurement is a mess: a 2020 review found 24 different operationalisations of flow across 42 studies and noted that the commonest yardstick classifies 25% or more of all daily experiences as flow — far too many for a state described as rare.
Unsupported. Deep work as a method. I looked for a controlled study testing the whole protocol — long blocks, ritual, isolation — against some alternative. I did not find one. The method is a work philosophy resting on adjacent findings, which is no crime — it simply is not the same thing as evidence.
A version that fits a day you do not control
What is left once you drop the quota
What remains is what no adjustment of method solves: part of your day legitimately belongs to other people. A parent of a small child, an analyst on call, someone in treatment — they do not have three hours because their lives do not, and the book implying otherwise describes a life, it does not prescribe a method.
The honest version of the advice is modest: fewer switches, clean transitions, one protected item a day. It will work less well than you hoped and better than the alternative, which is going on charging yourself for a quota that was never yours. I would like to promise more and cannot.
Why this matters for what I built
I spent years treating my fragmented days as a character flaw, and all the guilt I piled up there produced not one line of deep work. What helped, when anything did, was having somewhere to set the interrupted thing down knowing it would come back.
That small piece is what I wanted in ASB: someone you tell where you stopped, who keeps it and brings it back on time — so your head does not hold the residue while it tries to work.