Introduction


How We Talk About Work Is How We Work

All work follows a story. We work, consciously or not, according to a narrative that describes what we do, why we do it, and what it accomplishes. There are two different types of work in a basic narrative — work that we think needs to be done, and the effort to figure out what to do.

Most narratives focus on the first type, work which applies directly to a problem. Given a task that takes longer to understand than to accomplish, we will often lament how much time we wasted getting to the “real work.” However real or fraudulent work may feel, we define it by what we can understand of it.

Nobel physicist Richard Feynman was fond of telling a story from his youth in which a third, important type of work emerges from the whole process of picturing, doing, and understanding work. We will call this kind of work conceptual labor.

As an inquisitive child in the Great Depression, Feynman turned his early fascination with electronics into a side-hustle fixing radios. One day he was confronted with a radio that made a terrible noise he hadn’t heard before. He had to really think about what could be wrong, pacing back and forth while modeling the radio. Suddenly the owner of the radio shouted at him, complaining that he was just walking around, doing nothing. Young Feynman shot back “I’m thinking!” Once Feynman fixed the radio, the previously impatient owner became his biggest fan. He found Feynman many more jobs while bragging up his ability — “He fixes radios by thinking!”


Looking at the Whole Problem is a Different Type of Work

Feynman told this story not to talk about radios, but to talk about a way of working and thinking. The first section of his memoir, Surely You’re Joking Mr. Feynman, opens with the radio story and flows into examples of how the same kind of work applies to social situations, public education, and math, at one point connecting the FORTRAN language to how he repaired mechanical typewriters. In each case, Feynman’s infectious, joyful curiosity was sparked by “the whole problem of discovering what was the matter, and figuring out what you have to do to fix it.” He even invented his own system of mathematical notation because the conventional terms and symbols didn’t satisfy his needs. And why shouldn’t he? If doing math was a way for Feynman to engage the whole problem in front of him, the properties of math itself are fair game.

By asking us to look at the whole process of working on a complex problem, Feynman suggests a fundamentally different way of working that cannot be fully described by a back-and-forth between the two basic types of work — Type 1 being direct, understood work and Type 2 being the effort to understand that work. Rather than striving to find a clear narrative and stick to it, in this work one has to continuously create potential narratives of work while adapting to the circumstances in which the work occurs.

This conceptual labor integrates Type 1 and Type 2 work with a third dimension of skills and practices. These include the labor to define what work is, to learn how it must be done, and to invent new ways of working. These skills interact with technical knowledge, but extend beyond the limits of specialized training. After all, the radio owner was surprised by how fixing worked, not how the radio worked.

He followed the conventional narrative of labor, in which we judge work by available knowledge and prioritize Type 1 work. This narrative of work tells us that writing a novel is more like reporting than it is like driving a race car because of how these activities and their results look from the outside. Feynman didn’t see things this way, and he was freer for it.

Conventional narratives can serve us well, but they fail us all the time, often at great cost. At some point, everyone has done work which felt productive at the time only to discover that it had actually accomplished nothing. The feeling of productivity came from following the narrative, not meeting the actual requirements of the task. When this happens, it is easy to believe that we had the wrong instructions, and the right ones are out there somewhere. If we could just find them, we’d be able to follow them faithfully to the solution.

It is much harder to consider that some work has no straightforward story. Work that occurs in the absence of a narrative is confusing, demoralizing, and difficult. (Unless we are allowed to make it into play.) Regardless, some work can only be understood as one does it, some work continuously changes so it will never be fully described, and some work must be invented as it occurs. Such work frequently involves other people, learning something new, or understanding the behavior of a complex system. When we attempt to apply conventional narratives to this kind of work, we create new problems and fail at our original task.

Expertise Requires More Than Conventional Narratives Can Offer

The unique value of the work done by experts like Feynman comes from their ability to navigate, understand, and describe this kind of work1, let alone execute it.

Everyone is an expert in something, even if it is only where to find things in their kitchen. At some point we have all struggled to explain, to ourselves or to others, why the job that non-experts think we are doing is completely different than the real, tricky thing we are trying to accomplish, even if it looks similar from the outside. We have no guarantee that the conceptual labor we have done to make our expert work possible will be visible to observers, and it is difficult to remember how a novice sees the task. Likewise, we have all been confronted by a complex task that, when prepared and executed by an expert, looks like child’s play.

