History of the 2-Axis Spectrum: From 1789 to the Nolan Chart

The left-right line was an accident of French seating arrangements. The second axis took a century and a half to arrive, and it came out of psychology rather than political theory.

Écrit par Ottavio Malatacca · Publié 2026-06-08 · Mis à jour 2026-08-30

Every political vocabulary carries the fingerprints of the circumstances that produced it. The left-right distinction, now so natural that it feels like a fact about the world rather than a metaphor, originated in a seating arrangement in a single room in 1789. The second axis that turns that line into a compass arrived a hundred and sixty years later, and it came from experimental psychology rather than from political philosophy. Knowing that history makes clear both what the model captures and where it should not be trusted.

1789: an accident of furniture

In the French National Assembly of 1789, delegates debating the extent of the king's veto arranged themselves by conviction. Supporters of royal authority and the established church gathered to the president's right; opponents, who favoured radical constitutional change, gathered to the left. The arrangement was practical — deputies wanted to sit with allies and avoid being shouted at — and it was initially regarded as a regrettable sign of faction.

It endured because it encoded something real about that specific dispute: a single dimension running from preservation of the existing order to its transformation. The vocabulary spread through nineteenth-century European parliaments, and by the time of the mass suffrage era it had been thoroughly reinterpreted. The axis was no longer about the monarchy but about the economy, with left meaning collective provision and redistribution and right meaning private property and market allocation.

That reinterpretation is worth noticing, because it shows that the terms were never anchored to fixed content. They mean whatever the dominant cleavage of the period makes them mean, which is why arguments about whether a given position is really left or right are usually unresolvable.

The problem the line could not handle

The single axis broke down under the pressure of twentieth-century experience, and the breaking point is easy to identify. Placed on a left-right line, Stalin's Soviet Union and Hitler's Germany sit at opposite extremes. Anyone who had lived under either recognised that they shared a great deal: a single party, an omnipresent security apparatus, the suppression of independent civil society, a leader cult, and the subordination of the individual to a collective project.

The single axis also could not distinguish between the Scandinavian social democracies and Soviet state socialism, which shared an economic direction and differed completely on political liberty. Nor could it place a nineteenth-century classical liberal, who favoured free markets and opposed both established religion and state censorship, and who consequently registered as right-wing and left-wing at once.

By the late 1940s these were not academic puzzles. Explaining how mass support for totalitarian movements had been assembled in literate, industrialised societies was among the most urgent intellectual tasks of the period.

Eysenck: the second axis arrives from psychology

Hans Eysenck's The Psychology of Politics (1954) applied factor analysis to survey responses on political and social attitudes. The technique asks a purely statistical question: how many underlying dimensions are needed to account for the observed pattern of correlations between answers?

The answer was two, not one. The first factor was the familiar economic radical-conservative dimension. The second, which Eysenck labelled tough-minded versus tender-minded, was independent of it and tracked attitudes toward authority, punishment, discipline and conformity. Crucially, tough-mindedness was found at both economic extremes: fascists and communists scored similarly on it and differed sharply on the first factor.

This is the origin of the modern compass, and its origin is significant. The second axis was not proposed because a theorist found it philosophically appealing. It emerged from the data as the smallest number of dimensions that could account for how people's attitudes actually clustered.

Eysenck's work attracted sustained criticism, some of it justified. His samples were unrepresentative, his tough-minded factor may have been partly an artefact of item selection, and his later research on other subjects damaged his general credibility. Milton Rokeach argued in The Open and Closed Mind (1960) that the relevant variable was dogmatism — the structure of belief rather than its content — and that closed-mindedness appeared across the political spectrum. The specific labels remain contested. The two-dimensional finding has been replicated many times with better methods.

Nolan: the diagram, from a political movement

David Nolan, a founder of the American Libertarian Party, published his chart in 1971 with an explicitly political purpose. He argued that the conventional single axis made libertarianism unintelligible — a position combining economic freedom with personal freedom simply had nowhere to sit.

His solution used economic freedom and personal freedom as two axes of a square, rotated forty-five degrees so that the conventional left and right occupied the side corners while libertarian and authoritarian positions occupied top and bottom. The diagram made a case as much as it made a measurement: it presented libertarianism as a coherent quadrant of equal standing rather than an eccentricity.

This dual character — analytical tool and rhetorical device — is present in every version of the compass since, including ours. Where the axes are placed, how the questions are worded, and which positions end up near the centre are all choices, and different defensible choices produce different results for the same respondent. Any test that presents its output as a purely objective discovery is misrepresenting what it does. Our methodology page states our own weighting decisions explicitly for that reason.

From chart to internet phenomenon

The compass reached a mass audience through the web. The best-known implementation appeared in 2001, using an economic left-right horizontal axis and a social authoritarian-libertarian vertical axis, and the visual grammar it established — four coloured quadrants, a single plotted point — is now near-universal.

Its popularity brought recurring criticisms that any user of such a test should know:

• Question wording carries assumptions. Agree-disagree items frequently embed a premise, and a respondent who rejects the premise has no accurate way to answer.
• Scoring is rarely disclosed. Most public tests do not publish their weightings, so the placement cannot be independently checked or reproduced.
• Results cluster in one quadrant. Several widely used tests place the large majority of respondents in the libertarian-left quadrant, which suggests calibration rather than a genuine discovery about the population.
• Two dimensions are a choice, not a finding. Some factor analyses support three or more dimensions, and the emerging cosmopolitan-communitarian cleavage fits awkwardly onto either existing axis.

Alternatives worth knowing

The two-axis compass is not the only serious instrument. The Political Compass approach shares a lineage with academic expert surveys such as the Chapel Hill Expert Survey, which positions parties rather than individuals using coded assessments from country specialists. The Manifesto Project takes a different route again, coding the actual text of party programmes sentence by sentence to derive positions from documents rather than from opinion.

These academic instruments are more rigorous and considerably less accessible. A public test occupies a different niche: it is an entry point into political self-examination rather than a research instrument, and it should be judged on whether it prompts more careful thinking rather than on whether it produces a publishable coordinate.

Conclusion

The left-right line survived from 1789 because it captured a real cleavage, and it became misleading when the cleavage it described stopped being the only one that mattered. The second axis was added because the first could not distinguish arrangements that people living under them could tell apart immediately. Two axes are demonstrably better than one. Whether two are sufficient for the politics of the 2020s — with sovereignty, technology and ecological questions cutting across both — is an open question, and one worth keeping in mind while reading any result, including the one this site produces.