The laws in Physics summarized many experimental results after measurements or general observations. One simple law is that every mass is attracted by the Earth and it follows at the same acceleration independently of how big it is.

    This is not at all a model, although the any physical model must follow them and even they serve to create them. The previous law allowed Einstein to think in the Principle of Equivalence to interpret the gravitation interaction as a geometrical change of Minkowski's flat space-time in another Riemann provided with curvature. The important thing is that this theory allow us to predict new explanations to observations. As can be the blueshift of the light falling in a gravitational field or redshift if it goes against the gravitational potential due to have different time rates associated to different points of the gravitational field, which would be impossible to explain in Newton's gravitation. 

Maybe you could start by giving an example of any statement that is ``true by itself''? Let me emphasise that I do not say  this to be nasty, or sarcastic: I truly believe clarifying what you mean by ``truth'' can help answer the question, which I hold to be important.

True by itself might be limited to tautologies, where what we think of as truth is a relation of some kind.  And what we call true in science / physics can also be provisional, or what we might call appropriate until a more accurate theory comes along.

It is a good point F.Leyvraz and Edward, I am not really sure what does the word "true" mean. But i can think of a true law as an ultimate law which is always correct and will be always correct, so nature itself follows the law and not vise versa.

  When you say true, you might mean that you have made a choice between possibilities and that is not the case in Physics. You just choice facts and summarize them; for instance, you can tell when the sun arise at a certain date and place with a certain accuracy (every physicist learns how to measure and calculate accuracy). This is not "true" or "false" because it is not an statement within a "logic".

  Finally, all the physical laws have remain valid and will remain in the future. They only can change as being a special case of others more general that can be discovered.

We must distinguish between law, model, and theory.  A law is just a relation between two or more measured quantities, in a given condition.  As such, it is true if verified with a high precision, which is also called "valid."  If a law remains valid, on the other hand the definition of the quantities may vary drastically according to the theory in which they appear, and the domain of validity may be restricted.

   Dear Claude,

   I do not understand your answer, over all the sentence:

As such, it is true if verified with a high precision, which is also called "valid."

   The concept "true" is without meaning in this context, because there are not a "false" physical law, independently of the accurecy that is measured. In the worse case, this law would not exist.

It is true that the acceleration is proportional to the force, for velocities very small with respect to the speed of light, and with an error smaller than 0.1%.

Dear Daniel, Thanks for you answer. How would we know that the laws of physics will remain valid ? for centuries we believed in the determinism and the Newtonian laws, while today we fully believe in the Quantum physics and its probabilistic nature ! It might be a model that will keep evolving. 

Dear Claude, thanks for you answer. I think verifying a physics law with high accuracy does not mean it is true in necessary. I think that a true law need to be be self-proving regardless of experimentation error.

Akram Hourani ,

I take a theory to be a statement about Nature.  And I do not take the theory to describe Nature in itself.  It is not because the theory is not the ultimate theory or that because the theory is only approximative.  It is because of the essential difference between what the theory predicate about Nature in terms of the theory itself and Nature itself.   I is like the difference between a map and the territory.  It is not by increasing the accuracy of the map that the map become the territory.  The map or the theory is a guide to how I can move on the territory and it does not describe everything that can be described of what exist on this territory.  So theories are created by us for interacting with the world, our getting around the territory but are not the territory description but usefull tool of interaction with the territory. 

Imre Lakatos would maintain that theoretical laws are true because they embrace mathematical models capable of predicting new experimental results in non-ad hoc way

Akram, determinism isn't a law, it is a first principle.  A law in physics is a fact, on which theories and models are built. It seems your question is about theories, and not laws.

Rinat, a theorie can both have a predictive power, and eventually fail an experimental test. It can then not contain any truth, even a partial one, since its concepts can have changed.

For example, Newtonian mechanics could predict the existence of unknown planets, but failed for velocities comparable with light velocity. All what Newtonian mechanics said about space and time is false because these have been redefined in special relativity. Newtonian gravity is an approximation of general relativity only because it rests on observations, not because of some obscure metaphysical property.

Rinat, I think you hit a very important point. The deducibility in mathematics makes it the most trustworthy science, example: Pythagoras's was true and will remain true not like Newtonian gravity. However the dilemma is whether mathematics itself is true or not !! does our world really follow mathematical laws, or they happened to be good enough to be describe it. This might sound ridiculous, but the truth is that no one ever can prove that their exist a contentious space-time or a contentious mathematical functions.

   Dear Akram,

   The three Newton's laws of Physics are absolutely right today. The students of Physics follow to study them because they summarize important facts which are going to happen independently of the theory that explain them . They are not going never to be "false", this is a context out of the what is meant by physics laws. Another thing is a theory as Newton's one that nowadays is just a special case of General Relativity (for gravitation) or Special Relativity (for Kinematics or Dynamics)

   By the way, Pythagoras theorem has changed in time, although it belongs to realm of Mathematics. It is not the same in an Euclidean space as in Minkowski, Riemann or Lobachevsky geometries. The concept of geodesic or even distance is quite different.

