Analytic geometry
What becomes possible once we assign numbers to space?
What changes under a change of coordinates while the geometry itself stays the same?
Rigid geometry has given us the right notion of motion. Distances, angles, and shapes are preserved by isometries, and the symmetries of a space organize those motions into a group.
That picture is structurally clean, but it is still hard to compute with.
Take a line in the plane. Synthetically, we can describe it as the unique straight path through two points, or as the set of points satisfying a geometric relation. That description is meaningful, but it does not yet give us an easy way to calculate intersections, compare slopes, or solve for unknown points.
Now choose axes.
Once the plane has a coordinate system, each point can be recorded by a pair of numbers. A line can be expressed by an equation. A circle can be expressed by an equation. A transformation can be expressed by formulas acting on coordinates.
Analytic geometry begins at exactly that translation.
The geometric space is still the object of study. Coordinates are a chosen description of that space. They turn geometric relations into algebraic ones, so that shapes can be studied by equations and transformations can be studied by formulas.
Geometric space
│ choose coordinates
▼
R^n with metric
│
▼
Equations representing shapes
The space comes first. Coordinates give it a numerical presentation. Equations then describe figures inside that presentation.
This is the same representational discipline we already met in linear algebra. A basis made vectors and linear maps computable without becoming the vector space or the map itself. Here a coordinate system makes geometric points and figures computable without becoming the geometry itself.
A point in the plane can be represented as a pair of numbers because it receives that pair after a coordinate system has been chosen.
That distinction matters because the same point can have different coordinate descriptions. Move the origin, rotate the axes, or rescale the coordinate grid, and the numbers assigned to a point may change. The point has not moved merely because its description changed.
The same is true for equations.
The unit circle centered at the origin may be written as
x^2 + y^2 = 1
in one coordinate system. Shift the origin, and the equation changes. Rotate axes, and the formula may take another form. The geometric circle remains the same figure in the same space while its algebraic description changes with the coordinate frame.
This is the central discipline of analytic geometry: coordinates make geometry calculable, but coordinate expressions are representations.
Once that discipline is in place, the gain is enormous. Lines can be compared by their equations. Intersections become solutions of simultaneous equations. Circles, parabolas, ellipses, and more general curves become sets of points satisfying algebraic conditions. Transformations become coordinate formulas, and rigid transformations can be tested by checking whether distance is preserved numerically.
For example, the distance between two coordinate points p = (x1, y1) and q = (x2, y2) in the Euclidean plane can be calculated as
d(p, q) = sqrt((x2 - x1)^2 + (y2 - y1)^2)
That formula expresses the Euclidean metric in the chosen coordinates. The metric is geometric structure; the formula is its coordinate record.
This lets rigid geometry become computational.
If a transformation sends coordinates (x, y) to new coordinates (x', y'), we can ask whether the distance formula gives the same value before and after the transformation. When it does for every pair of points, the transformation represents an isometry. A geometric condition has become an algebraic test.
The same method works for figures. To understand where a line meets a circle, solve their equations together. To describe a rotated or translated figure, transform the coordinates and rewrite the equation. To compare geometric descriptions, track how equations change when the coordinate system changes.
So analytic geometry adds a numerical language for the spatial content already present.
That is why the change-of-coordinates question is unavoidable. If coordinates are choices, then geometry must be what survives valid changes of description.
Under a change of coordinates, the numbers assigned to points can change. Equations can change. Formulas for transformations can change. But incidence, distance, angle, shape, and symmetry are the geometric facts that must be expressible in any coordinate system.
This separation is the structural core. At the top level, we have spaces and figures with geometric structure. At the representational level, we have coordinate tuples, equations, and formulas. Coordinate changes move us between descriptions of the same underlying geometry.
space / figure
│
├── coordinates A ── equation A
│
└── coordinates B ── equation B
same geometry, different descriptions
The two equations may look different, but they describe the same object when they are related by the coordinate change. Equality of formulas is therefore too strict a notion for the geometric level. The right comparison asks whether the formulas represent the same figure after translating between coordinate systems.
This is where analytic geometry becomes more than a computational trick. It teaches a recurring mathematical lesson: once a structure can be represented in many ways, invariance under change of representation becomes part of the subject.
Rigid geometry tells us what should be preserved by motion. Analytic geometry adds a second kind of discipline: what should be preserved when the description changes.
Those are not the same operation. A rigid motion moves figures within space while preserving geometric structure. A change of coordinates changes how the same space is described. In practice they can be written by similar formulas, which is exactly why the distinction has to stay clear.
One formula may describe an active motion of points. The same algebraic pattern may also describe a passive change of coordinates. Geometry stays coherent by tracking which level is being discussed.
The benefit is that calculation and structure now cooperate. Coordinates let us compute, while invariance tells us which computations express facts about the geometry itself.
The structural move can now be stated plainly. The objects are geometric spaces and figures equipped with distance and angle, while the representational layer consists of coordinate systems, coordinate tuples, equations, and formulas. What composes at the representational level are coordinate transformations and algebraic substitutions, and these compose because changes of description can be performed successively. The invariants are the geometric facts that survive those changes of coordinates: incidence, distance, angle, shape, and symmetry. The defining relation is representation: equations and coordinate formulas describe geometric objects relative to a chosen coordinate system. Equality of coordinate expressions belongs to a fixed representation; geometric sameness asks whether the represented object is the same after translating between coordinate systems. Continuous change, derivatives, and local approximation are still absent.
Analytic geometry therefore completes the rigid geometric stage. Space has structure, transformations preserve it, symmetries organize it, and coordinates make it computable.
But geometry is still mostly static.
What happens when the figures themselves begin to vary continuously?
And how can change be studied when a curved process is visible only through its local linear behavior?