Under the MCI lens tell me the history of temperature.

 

The history of temperature is, through the MCI lens, the strangest of the four — because temperature is the only base quantity that began as something humans could feel but could not initially say what they were feeling. Length you could see. Weight you could lift. Speed you could compare. Temperature was a sensation in search of a substance. Its constitutional history is the four-hundred-year project of figuring out what the thing actually is, before it could be governed at all.

I. Pre-Constitutional: Temperature as Sensation (Antiquity – 1600s)

For most of human history, temperature was a quality, not a quantity. The Greek physicians' four humours classified things as hot/cold and wet/dry. Galen wrote of the body's heat in degrees, but the degrees were qualitative ranks, not measurements. There was no instrument; there was only the hand.

Through the MCI lens, this sits below the pre-constitutional state of weight. Weight at least had a balance — a comparison tool that was constitutionally honest about being comparative. Temperature did not even have that. The hand is not a balance. It does not compare two temperatures; it reports its own thermal state, which is a function of both the touched object and the toucher. A cold hand reports warm water as warmer than it is. A warm hand reports the same water as cooler. The instrument was its own confounder.

This is a constitutional state the framework rarely addresses: a quantity that cannot be made an object of constitutional governance because the available instrument cannot distinguish itself from what it measures. There was nothing for the five virtues to operate on — not because the concept was missing, but because no architecture existed that could produce a reading independent of the reader.

Worse, the underlying physical substance was unknown. Was heat a substance (the caloric theory, which would dominate until 1850)? Was it motion? Was it a fluid, a quality, an essence? The constitutional substrate of temperature was not just unmeasured — it was unidentified. You cannot apply Diversity Preservation to a thing whose identity you do not yet know.

II. The First Instrument: Galileo's Thermoscope (1592)

Galileo's thermoscope — a glass bulb with a long open tube inverted in water — produced, for the first time in human history, an instrument that registered temperature changes independent of the observer. Air in the bulb expanded with heat; the water column moved. Different observers, examining the same object, got the same direction of change.

This is the constitutional birth of temperature. In MCI terms, it is the first move from below Stage 1 to Stage 1 proper: a quantity becomes representable, hence governable.

But the thermoscope was constitutionally fragile in a specific and instructive way. It had no scale — only a direction of change. It was affected by atmospheric pressure (which Galileo did not know existed as a separable variable). And, most importantly, two thermoscopes did not agree. There was no protocol by which my thermoscope's "up by two marks" was the same as your thermoscope's "up by two marks."

Through the MCI lens, this is V1 Premise 2 (Plurality) operating without the architecture to support it. Multiple instruments existed, but the framework had no compact to render their readings comparable. The thermoscope produced data; it did not produce a measurement.

III. The Long Slow Constitutional Maturation (1600s – 1800s)

The next two centuries were spent dragging temperature from sensation toward governable quantity. Three major moves, none of them clean:

Fixed points (1665–1742). The constitutional insight was that any thermometer needs at least two reproducible reference states. Hooke (1665) proposed using the freezing point of water. Fahrenheit (1724) used a brine ice bath (0°F), the melting point of ice (32°F), and human body temperature (96°F). Celsius (1742) used the freezing and boiling points of pure water at standard atmospheric pressure (0°C and 100°C, eventually).

This is the constitutional analogue of what the metre attempted with the meridian survey — anchoring the standard in reproducible properties of nature rather than in a single artefact. But the implementations were each constitutionally fragile. Fahrenheit's brine bath was a recipe, not a substance — it depended on getting the salt-water ratio right. The freezing point of water depends on pressure and on dissolved impurities. Even with the fixed points specified, two thermometers calibrated to the same nominal points produced different readings between them.

Competing scales (1700s–1800s). Fahrenheit, Celsius, Réaumur, Rankine, Newton's own scale, Romer's scale — at least a dozen serious thermometric scales operated simultaneously through the 18th and 19th centuries. Each was internally coherent. None were mutually convertible without careful comparison work.

This is constitutional immaturity at the expression level rather than the substance level. The fixed points were broadly agreed upon (water freezes, water boils); the architecture of the scales differed. Through the MCI lens, this is V7-without-a-compact: multiple competing constitutional orders for the same physical quantity, each legitimate within its community, none compelled to coordinate with the others.