Claude,

it is impossible to refute a theory of such level of generality as classical mechanics (or classical electrodynamics).Newton's laws refer only to the basic ideal model (the Fundamental Theoretical Scheme) that depicts the relations between such abstract ideal objects as a Material Point,  Force and an Inertial System of Reference. The key point is that if one wants to describe experimental data s/he has to construct theoretical objects of partial theoretical schemes (and after that to pass to empirical schemes). And there are no strict and definite rules to do that, as Thomas S.Kuhn had exhibited in 1969 edition of his "Structure of Scientific Revolutions".For instance, to arrive at the oscillator equation, one has to make some important assumptions on the character of force, etc.

P.S. The second obstacle is of course the Duhem-Quine thesis

     Dear Rinat,

    Let me summarize the Newton's laws of mechanics:

        First law:

   In an inertial reference frame, an object either remains at rest or continues to move at a constant velocity, unless acted upon by a force.

       Second law:

    The vector sum of the forces F on an object is equal to the mass m of that object multiplied by the acceleration vector a of the object: F = ma.

        Third law:

     When one body exerts a force on a second body, the second body simultaneously exerts a force equal in magnitude and opposite in direction on the first body.

      Where do you find concepts as Material Point ? They are only for teaching and the above laws only summarize observations till XVII century in mechanics and needed to take as assumptions for his mechanics. Whera are the refutions to them, even today?

Rinat, you know that a theory that is not refutable (falsifyable) is not considered a theory at all.  Of course classical mechanics has been refuted, in the atom.  Electrodynamics has been quantized too.

Dear Daniel,

"inertial frame of reference" is an obvious example of idealization. It is connected with a body with all the interactions cut off. 

Claude,

it was Sir Karl Popper who had conjured up the principle of falsification. For modern post-Kuhnian and post-Lakatosian philosophy of science the principle is as old as mountains. May I advice You to steep in Imre Lakatos's "Falsification and the Methodology of Scientific Research Programmes" or in Elie Zahar's "Einsteinian Revolution"? Michelson-Morley experiment did not refute the classical mechanics.Nor did the Lummer-Pringsheim or Rubens-Kurlbaum ones.

That's right, the mere levels in the atom shattered classical mechanics into pieces. Tell the story to Zahar and Co, or perhaps are they philosophers of the 18th century?

Nope.Lakatos died in 1974, while his pupil Elie Zahar is still alive.But what is more important is that one cannot "refute" classical mechanics alone. It is "refuted" only in conjunction with classical electrodynamics (the Duhem Quine thesis). Usually in the cases like that one has to decide which part of the pair should be thrown out.

That's not true, relativistic classical electrodynamics still works, the second quantization is thought superfluous by many people.

   Dear Rinat,

   For you to choice an inertial observer of a phenomenon means that you make or accept a theory? All the measurements must be referred to a inertial observer (Galileo or Lorentz) and without that all the Physics could be no-meaning. False is a concept that you are find very rarely in papers or books devoted to real physics and when this is done, always is with respect to a theory and no with respect to laws (based only in facts).

   By the way Classical Mechanics or Classical Electrodynamics are everyday used and studied with publications on them. They are so alive as Relativistic Mechanics or Quantum Electrodynamics. None of them is a false theory at all, they only apply to different physical conditions. Notice that the classical mechanics or electrodynamics belong to the curriculum of every physicist while Relativity and Quantum Electrodynamics don't do it because the conditions of applicability are much more restricted.

All I want to say is just that.It is impossible to refute such extremely complicated and abstract theories as classical mechanics, classical elrctrodynamics, quantum mechanics, etc.The crux of the argumentation consists in the statement that the connections with experimental results are very intricate since each mature theory consists of many pieces with weak cohesions between them.More exact answer is given in my book "Reconstruction of Mature Theory Change" (Peter Lang, Frankfurt, 1999).

Classical mechanics has been refuted as a theory.  It is still useful as a model, of course, but that's not what we are speaking about.  Classical mechanics isn't an approximation of quantum mechanics, they are different theories, it is just a rule of thumb.

Classical mechanics was not refuted. It has been exchanged by relativity and quantum mechanics cause they had posessed more mathematical powers to solve the new puzzles

   If we work at a level of 6.63 x 10^(-34) Js as action, then it is necessary to employ Quantum Mechanics or Quantum Field Theory. But if the scales of action are much higher, as it happens with most macroscopic phenomena, then Classical Mechanics or Classical Electrodynamics are the proper theories to employ.