[ 00 ] · The Atelier · Method · 31° 14′ N
An instrument is arithmetic poured into mass.
How Obliquity turns the geometry of an axis into a leaning monolith — the orbital math, the survey that fixes true north, the concrete that holds for centuries, and the single solstice noon when the sky confirms the sum.
[ 001 ] · Premise
Engineering at the scale of patience
Most engineering is optimisation against a deadline. We design against the precession of the equinoxes — a clock whose hand completes one full circuit of the sky in 25,772 years.
At that scale the usual questions invert. The problem is never how to build something quickly; it is how to build something that will still be saying the same true thing long after everyone who poured it has been forgotten, and after the words on this page have decayed into a language no one reads.
The work resolves into five disciplines that must agree with one another to within an arc-minute: the orbital geometry that defines an alignment; the survey that pins that geometry to the Earth and finds true north without ever trusting a compass; the concrete, massive and durable enough to hold the angle for centuries while the desert tries to take it back; the bore and the morning of first light, when the instrument is proofed against the sky itself; and the stewardship that carries the record forward. None of the five tolerates the others' error.
STELE I, on the Carrizal Playa, is recently begun. What follows is method and intention, not a ledger of completed feats. The instrument is conceived as a hundred-year build; the foundation work now underway is the first of many phases, and the day the amber beam first lands true will fall to hands we will never shake. We write this in the future tense on purpose.
[ 002 ] · ε = 23.44°
The geometry of an alignment
An alignment begins as a single question asked of a single latitude: where, in this sky, over this ground, does a chosen celestial event stand? Everything downstream — the lean of the slab, the angle of the bore, the position of the one engraved mark — is the long-form answer, expressed first in degrees and only later in concrete.
Consider the pole. At the Carrizal Playa the north celestial pole sits 31.24° above the level horizon, because the altitude of the pole equals the latitude of the observer, exactly and without exception. The polar bore is cut to that angle so its line of sight holds the pole while the Earth rotates beneath it; over one sidereal day — 23 hours, 56 minutes, 4.091 seconds — the entire field of stars wheels once around that fixed, empty point. We aim at the point, not at the star. Polaris presently lies about 0.7° off true and is itself only a temporary tenant: precession handed the role to Thuban in the deep past and will hand it to Vega around the year 13,700.
Now the solstice. The monolith's outward lean is fixed at 23.44° from vertical — a deliberate quotation of the Earth's present axial obliquity, the tilt that makes seasons exist at all and that gives this studio its name. A separate south-facing port admits the winter-solstice noon sun, which at this latitude climbs to only 35.32° before falling back toward the longest night. For a few minutes around local noon that low light rakes the length of the aperture and lays the amber beam across one engraved line cut into the inner wall. That is the instrument's single true tick of the year, and Solstice Amber is the only colour the instrument is permitted to make.
The numbers are not eternal, and the instrument is built to admit it. The obliquity itself is slowly decreasing, by roughly 0.47 arc-seconds a year; the equinox drifts at 50.29″ annually. Each instrument is calibrated to a stated epoch — STELE I to J2026.0 — and the slow divergence of sky from stone across the coming millennia is not an error to be corrected but a further reading the instrument was always meant to take.
| Quantity | Value | Note |
|---|---|---|
| Axial obliquity (ε) | 23° 26′ 07″ · 23.4353° | decreasing ≈ 0.47″ / yr |
| Site latitude (φ) | 31° 14′ 22″ N | pole altitude = φ |
| Polar-bore inclination | 31.24° above horizon | aimed at the geometric NCP |
| Monolith lean | 23.44° from vertical | a quotation of ε |
| Sidereal day | 23h 56m 04.091s | one rotation against the stars |
| Mean solar day | 24h 00m 00s | noon to noon |
| Winter-solstice noon altitude | 35.32° | the year's lowest noon sun |
| Solstice-port azimuth | 180° 00′ 00″ (true S) | the amber-beam axis |
| Precession period | 25,772 yr | one circuit of the pole |
| Precession rate | 50.29″ / yr | drift of the equinox |
[ 003 ] · 31° 14′ 22″ N
Finding true north
Magnetic north is a moving error of several degrees that wanders year to year; we never use it, on any instrument, for any purpose. True north is recovered from the sky and from the Earth's own rotation, and then driven into the ground as a permanent reference before a single form is built.
