Force as interaction

In brief

Push a stalled car and it starts to move; stop pushing and, on a level road, it drifts to a halt from road friction and air resistance, not because the push "ran out." Something happened between your hands and the car, and something different happens between the car's tires and the road. Neither of these is a quantity the car owns. A force is not stuff stored inside an object, waiting to be spent, it is what happens between two objects while they interact. This matters because the everyday habit of saying "the ball has force" or "the engine has power to push" quietly smuggles in the wrong picture, one object holding a private supply. The right picture has no single-object forces at all: every force names an interaction, and an interaction always has two participants.

The full treatment

First look: what changed, and between what

Stand a book on a table. Nothing appears to be happening. Yet the table is pushing up on the book and the book is pushing down on the table, right now, with no motion at all. You can feel the same thing by pressing your two hands together: each hand pushes the other, and you cannot press on anything without something pressing back. This is the first clue that a force is not "possessed," it is exchanged, or more precisely, it is a description of what one thing is doing to another at a given instant. Whenever you are tempted to say "X has a force," stop and ask "what is X interacting with, and what is X doing to it?" If you cannot name a second object, you have not yet found a force, you have found a word standing in for one.

Building the idea: force needs two names

Consider the alternative, more common intuition: that force is a kind of fuel a moving object carries, and motion continues only while the fuel lasts. This was Aristotle's picture and it seems to fit thrown objects, which do slow down. But it fails a simple test: it cannot explain why a puck sliding on smooth ice barely slows at all, while the same puck sliding on carpet stops almost immediately. If motion needed internal fuel, the puck's own supply should run out at the same rate regardless of surface. It does not. What differs is the interaction between puck and surface, namely friction, and friction is plainly a relationship between two objects, not a property of the puck alone. Once you see this, the whole picture reorganizes: every force in a diagram must have an identifiable source object and an identifiable target object. "Gravity pulls the ball down" is shorthand for "the Earth pulls the ball, and the ball pulls the Earth back." "The rope holds the box" is shorthand for "the rope pulls the box, and the box pulls the rope."

The formal model: force as a description of an interaction

Formally, a force is a directed, measurable push or pull that one object exerts on another, arising from some physical interaction (contact, gravity, electric attraction, and so on). Write a force as F, with a magnitude (how strong) and a direction (which way), acting from object A on object B. The convention "force of A on B" is not decoration, it is the content: drop either name and you no longer have a force, you have an unattached number. A free-body diagram makes this discipline explicit: pick one object, draw only the forces that other objects exert on it (never the forces it exerts back on them), and label the source of each arrow. If you cannot label a source, erase the arrow. This single habit, insisting that every force arrow answers "from what?", catches the majority of beginner errors in mechanics, such as drawing a mysterious forward "force of motion" on a coasting object that is, in fact, interacting with nothing that pushes it forward at all.

Mechanism: interaction, not possession

Why does this reframing matter mechanically and not just linguistically? Because it fixes what counts as a complete explanation. A correct account of any motion must be a list of interactions: this object touches that one (contact force, e.g. normal force or friction), this object is near a mass (gravitational interaction), this object is near a charge (electric interaction), and so on, each with its source named. Nothing is left as an unexplained residue belonging to the object itself. This is also why forces come in the pairs described elsewhere as action and reaction: since a force is defined as an interaction between two bodies, describing "A's force on B" already implies there is a corresponding "B's force on A" from the very same interaction, not a separate coincidence layered on top.

Lineage

The everyday intuition that motion needs a continuously supplied cause goes back at least to Aristotle's physics, where a projectile's "impetus" was imagined as an internal quantity that depletes. Medieval impetus theorists (notably Jean Buridan in the fourteenth century) refined this picture but kept force as something carried. The decisive break came with Galileo's studies of inclined planes and rolling bodies, which suggested that unforced motion persists rather than decays, and with Newton's Principia Mathematica in 1687, which formalized force as the interaction that changes motion, and crucially, insisted (in the third law) that such interactions are always mutual between two bodies. This is the ancestor of the modern free-body diagram and the "force of A on B" notation used in every physics course since.

The strongest case for it

Treating force as strictly relational, always between two named objects, has enormous diagnostic power. It turns "why does this move the way it does" into a checklist: list every object touching or otherwise interacting with the object of interest, name the force each one exerts, and you have the complete cause. This is exactly the method used to design bridges (identify every load path), analyze collisions (identify every contact interaction), and predict orbits (identify every gravitating body). The approach generalizes cleanly to fields you cannot see or touch, such as gravity and electromagnetism, precisely because it never depended on contact in the first place, only on two identifiable participants and a measurable effect. Its reach, from a shove on a cart to the pull between the Earth and Moon, is why this is the starting definition adopted by every serious treatment of mechanics rather than a simplification for beginners.

The strongest case against it

The framing idealizes away some real complications. First, in continuous media, such as a fluid pushing on a submerged object from every direction, "naming the two objects" still works in principle but becomes unwieldy; physicists switch to describing a continuous field of stress rather than tracking every fluid parcel as a separate interacting body. Second, at the boundary of the theory, general relativity reinterprets gravity not as a force between two objects at all but as the curvature of spacetime, meaning that "force as interaction between two bodies" is a Newtonian approximation, extremely accurate at ordinary speeds and masses, that stops being the deepest description in extreme regimes such as near a black hole. Third, a persistent misconception is to treat "reaction forces" as somehow weaker or delayed, when in fact both members of an interaction pair exist at the same instant with equal magnitude; another is to draw a force "for" motion itself (a phantom forward force on a coasting object) rather than restricting arrows to forces actually being exerted by some other identifiable object right now.

Where it stands now

The relational definition of force is settled physics, foundational to classical mechanics and unchallenged at the level this entry addresses. It is also the entry point to more advanced pictures, quantum field theory recasts fundamental interactions as the exchange of particles between two systems, which is a deeper but structurally similar story: still two participants, still a describable exchange. Nothing in later physics removes the requirement that a force be traceable to an interaction; later theories only refine what "interaction" ultimately consists of.

Test yourself

A person stands still on a stationary rowboat and then steps toward the dock, and the boat drifts backward away from the dock even though nothing outside the boat-person system touched it. Draw a complete interaction map: list every pair of objects involved, name the force each exerts on the other, and use it to explain why the boat moves. Then apply the same method to a rocket in deep space, far from any planet, accelerating forward by ejecting exhaust gas backward, and identify the two interacting bodies whose mutual push explains the rocket's acceleration despite there being nothing outside the rocket to push against.