What Physics Is & How We Do It

Physics is the study of matter, energy, and the forces that tie them together — and it's the most fundamental science of all. Here's the mind-blowing part: the same handful of rules that make an apple fall also steer entire galaxies. That's what physicists are hunting for — those deep, universal rules. And the one thing that separates physics from just having opinions is this: every claim has to survive measurement. If the numbers don't back you up, you're wrong. No exceptions.

Physics isn't a pile of facts to memorise — it's a method for interrogating the universe, and it runs in a loop. You observe something puzzling, ask a sharp question, and propose a hypothesis — a prediction you could actually prove wrong. Then you run an experiment, analyse the data, and conclude. If your hypothesis fails? That's not a dead end — you refine it and go round again. Follow the loop:

That word 'falsifiable' is doing a lot of work. A real hypothesis must be something a test could prove wrong. 'Heavier things fall faster' is scientific — we can test it (and it turns out to be false in a vacuum!). 'The universe is beautiful' isn't science — no measurement could ever disprove it.

Here's the trick that makes an experiment trustworthy: change only one thing at a time. The independent variable is the single quantity you deliberately change. The dependent variable is what you measure in response. And the controlled variables are everything else you deliberately keep the same. Change two things at once and you've ruined it — you'll never know which one caused the effect.

In everyday chat, 'it's just a theory' means a wild guess. In science it's the complete opposite! A law describes what happens, usually as a tidy equation — like Newton's F = ma. A theory explains why, and it's a huge, battle-tested framework — like the theory of relativity. A theory isn't a weak law; they answer totally different questions. Look at the difference:

So next time someone says 'evolution is just a theory,' you can smile — in science, becoming a theory is about the highest honour an idea can earn.

Physics is enormous, so we carve it into branches, and it helps to know the map before you start the journey. Mechanics studies motion and forces, and it is where this course begins. Thermodynamics deals with heat, temperature and the flow of energy. Electromagnetism unites electricity, magnetism and light into a single framework. Optics looks at how light bends, reflects and forms images. Waves and acoustics cover everything that oscillates, from a guitar string to an earthquake. And modern physics, meaning relativity and quantum mechanics, handles the very fast and the very small. Different chapters, but one shared toolkit of careful measurement and mathematics runs through them all.

You may have heard physics called the science that speaks maths, and there is a good reason for that reputation. A sentence like 'the ball speeds up as it falls' is true but vague; the equation v = gt says exactly how much faster, at every instant. Mathematics lets a physicist compress a whole paragraph of behaviour into one compact, testable statement, and, crucially, it lets us make numerical predictions we can check against a stopwatch or a ruler. That is the deal physics offers: give me the starting conditions and the right equation, and I will tell you what happens next, to as many decimal places as you care to measure.

Here is a secret that surprises people: physics almost never describes the real world exactly. It describes a simplified version called a model. When we analyse a falling ball we often ignore air resistance and treat the ball as a single point mass, as if all of its matter sat at one dot. Is that wrong? A little, but it is wrong in a way that barely matters and makes the maths tractable. The real skill of a physicist is choosing which details to throw away: keep the ones that shape the answer, and drop the ones that do not. A good model is like a good map, useless if it shows every pebble, and priceless precisely because it leaves most of them out.

No measurement is perfect, and the two ways it can go wrong behave very differently. A random error scatters your readings unpredictably, sometimes high and sometimes low, so repeating the measurement and averaging tends to cancel it out. A systematic error pushes every reading the same way: a ruler that starts at 2\ \text{mm} instead of zero, or a clock that runs slow, biases the entire data set in one direction, and no amount of averaging will remove it. Telling the two apart matters, because their cures are opposite. You fight random error with more trials, but you fight systematic error by hunting down and fixing the flaw in your instrument or method.

One of the most valuable habits science can give you is a refusal to leap from 'these two things move together' to 'one of them causes the other.' Ice-cream sales and drowning deaths both rise in summer, yet ice cream does not cause drownings; a hidden third factor, hot weather, drives both. This is exactly why the controlled experiment, in which you change one variable and hold the rest fixed, is so powerful: it is the cleanest way to turn a mere correlation into genuine evidence of cause. Whenever someone claims that X causes Y, the trained scientific reflex is to ask what else changed, and whether anyone actually ran the test.

