How Science Works | Physics — Matter, Energy, Motion, Fields and the Search for General Laws

1. The Scientific Job of Physics

The central job of physics is to build and test models of matter, energy, space, time and interaction. A useful physical model tells us what quantities matter, how they relate, what should happen when conditions change, and where the model is expected to stop working.

A model therefore does more than fit yesterday’s data. It must expose itself to possible failure. If the model says a pendulum’s period depends on its length in a particular way, a new length gives a new test. If a theory predicts a spectral line, a measured spectrum becomes evidence. Prediction is powerful because it binds explanation to consequences that can be checked.

Physics is especially interested in relationships that remain stable across many cases. That is why conservation laws, symmetries, scaling relations and field equations are so important. They turn local observations into reusable structure.

2. A Systems Lens: Object, State, Occurrence and Relationship

One way to keep physical reasoning clean is to separate four jobs. First identify the object or system: a ball, gas, circuit, beam of light, atom or galaxy. Then define its state: position, velocity, temperature, charge distribution, pressure, quantum state or another measurable description. Next identify the occurrence: collision, acceleration, heating, emission, decay, phase transition. Finally identify the relationship: the rule connecting variables before, during or after the change.

Many weak explanations collapse these jobs into one sentence. “The object moved because of energy” is incomplete because it does not specify the interaction, the system boundary, the energy transfer pathway or the state change. Physics improves when nouns, states, events and relations are kept distinct long enough to show what actually changed.

3. Measurement Comes Before the Equation

Every physical quantity must connect to an operational way of being measured or inferred. Length can be compared with a calibrated scale. Time can be counted by a clock process. Voltage can be measured as electric potential difference. Temperature is inferred through calibrated thermometric properties. Some quantities, such as the mass of a distant star or the energy of a particle, are not measured directly in a simple sense; they are inferred through models and instrument signals.

This matters because a number without a measurement story can become decorative mathematics. Good physics asks: What instrument produced this value? What was calibrated? What resolution was available? What assumptions convert the raw signal into the reported quantity? What sources of uncertainty remain?

4. Units Are Part of the Meaning

Units are not labels pasted onto numbers after calculation. They constrain what a physical statement can mean. A speed in metres per second, an acceleration in metres per second squared, an energy in joules and a power in watts belong to different dimensions. Dimensional analysis can expose an impossible equation before any experiment is performed.

Suppose someone proposes that distance travelled is proportional to acceleration multiplied by time. The dimensions would be length per time squared multiplied by time, leaving length per time, not length. The expression cannot represent distance as written. Physics gains reliability because algebra is forced to respect physical dimensions.

5. Laws, Models and Theories Are Different Jobs

A law usually states a regular relationship: for example, an idealised proportionality or conservation principle. A model is a deliberately simplified representation used to reason about a system. A theory is a broader explanatory framework capable of integrating many observations, models and laws.

These are not ranks in which a theory eventually “graduates” into a law. Newtonian mechanics remains extraordinarily useful even though relativity provides a deeper account in regimes where Newtonian approximations fail. The question is not whether an older model becomes false everywhere, but where its error becomes significant for the task at hand.

6. Idealisation Is a Feature, Not a Lie

Physics often begins by deliberately removing detail: point masses, frictionless surfaces, massless strings, ideal gases, perfectly rigid bodies, monochromatic waves. These objects do not exist exactly in nature. Their value lies in exposing a mechanism or relationship without every real-world complication arriving at once.

The discipline lies in returning the omitted effects when they matter. A frictionless model may explain the basic exchange between kinetic and potential energy. Real bearings, air resistance and deformation can then be added as correction terms. A good idealisation has an explicit return path to reality.

7. Motion: Describing Change Before Explaining It

Kinematics describes motion using position, displacement, velocity and acceleration without yet deciding what causes the motion. This separation is powerful. A motion graph can be analysed before the force model is known. Velocity is the rate at which position changes; acceleration is the rate at which velocity changes. An object moving rapidly can have zero acceleration, while an object moving slowly can have large acceleration.

This is an example of scientific role discipline: description first, mechanism second. Mixing the two too early encourages circular explanation.