The fix proceeds in layers. A primary reference monument is anchored to bedrock below the playa's seasonal moisture line. From it we take repeated transit observations — Polaris and circumpolar stars by night, timed solar sights by day — and reduce them until the meridian is established to within ±2 arc-seconds. A gyrotheodolite gives an independent determination from the spin of the Earth alone. Static GNSS sessions then tie the whole site into the global geodetic datum, so the alignment can be recovered even if every surface marker is lost.
Then we account for the ground moving. The North American plate carries this site roughly 20 mm to the southwest each year — about two metres over the instrument's first century. The alignment to the sky is untouched by this, because the bore points at the Earth's rotation axis rather than at any coordinate on its surface; the pole does not drift with the continent. But the survey record fixes the datum epoch precisely, so the stewards of the year 2200 can read the accumulated plate motion directly off our reference monument.
[ 004 ] · 184 t
The concrete, and how it is meant to age
Concrete is the only common building material that grows stronger for a century and asks for almost nothing in return. That patience is the whole reason it is here.
The hard problem is not strength but permanence: ordinary reinforced concrete fails from the inside out, as embedded steel corrodes, swells, and spalls the cover away. So there is no steel anywhere in the body — reinforcement is basalt fibre, non-metallic and non-corroding, leaving nothing inside the slab that can rust, expand, or split it across the long horizon of the build. The binder is chosen for low hydration heat and low permeability rather than for speed, and the monolith is cast as a single continuous lift of roughly 22 hours so it cures as one body, with no cold joints for water to find.
At 184 tonnes — about 73 cubic metres — its mass is itself a structural argument: enough inertia to hold its angle against frost heave, gale, and the slow creep of the ground, and enough thermal lag that the desert's daily swing never reaches its core.
The surface is left board-formed and unsealed, and this too is deliberate. The slab will carbonate inward at roughly 18 millimetres a century — a figure built into the depth of both the protective cover and the engraved mark — and over generations it will take on the dark mineral patina, the desert varnish, the wind-cut texture of everything else that stands a long time in that place. We do not protect the instrument from age; we design it to wear its age legibly.
| Property | Specification | Rationale |
|---|---|---|
| Binder | CEM II/B-LL · 40% GGBFS · 8% silica fume | low hydration heat; low long-term permeability |
| Water / binder ratio | 0.34 | density and durability over easy workability |
| Coarse aggregate | local basalt, 20 mm graded | thermal match to the playa; native colour |
| Reinforcement | basalt-fibre bar — no steel | nothing to corrode, rust, or spall |
| Design strength | 60 MPa at 365 days | concrete that keeps gaining |
| Primary mass | 184 t (≈ 73 m³) | inertia against frost, wind, and time |
| Pour | single continuous lift, ≈ 22 h | one body, no cold joints |
| Cure | 90-day shaded fog cure | shrinkage held below cracking |
| Surface | board-formed, unsealed | left to weather and patina honestly |
| Carbonation allowance | 18 mm / century | built into the cover and the mark depth |
[ 005 ] · First light
First light: when the sky checks our arithmetic
Every figure on the page upstream of this one is a hypothesis. The sky is the only authority that can confirm it, and it does so on a single morning a year.
First light is the proof: the winter-solstice noon sun either lays the amber beam across the engraved mark, or it does not, and there is no arguing with the result. We build in the tolerance to be right. The solstice port and the polar bore are first cut deliberately undersized; across the days bracketing the solstice we observe the actual beam where it falls, measure its offset from the intended line in millimetres, and only then ream the aperture to walk the light onto the mark. The polar bore is proofed the same way after dark, sighted against the circumpolar stars until the field is seen to turn cleanly about the centre of the aperture.