We cannot understand expertise or innovation without conceptual labor. Painter James McNeil Whistler was highly influential on the development of modern painting, but his work was once ridiculed for its simplicity by critics who preferred classical technique. One such review led to a libel trial, where he famously defied the idea that the value of a painting was tied to the time spent painting it. When pressed if he asked market prices for “the labour of two days”, he responded: “No, I ask it for the knowledge I have gained in the work of a lifetime.”

A much more recent anecdote from economist Dan Ariley about a locksmith who lost wages as he gained mastery shows that the difficulty of putting a price on wisdom isn’t just a problem for professional artists. When the locksmith was still a novice, he would struggle with locks for hours at a time, sometimes breaking them. His customers waited and watched the whole process, and tipped him well for all of his hard work, even if they had to replace their lock. Once he became an expert, however, he could open locks in a fraction of the time, with apparent ease and without damaging them. Not only did this cause his customers to tip him less, but they started complaining about his base fee, even though they were getting a more valuable service. This locksmith had encountered his profession’s version of a refrain that painters have been hearing since at least Whistler’s time: “Well I could do that.

Conceiving of Work Is Work

The point here is that simply conceiving of work can be difficult. When we encounter work that lies beyond our imagination, we still judge it within the limits of our current understanding. Novices only see the work once it has been conceived, and without the domain knowledge, experience, and judgement of an expert, they fail to see how the work must be continuously re-conceived as it is performed. They don’t appreciate the work it took to produce what’s in front of them at any given moment because, in part, they can’t see it without doing at least some of the work itself.2. Conceptual labor creates the very ground on which it is performed.

This is why it’s not hard find experts who advocate for a new way to consider their work, regardless of discipline, that focuses on the conceptual labor involved.

A good starting point is any time someone uses the phrase “the art of,” such as Richard Hamming’s The Art Of Doing Science and Engineering. Lewis Thomas’s essay The Planning of Science addresses conceptual labor in the biomedical field, Zadie Smith calls writing the “craft that defies craftsmanship” in her excellent essay Fail Better, Bret Victor’s talk Media for Thinking the Unthinkable challenged coders and designers to deconstruct the media that encodes their ideas, Chris Argyris’s Teaching Smart People How To Learn brings it to the field of corporate management, and in math we have Paul Lockhart’s A Mathematician’s Lament. Rebecca Solnit’s profound and surprising tour around the territory of this way of thinking, A Field Guide to Getting Lost, digs into a challenge posed by Meno to Socrates: “How will you go about finding that thing the nature of which is totally unknown to you?”

This, surely, is hard work, but can we recognize it when we and others are doing it? To paraphrase Burton Rascoe, the spouses of writers will never understand that they are hard at work when staring out the window.

The Endless Argument

We have all been the writer, and we have all been the spouse, often both during the same project. As we produce and examine narratives of our own work, we are trying to see more than we can see.

Seeing the narrative of work for what it is rather than what we imagine it to be is the true difference between expert and novice. According to the novice, experts work beyond the boundary of confusing work, where all the work makes sense to those that know enough about it. The expert, however, knows that they will continually lose sight of the work as it evolves, and they have the skills and attitudes that let them move back and forth across that boundary of understanding.

We will never fully appreciate the work we are trying to invent before we have invented it. The disagreement between an objecting, impatient novice and an expert trying to explain the transformative nature of their work plays out in our own heads as we do difficult, conceptual labor.

Conventional narratives of work and the strategies that go with them assume that this will be an argument, and emphasize determination, persistence, and knowledge. These are fine qualities in the abstract, but also how you break a lock. A general theory of conceptual labor should provide rules of order to this argument so that it can become a lively conversation, full of productive debate and surprises.

> The Theory

  1. Feynman prided himself on being able to explain complex ideas simply, at one point declaring that, since he wasn’t able to describe an aspect of quantum mechanics to a freshman class, it was not fully understood by science. David L. Goodstein relates this anecdote in his introduction to a lecture by Feynman in which he used only high-school geometry to describe planetary motion. ↩︎

  2. Daniel Kahnamen’s research into how we frame our decisions based on available knowledge provides thorough support for this assertion. After establishing the influence of framing on problem solving strategies through years of studies, he and his contemporaries now refer to the phenomenon with the acronym WYSIATI — what you see is all there is. Citation Likewise, George Lakoff’s research addresses the effects of linguistic framing on the meaning of statements. There has been a resurgence in interest in his work since the 2016 US election, as it describes many of the mechanics behind propaganda and messaging used by politicians and the media. Citation ↩︎