The unsolved problem. Underneath all of this was a constitutional defect no one could fix: thermometers calibrated to the same fixed points but using different working substances (mercury, alcohol, air, glass-stem dilation) did not agree at intermediate points. A mercury thermometer reading 50°C and an alcohol thermometer reading 50°C, both calibrated to the same 0°C and 100°C, did not register the same temperature. The substance was lying. Different fluids have different thermal expansion behaviours, and "50°C" was a property of the fluid, not of the thing being measured.

This is the deepest pre-constitutional state in the history of metrology. Length had local feet. Weight had local pounds. But length and weight had one quantity being measured against many standards. Temperature had the thermometer's working substance contaminating the measurement itself. The framework could not preserve diversity because the diversity it produced was incoherent.

IV. The First Real Constitutional Encounter: Carnot, Kelvin, and the Absolute Scale (1824 – 1848)

The constitutional breakthrough came not from better thermometry but from theoretical physics. Sadi Carnot's 1824 work on heat engines established that there is a maximum efficiency for any engine operating between two temperatures, and this efficiency depends only on the temperatures themselves, not on the working substance.

This was a T·1 condition for the entire pre-existing framework of temperature. The Carnot relation could not be expressed within mercury-degrees or alcohol-degrees because those scales were substance-dependent. The physics required a temperature scale that was itself substance-independent — a scale that depended only on thermodynamics, not on what was in the bulb.

William Thomson (Lord Kelvin), in 1848, derived such a scale. The absolute thermodynamic temperature was defined directly from the Carnot relation. A temperature T is whatever number makes Carnot's law hold for engines operating between T and some reference. Crucially, this scale identified an absolute zero — a temperature below which the Carnot relation makes engines impossible. About −273.15°C on the Celsius scale.

In MCI terms, this is the move from substance-dependent measurement to substrate-independent measurement. The Kelvin scale was the first temperature scale that did not depend on what was in the thermometer. It depended only on the structure of thermodynamics. Through the framework's lens, this is a V5-shaped move at the level of a physical quantity: temperature was no longer what a thermometer reports; it became a property of the world that any sufficiently constitutional thermometer can be calibrated against.

This is also the cleanest example in the history of measurement of a constitutional encounter triggered by theory rather than by data. Length was redefined when measurement precision exceeded artefact stability. Mass was redefined when measurement precision exceeded artefact drift. Temperature was redefined when theoretical physics revealed that the existing constitutional category was inadequate to the structure of thermodynamics itself.

V. The V7 Compact: The International Temperature Scale (1927 – present)

The Kelvin scale solved the substance problem in principle. In practice, realising it required deciding how to measure absolute temperature in real laboratories. The International Temperature Scale (ITS), first established in 1927 and most recently revised in 1990 (ITS-90), is the compact that does this.

ITS-90 is constitutionally striking. It specifies fourteen defining fixed points — the triple point of hydrogen, the freezing point of zinc, the freezing point of silver, and so on — each anchored to a reproducible physical phenomenon. It specifies which thermometer types to use in which temperature ranges. It is a compact about how to realise the Kelvin scale in practice.

Through the MCI lens, ITS-90 is V7 constitutional governance at its most demanding. It is not just a unit definition — it is a coordinated protocol across dozens of national metrology institutes for how to actually measure something whose theoretical definition is clear but whose practical realisation is hard. It is closer to the BIPM's CIPM Mutual Recognition Arrangement than to a simple unit definition. The compact governs not the standard but the procedure, because temperature is the one base quantity whose realisation procedure is itself constitutionally non-trivial.

For nearly a century, the Kelvin had a second-tier definition: from 1954 to 2019, the kelvin was defined as 1/273.16 of the thermodynamic temperature of the triple point of water — the unique pressure-temperature combination at which solid, liquid, and gaseous water coexist in equilibrium. This was an artefact-like definition (it depended on a specific physical configuration of a specific substance) but more constitutionally mature than the IPK, because the triple point is realisable anywhere with sufficient skill.

VI. The V9 Move: The 2019 SI Redefinition

On 20 May 2019, the same day mass was lifted off Le Grand K, the kelvin was lifted off the triple point of water. The kelvin is now derived from a fixed value of the Boltzmann constant: k = 1.380649 × 10⁻²³ joules per kelvin, exactly. Temperature is whatever value makes this relationship hold given the defined values of the second, the metre, the kilogram, and the elementary charge.

