I have a question. I have a problem that when I learn science, I like to think 'how do they know this is right?'.

When we learn physics or science from a textbook, we read and understand it, and then we have exam. But how do you know what is given in a textbook or said in research is correct? Because we didn't do this experiment, or we don't know what the process is to get this scientific result by ourselves?

So, should I trust it and accept it? If not, what should I do?

  • 17
    $\begingroup$ Almost every science class has a lab portion. Doing those should be step 1. Further, up to high school level, pretty much almost all the physics you learn can be learnt by doing home-made experiments. Try those out yourself as well (I’m sure you can find instructions for these on YouTube somewhere). While such first-hand tests should always be prioritized, you can also alternatively read up on the history of science to understand how exactly a particular model or theory was developed. $\endgroup$
    – Prahar
    Commented Sep 17, 2023 at 8:33
  • 8
    $\begingroup$ Are you questioning the very concept of learning things second hand from reliable authorities? Do you believe history books despite all their claims being unverifiable? $\endgroup$
    – RC_23
    Commented Sep 17, 2023 at 18:30
  • 2
    $\begingroup$ I find it interesting as an engineer how people can be skeptical of science (which is fair), but also for instance walk in to a high rise building, watch helicopters fly, use electricity and so on. There's no 'well it must be magic' rationalisation for that stuff. $\endgroup$ Commented Sep 17, 2023 at 22:06
  • 1
    $\begingroup$ @LamarLatrell People are also skeptical of things that causes them inconvenience or worse. In this context, it may not be an actual, genuine concern with science, but rather feeling that education is "useless". If OP doesn't have to do homework and take tests, I suspect the concerns would go away. $\endgroup$
    – Nelson
    Commented Sep 18, 2023 at 3:11
  • $\begingroup$ You are asking a great question! There are no truths in science, only theories, waiting for improvement. What you have in undergrad books is not “true,” it's basic uncontroversial theories, not reflecting the latest knowledge as to when these theories work and break, and e.g. two main incompatible proposed solutions. Always question what you learn, look for tiny discrepancies. As an undergrad, you don't learn current frontier physics. You learn necessary basics, but, the most important, how to think like a physicist and how to teach yourself more physics later, when you're on your own. $\endgroup$ Commented Sep 22, 2023 at 18:22

10 Answers 10


Strictly speaking, you don't. All science is tentative and subject to change should new evidence or new insights arise. All conclusions are not only open to challenge, but it is part of the process that they are periodically challenged; systematic scepticism even of long-established orthodoxies is part of science's 'immune system'.

However, nobody has the time or inclination to check absolutely everything for themselves. So we build social processes and conventions that give us some assurance and reason to believe.

The thing to look out for is to check whether there are other people challenging and checking, who you consider competent to check and motivated to challenge, who are able to raise the alarm to the public if they see a problem, who will be taken seriously, and who you would expect to have heard about if the situation arises.

Scientific conclusions gain in confidence by surviving sceptical challenges. It is a case of 'survival of the fittest'. 'Authority' and 'expertise' of the authors or publishers mean nothing. What you need to see is people trying their hardest to knock the ideas down, and failing. And you need to see that if a hypothesis is challenged and its advocates cannot fully answer the challenge, it is subsequently treated as 'disputed' and uncertain.

While few people may be capable of checking the science for themselves (either because they can't do the maths, or because they can't afford their own 27 km particle accelerator), they can look at the human institutions around the science and see if they're applying the scientific method. Are sceptics and dissenters welcomed and encouraged?

As for textbooks, thousands of people are constantly reading them, learning from them, checking them, and asking questions here about the bits they don't understand. If there were obvious flaws, they would have been detected by now. Students are not the most expert reviewers, and are often discouraged from challenging the textbook in class, so it is of limited value, but it's usually pretty good. It's not perfect, of course. Yes, there are mistakes in textbooks. Yes, there are textbooks on entire subjects that are later debunked. Yes, there are entire areas of science where the application of the scientific method is 'shaky', to say the least. (See the 'replication crisis'.) But the human institutions and conventions around textbooks are such that in many cases we can expect only the best and most reliable to survive.

You start by checking and trusting the scientific process, that it is being properly applied, and as you learn more science, you gradually transition to being able to check more and more of the science itself for yourself. You never do reach a point where where you can check it all, where you absolutely trust it. If you ever do, you've gone too far! You're not doing science any more.

Have you read Richard Feynman's essay on 'cargo cult' science? It talks about what to look out for.


Where possible, you should check the claims of textbooks for yourself. It is certainly possible to reproduce experiments carried out by Galileo, Leonardo, Newton, Faraday and so on, with the added advantage that today we have easy access to accurate timing and measuring devices, precisely ground lenses and prisms, batteries, multi-meters etc.

