On the foundations of quantum physics Quantum physics has to be validated by experiments.  But experiments are to be interpreted in the context of quantum physics. Isn'it like a snake biting its own tail?
For example, using a scanning tunneling microscope, one can see individual atoms. This sounds like the definitive proof of existence of atoms. But the image we see is in fact computed by a computer based on the concept of quantum tunneling.
I could also take the example of quarks that cannot be directly observed or Higgs bosons. Their discovery require complex experiments whose interpretations are far from common sense and require theoretical analysis.
I agree that quantum theory and realized experiments form a consistent whole. Might a different history of science have led to a fundamentally different theory? Is quantum theory only an incidental historical construction instead of the closest approximation of the truth?
 A: No, it's not a snake biting its tail. It's a spiral. The history of science is always a spiral.
The electrons were found to produce interference patterns (experiment). So, they behave like waves. Then, other particles too should have such a behavior (theory). Then indeed, other particles should be tested e.g. silver atoms (experiment). So, this is a spiral. We have some experimental facts, we emit some theory to justify those facts, and the theory predicts new facts that we have to test for strengthening the theory, or disproving it.
And about "the image we see is in fact computed by a computer based on the concept of quantum tunneling." The microscope "sees" dark and less dark spots, or different colors.
A: I'll choose the example of quantum mechanics to try to explain how ideas are established in science.
Think of a guy called PhotonicBoom who has a theory which is, like any other theory in physics, highly mathematically dependent.  Lets call it Quantum Mechanics (QM). If the theory is mathematical, it has the advantage of relying on reason. You can use this to make predictions. In a hand-wavy way, predictions can be made by extrapolating the theory, and devising a way to test if those predictions are true. If the prediction is shown not to occur in nature, then you throw the theory away and start from scratch, it has been shown to be wrong. 
In our case of QM, people came up with the hypothesis that, for example, energy levels are quantised. They formulated the mathematics of the theory and then 'extrapolated' the theory to make predictions. And QM did predict a wide variety of phenomena, most of them extremely bizarre. 
But QM has survived every single test it has been thrown at it. This hints that we have the correct theory, whether we like it or not.
And now follows the obvious step. Making sense of QM. How do we make sense of extremely bizarre phenomena like entanglement, tunneling, superposition, state collapse etc? Well there is one easy and obvious step. We try to link concepts that we do know of, concepts that we grew up with and have hard-wired in our brains as "Status: Makes sense", even though these concepts have no reason to be more 'sensible' than other concepts.
And this is what we have done. We attached classical concepts to quantum mechanical ones, i.e think of spin, tunneling, wave-particle duality, state collapse, entanglement as an active link. All these concepts are hand-wavy, they are technically not 100% precise to what is really happening. But, they make sense, and also they are very useful at explaining what is going on at most, but not all, situations.
In your particular examples, we have attached the notion of a particle to what a tunneling microscope interacts with, but more precisely the microscope is actually detecting discrete jumps between potentials. But the atom model is an equally valid description of what is going on since these electrostatic potentials are what we call the atom in the first place! How we represent them on an image is arbitrary and is mainly used to give us intuition, it does not necessarily represent the truth.
Then you mention quarks and the Higgs. We might not be 100% sure that quarks or the Higgs exists but hey, these models have made predictions. These predictions have been experimentally verified. So what reason do we have not to believe in the existence of these particles? Just because we don't describe them with their exact technically correct terminology (that would be extremely painful, for example for the Higgs discovery: "We have successfully detected traces of the quantum excitation of the complex quantum scalar field with non-zero expectation value!") doesn't mean that the words 'particle', 'quark', or 'Higgs' are not enough to transmit the message. The point of this paragraph I guess is that the physics is there, whether we attach them some common sense term or not.
So to finish off, no QM is true (if its a precise theory or not is completely irrelevant) and has no dependence what so ever on history or the language we use to describe it. What would have been different are the interpretations, but those are just a construct to reduce the terminology to more "everyday common sense stuff" for communication reasons. You should be careful not to attach a very literal meaning to these interpretations, and definitely not to think of them as a definite truth. Things can be described in many different ways!
I hope I have been coherent in my long answer!
A: 
I agree that quantum theory and realized experiments form a consistent whole. Might a different history of science have led to a fundamentally different theory? Is quantum theory only an incidental historical construction instead of the closest approximation of the truth?