8. Forces: Interactions, Not Supplies of Motion

A force represents an interaction capable of changing momentum. Newton’s laws are powerful because they connect a system’s motion to the net interaction acting on it. The first law clarifies inertial motion; the second relates net force to the rate of change of momentum; the third expresses paired interactions between bodies.

A common misconception is that continuous motion requires a continuous forward force. In an inertial frame, an object with no net force continues at constant velocity. What often makes everyday experience look different is friction and drag. Once those interactions are identified, the apparent contradiction disappears.

9. Momentum: Tracking Motion Through Interaction

Momentum combines mass and velocity into a quantity whose change is tied to impulse. For a suitably isolated system, total momentum is conserved. This is why collision analysis can remain tractable even when the detailed contact forces are complex. The conservation law survives the messy event and lets us connect before and after states.

The important phrase is “suitably isolated.” Conservation statements depend on system boundaries. If external forces produce significant impulse, the momentum of the selected subsystem need not remain constant. Physics therefore requires the analyst to declare what is inside the system and what crosses its boundary.

10. Energy: A Bookkeeping Architecture for Change

Energy is one of physics’ most transferable abstractions. It allows motion, position, heat, radiation, chemical processes and electrical processes to be connected through a common conservation framework. Energy is not a visible substance. It is a state-dependent accounting quantity whose conservation places strong constraints on possible change.

When energy appears to be “lost,” the better question is where it went and whether the system boundary included the receiving store. Mechanical energy may decrease while internal energy rises through friction. Electrical energy delivered to a resistor becomes internal energy and is transferred to the surroundings by heating and radiation. Conservation does not guarantee usefulness: energy can remain conserved while becoming less available for a desired task.

11. Fields: Making Interaction Local

A field assigns a physical quantity to positions in space and time. Gravitational, electric and magnetic fields allow us to describe how an object would interact with its surroundings without imagining action as an unexplained jump across empty distance. Modern physics pushes this idea much further: fields are not merely convenient maps but central ingredients in fundamental descriptions.

Field reasoning changes the question from “What force does A somehow send to B?” to “What field state exists here, and how does an object with the relevant property respond?” That shift is one of the great conceptual reorganisations in physics.

12. Waves: Patterns That Transport Energy and Information

Waves involve organised variation that propagates through a medium or, for electromagnetic waves, through electromagnetic fields. Frequency, wavelength, amplitude, phase and propagation speed create a language that works across sound, water waves, light and many quantum phenomena.

Superposition allows waves to add. Interference, diffraction, standing waves and resonance emerge from that rule. Once again, a small number of relationships generates many visible phenomena. Physics becomes powerful when one mechanism explains what previously looked like separate facts.

13. Thermodynamics: What Large Numbers of Microscopic Degrees of Freedom Do

Thermodynamics studies energy transfer, temperature, work, heat and constraints on macroscopic change. Its laws are unusually general. The first law is an energy accounting statement. The second law introduces directionality through entropy and the statistical tendency of systems toward overwhelmingly more numerous macrostates.

This is why heat naturally flows from hotter to colder regions in ordinary conditions, why engines cannot convert all absorbed heat into work, and why many processes are effectively irreversible even though microscopic laws may be reversible. Statistical physics provides the bridge between microscopic possibilities and macroscopic regularity.

14. Electricity and Magnetism: One Electromagnetic Structure

Electric charge, electric fields, magnetic fields, currents and electromagnetic induction were once studied as partly separate phenomena. Maxwell’s equations unified them into one structure and predicted electromagnetic waves whose speed matched the measured speed of light. Light itself became part of electromagnetism.

This episode illustrates a recurring scientific success pattern: apparently separate domains become connected by a model that explains old observations and predicts new ones. Unification is valuable not because fewer equations are aesthetically pleasing, but because the new structure survives independent tests.

15. Relativity: When Space and Time Stop Being Separate Backgrounds

Special relativity reorganises space and time around the invariance of the speed of light and the equivalence of inertial frames. Measurements of time interval, length and simultaneity depend on relative motion, while spacetime intervals provide invariant structure. Energy and momentum are likewise unified.

