Progression: Finding Your Subject

Finding a subject is less like inventing something from nothing than noticing what has held your attention for years. In “Progression,” John McPhee looks back across his work and finds that most of his subjects connect to interests he had before college: sports, wilderness travel, canoes, geology-adjacent landscapes, and people with unusually deep knowledge.

Sources: “Progression,” The New Yorker and the official Draft No. 4 book page.

That observation does not mean a writer should remain trapped in childhood. It means durable curiosity is evidence. A subject you have returned to repeatedly already has emotional energy, vocabulary, memories, and unanswered questions attached to it. Those resources make it easier to recognize a promising angle and harder to quit when reporting becomes slow.

The sequence matters. For most nonfiction, the writer first identifies a subject, reports it until there is substantial material, and then discovers a structure that fits what was found. Beginning with a clever form and searching for facts to fill it is possible, but McPhee presents it as an unusual exception rather than a dependable method.

A durable interest is a topic, activity, place, problem, or kind of person that keeps drawing you back without external pressure. Making that interest visible is the first practical step because vague self-knowledge rarely produces a precise reporting plan.

Start with evidence from your life rather than aspirations about what a serious writer ought to cover. Review old bookmarks, photos, notes, hobbies, recurring conversations, jobs, journeys, arguments, and questions. Look especially for subjects that have survived changes in fashion or career.

Evidence to inspect	Question to ask	Possible subject seed
Activities repeated for years	What keeps me practicing after the novelty fades?	The hidden culture of a local climbing wall
Places revisited	What changes here, and who notices first?	A river crossing through four seasons
People you seek out	What knowledge do they possess that others miss?	The diagnostic habits of bicycle mechanics
Frustrations you keep naming	What system produces this recurring problem?	Why public forms are hard to understand
Childhood fascinations	What adult institution now surrounds this interest?	The economics behind amateur astronomy

Exercise — the ten-and-three inventory: Write ten durable interests without judging their importance. Circle three that contain a person, place, conflict, process, or change you could observe directly. For each circled item, finish the sentence: “I keep returning to this because I still do not understand …”

The result is not yet a pitch. It is a personal map of promising territory. A subject becomes useful only after you connect private interest to something a reader can enter.

A private fascination becomes a nonfiction subject when it contains a question that reporting can answer and a consequence another person can care about. “I love maps” is an interest; “how volunteer mapmakers update a city during a flood” is a subject with actors, action, stakes, and a route into the world.

Use four tests to move from fascination to subject:

Test	Weak version	Stronger version
Specificity	“Food systems”	“How one school kitchen replaces packaged lunches”
Reportability	“Why people fear change”	“How veteran nurses adopt a new handoff protocol”
Movement	“A portrait of a station”	“A station during its final month before closure”
Stakes	“A collector’s unusual hobby”	“The collector preserving records a town is discarding”

A useful subject is bounded but not exhausted. It is narrow enough to report in the time available and open enough to surprise you. If you already know every conclusion, you have an argument to illustrate rather than a subject to investigate.

Write a provisional subject statement in this form: I will follow [person, place, system, or process] while [change or pressure] reveals [larger question]. Treat it as a compass, not a contract. Reporting should be allowed to revise it.

Subjects often progress rhizome-like: a rhizome is a root system that spreads sideways and sends up new growth at multiple points. McPhee describes ideas emerging from relationships, places, or questions encountered while reporting earlier work. The next subject may be a person at the edge of the current story or a question the current assignment cannot contain.

This growth is not mere recycling. Each new piece needs its own center of gravity. A guide encountered during a river story might lead to a profile, but the profile must ask something that the river story did not. The link supplies access and informed curiosity; it does not supply the finished angle.

Maintain a sideways notebook while reporting. Reserve one page for people, claims, places, and tensions that matter but do not belong in the current piece. Record why each item caught your attention and what first reporting move would test it. This preserves possibility without overcrowding the work in front of you.

Reporting is not only a way to collect facts after choosing a subject; it is how you discover whether the subject can sustain a piece. A promising idea should become more dimensional as you speak with people, observe action, read documents, and test your assumptions.

Before committing months, run a small reporting probe:

1. Speak with one participant who knows the subject from inside.
2. Observe one real setting where the subject becomes visible.
3. Read one primary document, record, manual, dataset, or archive.
4. Identify a knowledgeable person who disagrees with your first source.
5. Write down three things that surprised you.