From the day it reads true, the instrument has no operator and no switch. The Earth turns; once each year the low amber light returns to the line; the reading is identical for a child in 2030 and a stranger in 3030. What we hand over at first light is not a machine to be run but an appointment the sky will keep on its own, indefinitely, whether or not anyone is there to witness it.
After first light there is nothing to switch on. The sky runs the instrument, and the instrument keeps only one appointment a year.
[ 006 ] · The studio
Who builds it
Obliquity is a small standing studio of six disciplines, augmented for each commission by local surveyors, contractors, and masons who know the particular ground.
No discipline outranks another and none works alone; an alignment is only ever as true as its least careful number, and the engraved mark is only as honest as the survey beneath it. The roles below are the irreducible set — remove any one and the instrument lies.
- 01Geodesist
- Fixes true north and the datum, and models the site's tectonic drift across the build's century.
- 02Observational astronomer
- Owns the alignment geometry, the ephemeris, and the first-light proof against the real sky.
- 03Structural & materials engineer
- Designs the mass-concrete body and the century-durability mix, with no metal inside it.
- 04Board-form mason
- Builds the formwork and the surface, and cuts the single engraved mark the beam must find.
- 05Optical technician
- Bores and reams the apertures; calibrates the polar sight and the solstice beam to the millimetre.
- 06Land steward
- Keeps the site, the weathering log, and the endowment meant to outlast the studio itself.
[ 007 ] · Principles
What we hold to
These are not aspirations; they are constraints we accept before a commission is taken, and they decide more of the form than any aesthetic preference does. When two of them conflict, the one closest to deep time wins.
- 01Build for the instrument, not the view
- Legibility to the sky comes before the photograph taken from the road.
- 02One true tick
- Each instrument keeps a single, unambiguous annual event, marked exactly once.
- 03Material honesty
- Unsealed concrete left to age in the open, reporting its own time as it goes.
- 04No moving parts
- The Earth and the sky do the moving; the stone is built only to wait.
- 05Repairable by hand
- Any future repair must be possible with the ordinary tools of any century.
- 06Documented to outlive us
- Every figure archived three ways — in the stone, on paper, and in living stewardship.
[ 008 ] · Commissions
Siting a new instrument
Obliquity accepts a small number of commissions from institutions, municipalities, observatories, and private patrons prepared to build at the scale of generations. A commission is not the purchase of an object.
It is the founding of an instrument together with the endowment that will keep it reading true long after the founders are gone — closer to chartering a small observatory than to buying a sculpture. We begin every commission with a latitude study, because the geometry follows from the site and not the reverse. A workable site needs an open horizon in the direction of the chosen event, skies dark enough that the alignment is not lost to glare, ground that will hold a foundation, and — most demanding of all — secure long-term tenure: perpetual or long-lease land, paired with a maintenance trust.
The phases below are indicative; the exact spans flex with latitude, ground, and weather. STELE I is presently between Phase 02 and Phase 03 — design proofed, foundation work commencing — and is the working reference for any new commission. First light is scheduled to a solstice, never to a convenient calendar date: the sky sets that appointment, and we keep it.
| Phase | Span | Deliverable |
|---|---|---|
| 00 · Inquiry & latitude study | 2–3 mo | feasibility dossier; event and epoch chosen |
| 01 · Site survey & sky audit | 6–9 mo | true-north fix, horizon profile, darkness baseline |
| 02 · Design & alignment proof | 9–12 mo | geometry, scale model, bore calculations |
| 03 · Foundation & north monument | 4–6 mo | bedrock anchor; permanent reference markers |
| 04 · Formwork & continuous pour | one season | primary monolith cast as a single body |
| 05 · Cure, finish & bore calibration | 12 mo | aperture cut and optically proofed |
| 06 · First light & dedication | the next solstice | amber-beam verification on the mark |
| 07 · Stewardship & endowment | perpetual | maintenance trust and observation log |