Working Complexity

Long-form thinking
on complex systems

These pieces are mine to own. The ideas, the positions, the arguments. The writing is a genuine collaboration with Claude — not assistance, not generation, but a working practice I find worth naming. In a field that values honesty about how knowledge is made, that seems the right place to start.

Published

Managing the Soft Drift

A three-part argument on chronic disease, AI, and the architecture of change in healthcare systems built around the wrong event. Why health systems are structurally blind to the slow transitions that matter most — and what it would take to change that.

Read the series →

Three parts

  • 01 Managing the Soft Drift — a reframing of care trajectories in chronic disease — start here
  • 02 The Attentive Machine — how AI can serve the soft transition, and the conditions that make it possible
  • 03 The Crooked Road — change vehicles, stuck systems, and where the soft transition might actually be managed first

Experiments with Uncertainty

A two-part exploration of what not-knowing really means in healthcare — and what to do about it. First, a scholarly analysis through four patient personas across the ILD journey. Then, a rigorous advisory board stress-tests every finding and updates its positions in real time.

Read the series →

Two parts

  • 01 Living with Not Knowing — five types of clinical uncertainty mapped through four patients, and a WorkingComplexity way forward — start here
  • 02 The Board Responds — six healthcare leaders react, debate, and update their positions on uncertainty, AI, and who the system is designed to serve

After the drift:
a wayfinding.

The Polynesian navigator does not know which wave comes next. She knows how to read the ocean. A synthesis of all five bodies of work — soft drift, clinical uncertainty, the certainty complex, what other traditions knew — and the regenerative principles that could guide better care. Not a roadmap. A bearing.

Read the essay →

Draws on

  • 01Managing the Soft Drift — event-based blindness and the soft transition
  • 02Experiments with Uncertainty — five types, four patients, the system's limits
  • 03The Certainty Complex — politics, medicine, and the same structural pathology
  • 04What the West Forgot to Carry — Polynesian navigation, Talmud, indigenous fire
  • 05Principles of Regenerative Economics — the frame for what better looks like

Heuristics in complex systems

When do simple rules outperform complex models? A journey through medicine, markets, and organisational design. A complete taxonomy of why heuristics work, when they fail, and how to preserve their wisdom as environments change.

Read the essays →

Three essays

  • 01 When Simple Rules Beat Complex Models — seven worked examples from medicine and finance — start here
  • 02 Taxonomy Edition — four types of heuristics and why they fail. Eleven examples. The full theory.
  • 03 Knowing When to Let Go — detecting decay, preserving wisdom, and what AI changes about heuristics

Personal
Personal essay · Why complexity · June 2026

I lost something. I also gained something.

A personal note on a lifelong attraction — from a schoolboy's disappointment that Newton's laws were "too simple", through Cynefin and complex adaptive systems, to the question that matters most in health care: what matters to you? Less a position paper than the reason behind them.

Read →

Essays & analysis
Essay · Care ethics & systems · June 2026

Noticing is not neutral. Attentiveness is the first act of care.

The soft drift series argued that health systems are structurally blind to slow transitions. Joan Tronto has been making the same diagnosis for thirty years — as a failure of attentiveness, the first phase of care. On billing codes as moral documents, AI that notices without owning what it notices, and the unpaid attentiveness workforce the event-based economy free-rides on. The essay that joins our complexity argument to her ethics of care.

Read →
Position paper · ILD & Complexity Science · June 2026

Complex by diagnosis. Reductionist by measurement.

We agree interstitial lung disease is complex — then track it with the simplest instruments we own. FVC as a scalar, the single endpoint, the composite that only aggregates reductions. A position paper on why reductionist measurement is the field's primary epistemology, and the complexity-science toolkit — early-warning signals, symptom networks, joint models, digital twins — that could change it.

Read →
Essay · Power & Epistemology · May 2026

The certainty complex. How politics and medicine make the same mistake.

Rainer Kattel argues that those who most desire power seek it to impose certainty on an uncertain world. Healthcare systems do the same — and suffer the same consequences. Two bodies of thinking, one diagnosis.

Read →
Essay · Uncertainty & Economics · May 2026

The price of not knowing. What we cannot buy down.

Value of perfect information is one of health economics' most useful concepts. Applied to progressive lung disease — and complex chronic medicine more broadly — it encodes assumptions that do not hold, and creates research priorities that systematically price uncertainty wrong.

Read →
Essay · Epistemology & Borrowed Wisdom · May 2026

What the West forgot to carry.

The Polynesian navigator, the Talmudic scholar, and the indigenous fire manager have been working with genuine complexity for millennia. Not as alternatives to rigour — as embodiments of it. What their structural features mean for healthcare design, and whether Western institutions are ready to receive them.

Read →
Synthesis · Systems & Practice · May 2026

Three arguments, one problem.

Hard transitions, clinical uncertainty, and the value of information are not three separate ideas. They are the same structural failure, seen from different angles. This essay maps how they connect — and names five practical moves any health system can make now, without waiting for a transformation programme.

Read →
Essay · Regenerative Economics · 2026

Do we need to rewild healthcare?

Regenerative economics proposes that health systems should operate more like living ecosystems and less like supply chains. Drawing on experiments in the US, China, Japan and Germany — and asking what legal architecture, evidence and AI need to contribute — we examine whether the UK might be among the first to find out.

Read →

Reference
Reference · AI-developed personas · 10 profiles

The cast. Citizens and professionals.

The analytical instruments behind the work. Four citizen personas — Harold, George, Francis, Victor — and six healthcare advisory board members, each with a committed position, a professional history, and a specific kind of uncertainty. None are real. All are true.

View the cast →

Coming next
Essay · Forthcoming

The Payer Problem

Why the most consequential AI in healthcare is being built for insurers, not clinicians — and what that means for who benefits.

Coming soon
Analysis · Forthcoming

Singapore's Experiment

Healthier SG is the world's most explicit attempt to build relational continuity into a national primary care system. What three years of data shows.

Coming soon
Argument · Forthcoming

What McKinsey Gets Wrong About Complexity

The conventional consulting approach treats system transformation as a complicated problem. It isn't. The category error is structural — and consequential.

Coming soon