General relativity goes further by describing gravitation through spacetime geometry. Matter and energy influence geometry, and geometry influences motion. Newtonian gravity remains an excellent approximation for many everyday and engineering situations, but precision timing, strong gravity and cosmology require the relativistic framework.

16. Quantum Physics: Probability Is Built Into the Predictive Structure

Quantum theory describes matter and radiation at scales where classical intuitions fail. Quantum states evolve according to precise mathematical rules, yet measurement outcomes are generally predicted probabilistically. Superposition, quantisation, interference, entanglement and uncertainty are not decorative oddities; they are repeatedly tested features of the framework.

Quantum mechanics does not mean “anything can happen.” Probabilities are tightly constrained. The theory can make extraordinarily accurate statistical predictions, and modern electronics, lasers, atomic clocks and much of materials science depend on those predictions being reliable.

17. Scale Decides Which Description Is Useful

No single level of description is always best. A fluid can be treated as a continuous medium even though it is made of molecules. A bridge can be analysed using continuum mechanics without tracking atomic wavefunctions. A gas law can predict pressure while ignoring the trajectory of every molecule.

This is not intellectual laziness. It is controlled coarse-graining. Physics asks which variables retain the information needed for the present question. Good modelling discards detail only after deciding that the discarded detail will not materially change the conclusion.

18. Experiments Are Designed Arguments

A physics experiment is not merely “doing something and seeing what happens.” It is an argument built from apparatus, calibration, controls, measurements, assumptions and analysis. The design tries to make competing explanations produce distinguishable outcomes.

Strong experiments manage nuisance variables, quantify uncertainty, distinguish systematic from random effects, and preserve enough methodological detail for independent checking. In some branches of physics, controlled experiments are possible. In astrophysics and cosmology, researchers often rely on observational tests because the objects cannot be manipulated. The scientific logic remains: models meet evidence under constraints.

19. Uncertainty Is Information

Every measurement has limits. Instrument resolution, calibration uncertainty, environmental variation, sampling, model assumptions and statistical noise all affect what can be claimed. Reporting uncertainty does not weaken science; it specifies the width of the evidence.

A result of 10.0 metres is not automatically more informative than 10 metres. The number of displayed digits must be justified by measurement capability. Precision and accuracy are different: repeated measurements can be tightly clustered yet systematically wrong. A laboratory culture that records uncertainty is harder to fool with attractive numbers.

20. Correlation, Mechanism and Causality

Physics often benefits from systems simple enough that causal mechanisms can be isolated with unusual clarity. Yet the same caution applies as elsewhere in science. Two quantities varying together does not by itself establish that one causes the other. A model must specify an interaction or dependency, and tests must discriminate that model from alternatives.

Mechanism matters because it gives the relationship structure. If increasing current raises heating in a resistor, the explanation connects electrical work, resistance, power and energy transfer. The claim becomes more than a graph: it becomes a testable account of how the state changes.

21. Mathematics Is the Constraint Language

Mathematics gives physics a precise way to express relationships, transformations, symmetries and limits. Equations are valuable because they force multiple verbal claims to coexist consistently. Calculus expresses rates of change and accumulation. Linear algebra organises states and transformations. Differential equations encode dynamics. Probability handles uncertain outcomes. Geometry becomes the language of spacetime.

But mathematics alone does not make a statement physical. The symbols must be tied to measurable quantities, the assumptions must fit the system, and the resulting predictions must survive evidence. Physics is not mathematics with units; it is an empirical science whose models are frequently mathematical.

22. Computation Extends What Can Be Solved

Many realistic systems cannot be solved exactly on paper. Numerical simulation lets physicists approximate fluid flows, plasma dynamics, climate components, galaxy formation, quantum systems and detector behaviour. Computation does not remove the need for scientific judgement. It adds new questions: Was the algorithm verified? Was the grid fine enough? Were parameters calibrated independently? Did numerical error masquerade as physical behaviour?

A simulation is a model executed, not reality observed. Its reliability depends on equations, boundary conditions, parameter choices and numerical implementation. The strongest work checks simulation output against analytic limits, benchmark cases and empirical data.