Where you cannot directly verify the claims of a textbook, you can cross-reference the claims of one textbook against another - one golden rule of science is never accept anything that has only a single source. For "modern" science (in the last 150 years or so) you can read the original papers of Einstein, Rutherford etc. Where you cannot reproduce an experiment, then you are ultimately placing your trust in the scientific method of publication and testability of objective results, as overseen by the scientific community as a whole.

  • 2
    $\begingroup$ It's probably not realistic to directly confirm any more than a handful of very basic experiments, since it takes a lot of time and often decent quality equipment. But checking consistency with everyday experience and what one already knows is a good exercise to go through constantly while learning physics. $\endgroup$
    – user34722
    Commented Sep 22, 2023 at 15:13
  • $\begingroup$ One thing one should also address is that the people you mentioned built their own measurement devices, while today as you mention we rely on modern measurement devices, which, by the scientific method, we should also check that they work as we want. This is a step-by-step process, you won't be able to understand everything accurately, but you can start somewhere and work your way up towards more accurate and precise levels of understanding. $\endgroup$ Commented Sep 25, 2023 at 8:06

To give you a satisfactory answer to your question would take a book -- or perhaps a six-part series from 1987 on PBS. And you are in luck, it's free on YouTube:

The Ring of Truth: How We Know What We Know

The series was written and hosted by Philip Morrison, a professor of physics at MIT, a key contributor to the Manhattan Project and the man whose 1958 paper gave birth to gamma ray astronomy. His PhD advisor was J. Robert Oppenheimer.

Morrison was particularly gifted in explaining principles of physics and how one discovery or principle leads to another. If you watch this series you will learn "how we know what we know" in physics.


It depends on what you are learning. If you are learning about for example the electromagnetic wave equation and are wondering where it came from, then you need to look at the derivation from Maxwell's equations and try to understand every step there. This however does not prove that it is correct in regards to what happens in nature. Here comes the importance of experiments. Some experiments as Prahar states can be easily done at home and some you can do at your school. Other experiment which require expensive and or large equipment you will simply need to trust the data which is provided by the ones operating it.


I encourage everybody to perform experiments for themselves. If you google "physics home made" (or similar) you will find many cool experiments, which help to build up confidence and therefore frees the path to a deeper understanding. This is why in high school classes and universities lectures the experiments are usually a key part of the presentation. In addition, physics students must reproduce many famous experiments as this helps to get used to the equipment. During these lab hours I experienced several times that pupils basically contradicted a well-known result. If the tutor was dedicated he/she pointed out the mistake. However, in some cases the students were left alone and became rather frustrated. Thus, if you want to try experimenting at home, I strongly recommend joining forces with others to keep the spirit high.

I believe that especially in high schools a key part of performing experiments is the visualisation. Thus, a second great resource is available: Simulations using computer programs. You don't have to write the code yourself, but there exists visual presentations for many topics.

Finally, I can strongly recommend a visit in a science museum. I visited the one we have here with my eight year old daughter and her friend. They had all kind of "do it yourself" experiments. They loved it and we had to revisit several times. I didn't explain anything to them, so I know the phenomena themself can be fun. They enjoyed the experience of experimenting and wonder. Hope you do as well.


A physics textbook is a boiled-down version of the results of many, many experiments which all point towards the truth. In each experimental case, the results of the experiment are exhaustively cross-checked and picked apart by the rest of the science community whose task it is to identify and eliminate errors. This is an essential part of the scientific process and physics textbooks are naturally focused on providing the results of that process, and not on all the arduous labor that went into obtaining and verifying those results.

This is why you can rely on the textbooks to furnish the best truth available at the time, without having to exercise faith in their correctness.


In addition to what the other answers have said about trusting the wider scientific community being necessary at some level (since none of us have the time or resources to reproduce all experiments), I want to also emphasize the importance of thought experiments. Novel scientific theories were often formulated by imagining interesting scenarios and taking new insights to their logical extremes. For example, Einstein's thought experiments with trains moving close to the speed of light were important for understanding his special theory of relativity. It's one of the jobs if a theoretical physicists to apply a theory to interesting and extreme scenarios and look for contradictions or bizarre predictions.

Applying textbook knowledge to imagined scenarios is a way to check internal consistency of a theory. A theory being consistent doesn't prove that it is correct, but physics is very logical and it's very difficult to formulate a logical theory without creating a contradiction. So if something is consistent with what you already know, the thought experiments you can conduct in your head, and relevant real-world experiences you can think of, then that should at least increase your confidence in it. I have occasionally found minor errors in textbooks by doing this. But more importantly, I very often find flaws in my own understanding!


"The best teachers are those who show you where to look, but don’t tell you what to see." - (maybe) Alexandra K. Trenfor

The goal of a physics (any science) book is not to claim, but to explain. I teach maths and are often asked: "Why do you show proofs for the methods at the lectures when we aren't gonna do the proofs but are gonna use the methods?"