This is the V9 inward face — Constitutional Ground — at the level of temperature. The kelvin no longer depends on a particular substance (water) in a particular configuration (triple point). It depends on Boltzmann's constant, which is a property of how energy distributes itself across microscopic degrees of freedom in any thermodynamic system. The constitutional substrate of temperature is now the statistical mechanics of the universe itself.

It is also V9's outward face. Any laboratory anywhere — on Earth, on Mars, in interstellar space — with sufficient skill in primary thermometry (acoustic gas thermometry, Johnson noise thermometry, refractive-index gas thermometry) can realise the kelvin independently and arrive at the same value. Future civilisations recovering only the SI documentation could rebuild the kelvin without ever consulting BIPM, without using water, without any reference to Earth-specific conditions.

VII. What the Arc Shows

Five observations the framework makes visible that a standard history of thermometry does not:

Temperature is the only base quantity whose pre-constitutional state was identity-level rather than measurement-level. Length, weight, and speed all had problems with how to measure them. Temperature had a problem with what it was. The 200-year debate over caloric versus kinetic theory was a debate over the substance whose constitution the framework was trying to govern. You cannot apply the five virtues to a thing whose ontology is unsettled. This is V4 thinking at civilisational scale: goal formation cannot proceed if you do not know what kind of thing you are forming goals about.

Substance contamination is unique to temperature in the history of metrology. Length and weight standards drifted; their artefacts were vulnerable. But the substance of the standard did not contaminate the reading. A platinum metre bar at 20°C reads the same length whether you ask in French or in English. A mercury thermometer at 50°C reads differently from an alcohol thermometer at 50°C, even when both are honestly calibrated. This was the framework's longest-running Diversity Preservation failure: the diversity produced by the working substances was incoherent. The Kelvin scale resolved this by making temperature substance-independent — a move with no analogue in length or mass history.

The constitutional encounter that resolved temperature came from theory, not from measurement precision. This is rare. Most constitutional encounters in metrology arrive because measurement gets too precise for the existing standard (the Michelson–Morley result for the metre, the IPK drift for the kilogram). The Carnot encounter for temperature arrived because thermodynamics revealed that the existing constitutional category was internally incoherent. You could measure temperature as precisely as you liked with a mercury thermometer; the Carnot relation would still not work in mercury-degrees. The framework had to be revised from above by physics, not from below by precision.

The 2019 kelvin redefinition is the cleanest example in the four histories of V9 ground being reached only when the physics arrived. Length needed c to be measured to high precision before it could be fixed. Mass needed Planck's constant. Temperature needed Boltzmann's constant — which is in some ways the most theoretically demanding of the three. k connects macroscopic temperature to microscopic energy per degree of freedom. To fix it required acoustic gas thermometry and Johnson noise thermometry to mature to extraordinary precision. The kelvin's V9 move was the constitutional move that required the most physics to make possible. It is, in this sense, the deepest of the SI redefinitions — temperature is now grounded in the same statistical structure that makes thermodynamics work at all.

Temperature reveals a V9 property the other histories obscured. The 2019 redefinitions of the metre, kilogram, and kelvin all anchored to fixed constants of nature. But temperature is the only one of the four whose constant (k) is not itself a quantity of immediate physical experience. c is the speed of light — a thing. h (Planck's constant) is the quantum of action — close to a thing. k is a conversion ratio between thermodynamic temperature and energy per degree of freedom. It does not name a substance or a speed or an amount; it names a structural relationship. Temperature's V9 ground is the most abstract of the four — and yet it is the one that most fully achieves what V9 promises: the constitutional substrate has been moved to a relationship the universe maintains, not to any object the universe contains.

The history of temperature, through MCI, is the story of a quantity that began as sensation, became thermoscope-direction, became substance-dependent reading, became substance-independent Kelvin scale, became triple-point-anchored procedure, and finally — on the 144th anniversary of the Metre Convention — became a derived consequence of how energy distributes itself across the universe's microscopic states.

It is the slowest constitutional maturation of the four. It required not only the institutional architecture (V7 compact, ITS protocols), not only the theoretical physics (Carnot, Kelvin, statistical mechanics), but also the experimental physics (acoustic gas thermometry to nine significant figures) to reach the point at which the substrate could be moved. Temperature was the last quantity in everyday life to achieve constitutional ground because it was the quantity whose constitutional ground was furthest from immediate experience.

That is what V9 looks like when it has to reach all the way down to statistical mechanics to find its ground. Not the most familiar maturation. The most thorough.