Long before you reach for a calculator, a good physicist asks 'roughly how big should the answer be?' These rough, powers-of-ten guesses are called order-of-magnitude estimates, and they are a superpower for catching mistakes. If a problem asks for the speed of a thrown ball and you get 3000\ \text{m/s}, you should feel alarmed, because that is faster than a rifle bullet. Estimation also lets you attack wild-sounding questions, like how many piano tuners work in a city, or how many breaths you take in a lifetime. You break the problem into pieces you can guess, multiply them together, and accept that being right to within a factor of ten is often all you actually need.

Perhaps the deepest idea in this whole lesson is that science is built to catch and repair its own mistakes. When a scientist reports a result, others repeat the experiment, which is called reproducibility, and before a finding is published, independent experts scrutinise it in a process called peer review. A single dramatic result proves very little on its own; it is the slow accumulation of many independent confirmations that turns an idea into accepted knowledge. And nothing is ever truly final: if better evidence arrives, even a long-trusted idea can be revised. That willingness to change its mind in the face of data is not a weakness of science, it is the entire source of its strength.

It helps to sharpen the boundary from the other side. Pseudoscience borrows the vocabulary of science, words like energy, frequency and quantum, but quietly abandons its rules. Its claims are often vague enough to explain any outcome at all, which means no experiment could ever prove them wrong; when a prediction fails, the story is patched rather than dropped, and glowing testimonials stand in for controlled data. Real science does the opposite. It makes sharp, falsifiable predictions, publishes its methods for anyone to check, and treats a failed prediction as a reason to change the idea. If a claim can never be tested, or is never allowed to fail, it has stepped outside science, no matter how technical it sounds.

The best hypotheses do not merely fit what we already know; they stick their neck out and predict something new and surprising, which is exactly what makes them so convincing when they pass. When Einstein's theory of gravity predicted that starlight would bend by a specific angle as it grazed the Sun, astronomers waited for the total eclipse of 1919, measured the star positions, and found the shift right where he had said it would be. A vague idea that only agrees with old data risks nothing; a sharp hypothesis that could easily have been wrong, yet turns out right, earns real trust. The riskier the prediction, the more a successful test actually tells you.

Students often assume that when a new theory arrives, the old one is simply thrown in the bin, but that is rarely how it works. Newton's mechanics did not become wrong when Einstein came along; it became a beautifully accurate special case, valid whenever speeds are small compared with the speed of light. Engineers still use Newton to design bridges and to launch spacecraft, because in that everyday domain his predictions are indistinguishable from the fancier theory. A mature theory usually has a well-mapped domain of validity, a range of conditions where it works superbly, and progress often means discovering the edges of that range rather than demolishing everything inside it.

When people are part of an experiment, expectation itself can skew the result, so scientists build in extra safeguards. In a blind trial the subjects do not know whether they received the real treatment or a harmless placebo, which prevents their beliefs from colouring the outcome. In a double-blind trial not even the researchers handing out the pills know who got what, so they cannot unconsciously nudge the data one way. Alongside these sits the control group, which receives no active treatment and provides the baseline that everything else is compared against. These are not bureaucratic hoops; they are precisely what lets a study separate a real effect from wishful thinking.

Here is a discipline worth building from your very first day: in physics a bare number almost never means anything on its own. If a problem says a rope can hold '500', you have to ask five hundred what, since newtons, kilograms and pounds each describe a completely different rope. A physical quantity is a number paired with a unit, and the unit is not decoration; it tells you what kind of thing you measured and on what scale. Dropping units is one of the most common ways students quietly lose marks, and worse, it hides mistakes that the units themselves would have caught. So write the unit every single time, and treat a naked number as an unfinished thought.

It is tempting to picture the scientific method as a straight staircase that ends at 'the truth', but a working scientist experiences it as a loop that never quite closes. Every good answer breeds sharper questions: confirming that objects fall together in a vacuum immediately invites the question 'but why does gravity treat every mass the same?', which eventually led Einstein to a whole new theory. Each turn of the cycle narrows the uncertainty and widens what we can explain, yet there is always another digit to measure and another regime to probe. That open-endedness is not frustrating once you embrace it; it is the reason physics still has room for you to discover something nobody has ever seen.

Because these three words cause so much confusion, it is worth lining them up one final time. A hypothesis is a single testable prediction, small, specific and easy to kill with one good experiment. A law is a regularity that has survived countless tests, usually written as a compact equation that says what nature does, such as F = ma. A theory is the big explanatory framework that says why those laws hold, tying many observations and laws together into one coherent story. Notice the ladder: a hypothesis can grow into part of a law, and laws live inside theories, but a theory is never a grown-up hypothesis waiting to be promoted to a law. They are different kinds of thing, doing different jobs.