Signal after the probe	What it suggests	Next move
Sources give only general opinions	The subject may lack observable action	Find a process, event, or decision to follow
Every answer confirms your premise	Your question may be too leading	Seek a rival explanation or counterexample
One detail opens several new questions	The subject is generating material	Follow the most consequential branch
Access is impossible	The subject may be sound but unavailable	Change the vantage point or narrow the scope
Stakes remain invisible	Readers may not know why it matters	Locate who bears the cost of the outcome

The goal is not to prove that your initial idea was correct. It is to learn whether reality pushes back. Resistance, contradiction, and surprise are often signs that the subject has enough depth.

Material is the body of usable reporting: scenes, facts, documents, descriptions, quotations, chronology, contradictions, and observed behavior. McPhee’s normal progression places this material before structure because a nonfiction form must be accountable to what actually exists.

If you impose a form too early, you begin rewarding only the facts that fit it. A neat three-part argument can make you ignore a fourth force that the reporting reveals. A planned heroic arc can cause you to flatten a source whose motives are mixed. Structure is essential, but it should organize complexity rather than pre-empt it.

At the same time, “report before structure” does not mean report forever. Define a stopping condition for the first collection phase: a date, a set of essential voices, or enough material to answer the central question from competing angles. Then reread the whole body before deciding how to arrange it.

A structure-first project begins with a formal relationship and then seeks subjects that can inhabit it. McPhee recalls rare works that started from abstract patterns, including a paired-profile concept and the multi-part relation he rendered as ABC/D. These examples are memorable partly because they were departures from his usual practice.

Starting mode	Use it when	Main danger
Subject first	A person, place, event, or process already compels inquiry	Gathering without eventually selecting
Material first	You inherited an archive or witnessed events before knowing the story	Treating everything as equally important
Structure first	The formal comparison is itself the genuine question	Forcing reality to fill predetermined slots

The discipline is to name which mode you are using. If the structure comes first, treat each slot as a reporting hypothesis. Be willing to abandon the pattern if the subjects become distorted by it. Formal elegance is never permission to misrepresent evidence.

Decision check: Explain your intended form to someone in one sentence. Then explain the subject without mentioning the form. If the subject sounds empty without the device, gather more reporting before committing.

A subject workshop converts reflection into a next action. It is useful when you have many interests but no assignment, or when every possible idea feels either too broad or too familiar.

Minutes	Action	Output
0–10	List ten durable interests	Unfiltered personal territory
10–20	Add a person, place, process, change, and conflict to each promising item	Concrete entry points
20–30	Choose three and write a public-facing question for each	Three provisional subjects
30–40	Score access, surprise, stakes, and scope from 1 to 5	A reasoned priority
40–50	Design a five-step reporting probe for the leader	Immediate fieldwork
50–60	Record two possible rhizome branches without pursuing them	A protected future-ideas list

End by scheduling one action that touches reality: request an interview, visit a location, retrieve a record, or observe a process. More brainstorming cannot substitute for contact with the subject.

After the probe, write a one-page memo with four headings: what I assumed, what I found, what surprised me, what I will report next. That memo makes progression visible and prevents the original idea from quietly controlling every later decision.

The practical lesson of “Progression” is that subjects develop through sustained attention rather than sudden inspiration. A writer notices durable interests, converts one into a reportable public question, gathers material, and only then chooses the structure most faithful to the evidence.

• Durable interests are evidence of where you already possess energy and sensitivity.
• A subject needs specificity, access, movement, and stakes.
• Earlier reporting can generate new subjects sideways, like a spreading root system.
• Reporting tests the subject and should be allowed to change the premise.
• In the usual sequence, subject and material precede structure.
• Structure-first projects are legitimate exceptions that require extra care.

Use this checklist before committing to a long piece:

• [ ] I can name the durable interest beneath the subject.
• [ ] I can state the subject as a question, not only a topic.
• [ ] At least one person, place, process, or change makes it observable.
• [ ] I know why a reader beyond my immediate circle might care.
• [ ] I have designed a small reporting probe that can surprise me.
• [ ] I am recording promising side branches without crowding the current piece.
• [ ] I will gather and reread material before locking the final structure.
• [ ] I can abandon an elegant form if the evidence does not support it.