Two-dimensional motion and projectiles

In brief

Fire a bullet horizontally from a gun at the exact instant a second bullet is simply dropped from the same height, and, ignoring air resistance, the two bullets hit the ground at the same moment, even though one has travelled far downrange and the other has fallen straight down. This is startling the first time you see it, since the fired bullet clearly moves much faster and farther than the dropped one, yet gravity does not care about that sideways speed at all. The reason is that a projectile's motion is not one complicated curved thing, it is two separate, simple things happening at once: a horizontal motion that never changes because nothing pushes on it sideways, and a vertical motion that falls exactly the way anything falls, regardless of how fast it is also moving sideways. This entry builds the curved path of a thrown or launched object entirely from that insight, treating the two directions as independent partners that never influence each other, sharing nothing but the clock that ticks for both.

The full treatment

First look: the dropped bullet and the fired bullet

Take the two-bullet scenario seriously as a testable claim rather than a trick. The dropped bullet's motion is pure vertical fall: it starts at rest vertically and speeds up downward at the constant rate any falling object does near the earth's surface. The fired bullet, at the instant it leaves the barrel, also starts with zero vertical velocity, because the gun points horizontally and gives it no vertical push; its only initial motion is horizontal. From that instant onward, gravity pulls straight down on both bullets identically, regardless of how fast either is moving sideways, so both vertical motions are identical, and both bullets reach the ground at the same instant. The horizontal motion of the fired bullet, meanwhile, has nothing pushing or pulling on it once it leaves the barrel, so it proceeds at whatever constant sideways speed it started with, entirely unaffected by the fact that the bullet is also falling.

Building the idea: two motions, one clock, no interaction

Generalize the two-bullet case into a rule: any projectile's motion splits into a horizontal component and a vertical component, and the two are independent, meaning what happens in one direction has no effect on the other. The horizontal component, with nothing acting on it sideways once launched, is motion at constant velocity, the simplest case. The vertical component, with a constant downward acceleration from gravity and nothing else, is exactly the uniformly accelerated motion already developed for a falling object. The one shared feature linking the two is time: both are functions of the same clock, so at any instant after launch, the object's true position is found by looking up where each component has gotten to at that time, then combining the two as components of a single position vector.

The formal model: building the trajectory from two ordinary motions

Let the object launch with velocity components, horizontal vx and vertical vy, at time zero, from position x0, y0. The horizontal component obeys constant-velocity motion: x equals x0 plus vx multiplied by t, with vx unchanged for the whole flight. The vertical component obeys uniformly accelerated motion under a constant downward acceleration g, roughly 9.8 meters per second squared: y equals y0 plus vy times t minus one half g times t squared, and vertical velocity changes according to vy at time t equals vy at launch minus g times t. Neither equation contains the other's variables, the algebraic expression of independence. The curved path traced out, plotting y against x instead of against t, works out to be a parabola precisely because x grows linearly in t while y grows with a term proportional to t squared, and eliminating t between the two equations produces the standard parabolic shape. Maximum height comes from the vertical equation alone, setting vertical velocity to zero, and total flight time and horizontal range follow by feeding that time, or twice it for equal launch and landing height, into the horizontal equation.

What stays independent, and where the model needs care

The deepest fact here is that launch speed in one direction never changes the object's motion in the other: a ball thrown gently sideways and one thrown hard sideways from the same height, with no vertical push either way, still hit the ground at the same time, differing only in how far downrange they land, exactly the two-bullet result generalized. This is easy to test directly, the appropriate way to confirm independence rather than accept it on faith. The model idealizes two things worth naming: a constant gravitational acceleration, an excellent approximation near the earth's surface over modest heights, and no air resistance, a much rougher approximation that a real experiment, timing a thrown or launched object, will expose the moment the object is light or fast enough for drag to matter.

Lineage

Galileo's Two New Sciences, in its Fourth Day, gives the founding argument directly: he treats the motion of a projectile as compounded from a uniform horizontal motion and a naturally accelerated vertical motion, argues the two can be considered separately, and derives that the resulting path is a parabola, confirmed with experiments rolling balls off inclined tables onto the floor. This broke with older accounts that treated a cannonball's flight as one unified, largely mysterious curve. Newton's Principia generalized the result within the broader laws of motion, treating any projectile's path as the combination of whatever motion it already had with a continuous deflection caused by an impressed force, gravity being the specific case, a framing general enough to extend later to orbital motion, where the "projectile" never comes back down because the earth curves away beneath it as fast as it falls.

The strongest case for it

The independence principle is what makes projectile motion calculable rather than merely observable. Splitting a curved, seemingly unified path into two ordinary one-dimensional motions, one already understood as constant velocity, one already understood as uniform acceleration, means no new physics is needed to handle two dimensions, only the discipline of tracking two components against a shared clock. This is exactly the technique that scales up to artillery ranging, sports science analyzing a thrown ball's trajectory, and the early phase of rocket flight, and it is confirmed the same way Galileo confirmed it, by launching real objects at known speeds and angles and checking that the predicted landing point and flight time match what is measured.

The strongest case against it

The two idealizations named above are real limits, not footnotes. Air resistance, ignored throughout the model, depends on speed, shape, and density in ways that noticeably shorten the range and steepen the fall of light or fast objects, a badminton shuttlecock or a fired artillery shell both depart from the simple parabola substantially, even though a thrown shot put over a short distance matches it closely. Treating gravitational acceleration as constant also fails once vertical distances become large enough that gravity noticeably weakens with height, a regime this entry does not reach. A common misconception is believing the horizontal and vertical motions must somehow "know about" each other to produce a smooth curved path; they do not interact at all, and the smoothness of the parabola is entirely a byproduct of both components being simple, continuous functions of the same time variable.

Where it stands now

The decomposition of curved motion into independent horizontal and vertical components has stood since Galileo's original derivation and is not revised by later physics; it remains an excellent approximation for any projectile whose flight is short enough that air resistance and the weakening of gravity with height can be neglected, which covers the overwhelming majority of thrown, kicked, and launched objects. Where the approximation breaks, long-range artillery, spacecraft, high-speed sport projectiles, the same component-based thinking still applies, only with air resistance or a varying gravitational pull added back into each component's own equation rather than independence itself being discarded.

Test yourself

A ball is launched from ground level at 20 meters per second at an angle of 30 degrees above the horizontal. Split the initial velocity into its horizontal and vertical components, then predict the time the ball spends in the air, the maximum height it reaches, and the horizontal range where it lands, using only the independent horizontal and vertical equations developed above and taking g as 9.8 meters per second squared. Then design a simple real experiment, using an object you could actually throw or roll, that would let you test whether your predicted range is correct, and state one specific reason your measured range might come out shorter than your calculation and which idealization in the model that reason violates.