Map and Territory

Writing in 2019, Michael Batty traces “the map is not the territory” to a 1931 paper by Alfred Korzybski.¹ A map of London is not London. A model is not the system it represents.

This seems obvious. It isn’t. The failure mode is constant: optimizing the map instead of the territory, mistaking model outputs for reality, forgetting that abstraction loses information.

Maps are useful because they’re simpler than territory. A 1:1 scale map of London, including every atom, would be London — and useless for navigation. Abstraction is the point. The question is which details to keep and which to discard.

Every map makes choices. Road maps show roads, not buildings. Topographic maps show elevation, not roads. Political maps show borders that exist only in human agreement. The mapmaker’s purpose determines what appears and what vanishes.

The same territory supports many maps. Several can be accurate for their stated use while showing different facts. A road map can be right about roads and silent about flood risk. The error is asking a map to answer a question it was not made to answer.

Goodhart’s Law is a map-territory confusion. The metric was a map of what you cared about. When you optimized the metric, you changed the territory in ways the map couldn’t see. The map diverged from what it was supposed to represent.

Long-Term Capital Management is a useful case with a common bad summary. Russia stopped payments on some debt in August 1998. The market shift then pushed LTCM toward collapse. Fourteen private firms supplied $3.6 billion in a deal arranged by the Federal Reserve Bank of New York; the Federal Reserve did not invest its own funds.² Russia defaulted. LTCM did not.

Science makes models. Newtonian mechanics remains a good approximation for many objects moving well below the speed of light. Relativity and quantum mechanics cover cases where its assumptions fail. A later theory can limit the domain of an earlier one without making the earlier calculations useless.

All scientific theories are maps. They compress reality into usable form. They work until you find territory they don’t cover. Then you need a new map.

The practical discipline: notice when you’re looking at a map instead of the territory.

Financial statements are maps of a business. The P&L abstracts away customer relationships, employee knowledge, technical debt. The map says the company is profitable. The territory may be rotting.

Resumes are maps of people. The resume shows credentials and job titles. The territory includes judgment, integrity, how they act under pressure. Hiring on the resume alone is navigating by a map that omits most of what matters.

Words are maps of concepts. The word “love” maps a vast territory of experiences, relationships, and feelings. Two people using the same word may be pointing at completely different territories.

You cannot reason without some model. State what a model leaves out, test it against observations, and change it when the misses form a pattern.

Go Deeper

Books

Science and Sanity by Alfred Korzybski (1933) — A fuller account of his system.
Language in Thought and Action by S.I. Hayakawa — Accessible introduction to general semantics including map-territory.
The Structure of Scientific Revolutions by Thomas Kuhn — On how scientific maps (paradigms) change.

Essays

Jorge Luis Borges, “On Exactitude in Science” — One-paragraph parable of a map that achieves 1:1 scale.
Gregory Bateson’s “Form, Substance, and Difference” — On the map-territory distinction in systems thinking.

Fiction

Lewis Carroll’s Sylvie and Bruno Concluded (1893) — Contains the idea of a map at 1:1 scale, predating Borges.

Sources

Michael Batty, “A map is not the territory, or is it?”, Environment and Planning B: Urban Analytics and City Science 46, no. 4 (2019), 599–602. ↩
Michael Fleming and Weiling Liu, “Near Failure of Long-Term Capital Management”, Federal Reserve History, 22 November 2013. ↩