That questions is answered by your question. If you don't see the proofs, then you aren't convinced that methods are indeed correct. Then you are blind. The goal of any student must be to understand, not to believe. Otherwise a falsehood will persist.

To avoid the issue you are feeling with doubts about the truthness of what books claims, texts and lectures and lessons in general will typically use a set of verification appoaches such as...

... showing references that you can look up to confirm the claim (and the references will contain references etc.)

In physics the starting point will be some experimentation that confirms a hypothesis, which then is confirmed further by several more experiments by other people (peers), and it should be possible to reach backwards in the array of references until the original publication of the original experiment is reached.

If you still don't believe this original publication of the original experiment, then you must be...

... redoing the experiment

Reproducability is a key part of the scientific method - we won't universally trust the outcome of an experiment as confirmation of a hypothesis or observation unless it can be reproduced by others.

Of this reason, experimentation is essential for learning what human kind knows about e.g. natural sciences. At your school or study, laboratory work is always (ought to always be) mandatory of this very reason. You have to see with your very own eyes that the claimed physical phenomena happen and that the claimed methods work.

... working their way through proofs, showing the points as connected pearls on a string, rather than just getting to the point and provide methods, claims and facts

This relates to the question from a student that I started with. A lecture should not start claiming things without at least refering to where verification can be found or without at least initiating the road through a proof or train-of-logic that will let you reach the conclusion yourself.

At the end of the day: Never believe, find understanding. Don't trust a teacher/book by their word, but use their words to guide your eyes and thoughts.


I think some of the challenge here is that we typically use a scientific-realism voice in textbooks. They state "gravity does this," "heat does that." Which leads to the sort of questions you are having. A more nuanced point of view would be to say "gravity is modeled like this" or "heat is modeled like that." This draws a distinction between then physical world around us and the mathematical models we build to describe it.

With this distinction, we can use standard verification and validation logic. We can ask a key question, "what is the intended use of this model?" We can as "what are its assumptions and limitations?"

Newtonian mechanics was originally thought to explain all of physical motion. It's intended use was "everything." Then we made some observations and realized something was amiss. It looked like things were "dragging" just a little. This lead us to develop the theory of relativity, another model. We now claim the intended use for Newtonian physics is "physical motion when things aren't moving really fast." And we can go through the same logic for quantum mechanics. Now, in an era where we have models of relativity and quantum mechanics, we claim the intended use for Newtonian physics is "physical motion when things aren't moving really fast and aren't really small."

If your textbook isn't claiming accordingly, then they're probably just trying not to overload you with the astonishingly detailed models we've built over the years. Start small. They're trusting that you won't have to fly a spacecraft at relativistic speeds before they get a chance to amend their statements. (As you get further into physics, the text books will be more explicit with their assumptions, so you should see more accurately claimed intended uses as you progress)

When it comes to modeling, we constantly weigh the cost of knowing versus the risk of assuming. You don't know a hockey puck is going to follow the path calculated using Newtonian physics. You can't truly know anything about the future (some would argue you can't know anything, period). But you can assume it, and act accordingly. If you're a hockey player, you might assume physics will do what you're used to, and "skate to where the puck will be." In this case, the cost of knowing is high (you would have to wait to see where it went, and might miss a goal), and the risk of assuming is low (at worst, you expended some extra calories hustling).

The only way to answer your "trust" question is to consider your own balance of costs of knowing versus risks of assuming. It's different for each person and different in each situation. If you're doing manned spaceflight, you assume quite a lot less than we did with our simple hockey puck.

When it comes to science, and more importantly the scientific community, we make a very strong argument that the social structures and institutions we put in place let you "buy down" the risk of assuming. It's a purely social thing, but basically other people have put a lot more on the line claiming that models are correct. When you're worried about the physics of steering on a wet road, you can have some comfort that the risk of these physical models being wrong is low because companies have literally put billions of dollars on the line, making the same assumption. A professor who has put their academic reputation on the line in defense of a theory is (circumstantial) evidence that a model is credible. Many professors putting their reputation on the line starts to be even more reliable.

But in the end, its all up to you. Don't "trust" the models, as much as understand the cost of knowing versus the risk of assuming. And then we can talk about why the risk of assuming is very low for reason A or reason B, without having to invoke "trust."


THe only way to know is by doing the experiment. There really isn't a substitute. Logical thought, exclusively, leads to multiple theories that typically result in conflicting results; and the only way to choose among them is experiment.

The only other alternative way I can think of would be to understand the consequences of your physics textbooks. Airplanes fly, semiconductors work to allow your computers to function, etc....You can draw a straight line from the operation of these things to your physics textbooks.


Not the answer you're looking for? Browse other questions tagged or ask your own question.