23. Worked Example: A Falling Ball

Consider a ball released near Earth’s surface. The first question is not “Which formula do I remember?” but “What system and accuracy do I need?” For a short fall at ordinary speed, we may model Earth’s gravitational field as uniform and neglect air resistance. The ball begins with a chosen initial velocity and experiences approximately constant downward acceleration.

Kinematics predicts position and velocity as functions of time. A force model explains the acceleration through gravitational interaction. An energy model connects the loss of gravitational potential energy with increasing kinetic energy. These are not competing explanations; they are different representations of the same process under compatible assumptions.

If the ball is light, broad or moving fast, air resistance may no longer be negligible. The model must change. The scientific strength lies not in pretending the simple model is universal, but in knowing the conditions under which it is adequate.

24. Worked Example: A Resistor in a Circuit

Suppose a resistor is connected across a source. A circuit model identifies potential difference, current and resistance. For an ohmic resistor under conditions where resistance remains approximately constant, voltage and current are proportional. Power can be expressed as the rate of electrical energy transfer.

If the resistor heats significantly, its resistance may change. The original proportionality can fail. Again, the “law” was never a magic sentence detached from conditions; it was a model of behaviour within a specified regime.

25. Worked Example: An Ideal Gas Becoming Real

The ideal gas model treats particles as point-like and neglects intermolecular forces except during elastic collisions. It predicts a compact relationship among pressure, volume, amount of substance and temperature. At moderate densities and suitable temperatures, many gases approximate this behaviour well.

At high pressures or low temperatures, particle volume and intermolecular attractions matter. More detailed equations of state are needed. This transition shows how physics builds layers: a simple model first, then controlled corrections where evidence shows the approximation losing accuracy.

26. The Most Common Physics Failure Modes

• Formula hunting: selecting an equation before defining the system and quantities.
• Unit blindness: treating incompatible dimensions as if algebra alone could make them valid.
• Boundary blindness: claiming conservation without stating what is inside the system.
• Model literalism: assuming an idealisation is reality rather than a controlled simplification.
• Graph literalism: seeing correlation or curve shape and inventing a mechanism afterward.
• Precision theatre: reporting more digits than the measurements justify.
• Scale confusion: using a model outside the regime where its assumptions hold.
• Explanation by label: saying “because gravity,” “because energy,” or “because quantum” without showing the chain of state change.

27. How to Think Like a Physicist

Start by drawing the system boundary. Name what is measured and what is inferred. Write units beside quantities. State assumptions before using equations. Use more than one representation when possible: words, diagram, graph, equation, limiting case. Ask what would change if an assumption were removed. Estimate the order of magnitude before trusting a calculator. Check whether the final result has the right units and plausible scale.

Most importantly, keep the return path from model to world visible. Physics is strongest when the chain can be traced from observation to representation to calculation to prediction and back to measurement.

28. Physics Connects Outward

Chemistry depends on quantum physics, electromagnetism and statistical mechanics. Biology depends on mechanics, transport, thermodynamics, optics and molecular interactions. Earth science depends on fluid dynamics, heat transfer, wave propagation, gravity and radiation. Engineering turns physical models into constrained designs. Medicine uses imaging, radiation, fluid mechanics, acoustics and instrumentation. Computing depends on semiconductor physics and information carried by physical states.

This is why physics is high in the first batch of the How Science Works subject map: many scientific disciplines inherit its measurement language and physical constraints, even when their central questions require additional levels of explanation.

29. What Physics Does Not Do Alone

A complete explanation of a living cell is not obtained merely by listing the motion of its atoms. A social institution is not usefully understood by reducing every participant to particles. Higher-level sciences identify stable patterns, information structures, functions, histories and organisations that are real explanatory targets at their own scale.

Physics constrains those higher levels, but constraint is not the same as complete explanation. Scientific maturity includes knowing which scale owns the question.

30. The Frontier Is Where Models Meet Their Limits

Physics continues to test the nature of dark matter and dark energy, the quantum structure of gravity, neutrino properties, the behaviour of strongly correlated materials, turbulence, high-energy plasmas, quantum information and many other frontiers. These are not signs that physics has failed. They are signs that the current map has edges.

A mature science does not hide its unknowns. It marks them, measures what can be measured around them, and designs the next discriminating observation.