One has to separate experiments, and the theories used to  model experiments.
Experiments work with "proxies". To measure the temeperature we use a thermometer, the height of the column is a proxy of the temperature. 
The reading of the column with our eyes enters our brain through the proxy of light shining on the column of the thermometer.
The knowledge  communicated by experimenter A to experimenter B and they agree at the "reading" goes through a series of biological proxies.
Thermometers are accessible proxies. Experimenters with particle physics are using more and more nested proxies, one has to look at an LHC detector to gauge the complexity.
The reason experimenters make sense of temperatures and pressures and the even more complicated quarks and gluon signals is because of the powerful tool of mathematics. Mathematics to start with allows us to model the data, and then a more general mathematical model is sought  called a physics theory, which will make predictions to be verified by new experiments and data gathering. Once the theory does this successfully it is validated. Quantum mechanics is a well validated theory, there have not been crucial predictions that have been falsified, and the theory stands as the underlying theory in the nested world of measurements and models describing measurements.
Mathematics cannot be different in different histories ; any other historical sequence would not reach different mathematical models on the same data so the answer is: "it is the closest with the mathematical tools we have up to now description of the behavior of nature" .
A more advanced civilization might have more advanced mathematical theories, but they would encompass the  essence of the present theoretical models.
A: Bob asks a good question. The interpretations of quantum mechanics are full of the type of circular reasoning which Bob identifies. The most glaring examples come from identifying photons as particles. Even Feynmann is guilty of this, as you can see in this video clip:
https://www.youtube.com/watch?v=eLQ2atfqk2c
He takes up the question around 36 minutes into the video, where he asks: "how do we know it's corpuscular?" Here is the essence of the circular reasoning:


*

*We have an instrument called a "photomultiplier" which detects the weakest possible quantities of light.

*Here is how it works: when a particle of light hits this plate, it knocks out an electron.

*That electron is accelerated to another plate, where it knocks out more electrons etc.

*This whole chain of events goes to an amplifier which drives a speaker. Therefore when you hear a click, you have detected a photon.

*Therefore light is made of particles.
It's circular because he uses the clicks in the photomultiplier as proof that light is corpuscular, but to explain how the photomultiplier words, he relies on the "fact" that light is made of photons. 
This is not the only possible explanation of how a light detector works. In this blogpost, for example, I explane how a photographic plate can detect light without relying on the photon theory: http://marty-green.blogspot.ca/2014/12/wave-function-collapse-explained-by.html 
It's not exactly the photomultiplier tube, but its the same basic idea.
Feynman then goes on (at 39:00) to back up his argument by talking about when the light is spread over an area, it only hits one detector or another, and never both. 
It's circular reasoning because it says:


*

*If light is made of photons, this is the kind of counting statistics you should expect in this experiment with two detectors.

*In fact, those are the statistics you get.

*Therefore, light is made of photons.
The problem with this argument is that any reasonable argument premised on the wave theory gives you exactly the same detection statistics as the particle theory. Feynamnn even alludes to this difficulty when he says "...if two go off together you have too many coming and you can't resolve it."  I explain why this argument fails in this blogpost: http://marty-green.blogspot.ca/2010/02/clicking-detectors.html 
On some level, even Feynmann knows these arguments are circular. You can tell by the uncharacteristic frustration in his voice around 39:30 when he says "i don't know how much I can emphasize this...it IS particles in every way..."
A: This is really good question. The way we interpret experiments in QM is 
sometimes confusing, another example is the vaccum state, some physicist talk about this using the concept of "virtual particles".
However this is not a problem only of QM, in all the other theories we do the same thing, every experiment is interpreted using the terms of the theory itself.
This "problem" is something related with the way science works.
On the other side, one can say that the sole purpose of physical theories is to reproduce the results, the way this is done doesn't matter.
QM can be formulated differently(historically this is the case) but every experimental result needs to be the same. 
With respect to the last question:
"Is quantum theory only an incidental historical construction instead of the closest approximation of the truth?"
The relation between physical theories and "truth" is not that simple.
