What is this “it” that you think QM says is changed? And why do you say “permanently”?
And what is a “property” of something anyhow?
I think what most physicists consider to be changed when they become aware of how a system has influenced some measurement apparatus is just their relationship to that system rather than the system itself. In particular, if they have a good theoretical model for how the system evolves, then they may be able to predict the results of future similar “measurement” processes (so long as the system does not interact with anything else in the meantime). The repeatability (or at least predictability) of such outcomes is analogous to that of measurement of some property of a classical system, but quantum systems don’t seem to have fixed “properties” in the same sense, so we generally refer to the results of measurements as “observables” rather than “properties”.
And there may be “complementary” observables about which they necessarily have less information after that first measurement (and for which a subsequent measurement of those complementary observables leaves us back in the dark about what to expect for future measurements of the original one).
But the question of how we came to accept this lack of fixed classical “properties” is a good one.
It’s a long and interesting story that I can’t possibly do justice to here. But it starts with some observations by Werner Heisenberg in the 1920s about the way looking at a small object, so as to determine its position very precisely, involves shining a very short wavelength light on it; and according to Einstein’s interpretation of the Planck’s idea of electromagnetic energy being restricted to discrete jumps (which get bigger as the wavelength gets shorter) interaction with shorter wavelength light causes a larger change in momentum that is not completely predictable because of the size of the lens needed to reliably collect a photon. It soon became clear that this “observation effect” was independent of any particular observation process and was actually a fundamental principle inherent in all the attempts that had been made so far (eg wave mechanics, matrix mechanics, Hilbert space etc (which are all the same really)) to come up with theories of matter that match our observations (of atomic spectra, what happens when streams of small particles are sent through pairs of slits in a barrier, and so on).
So by the 1930s, most physicists already accepted that the classical idea of fixed “properties” would have to be abandoned. But some (such as Einstein, Bohm, and others) considered this premature and suggested looking for “hidden” fixed properties that could restore certainty if only they were known, and would explain the uncertainties of quantum mechanics as just the result of a lack of knowledge of their actual values.
Various attempts have been made to predict the results of quantum mechanics as averages over unknown values of such hidden variables. But none of them provide any way of actually measuring the hidden variables, and so they are more like a theoretical tool (just like the wave function itself) rather than anything “real”.
Also, it turned out that all of the ways people discovered for actually getting the same results as quantum mechanics from a hidden variables theory (such as the ”pilot wave” theory of de Broglie and Bohm) involved instantaneous interaction between widely separated parts of the system, and so if the hidden variables were “real” this would allow things like faster than light communication (and if the speed of light is independent of that of the observer then this would lead also to things like time travel and the grandfather paradox).
For many years people wondered if perhaps it was just due to not being clever enough and whether perhaps someone might eventually discover a “local” hidden variables theory which didn’t have the ftl and time travel problem. But over time that has come to seem less and less likely,
In 1964 John Bell noted that, even for a very simple system, in any “local” hidden variables theory there would have to be certain relationships between the probabilities which are violated in quantum mechanics. Since then, various experiments have been done with ever more precision to test the quantum predictions (leading to a Nobel prize in 2022 for Alain Aspect, John Clauser, and Anton Zeilinger). So it has become more and more certain that no hidden variables theory can avoid the ftl effect and the paradoxes that would arise if the hidden variables could ever be observed.
Perhaps even more important is the fact that, according to a theorem of Gleason from 1957, and another (proved by Bell in 1966 and extended by Kochen&Specker in 1967), in order to match the quantum predictions the hidden variables would have to be “contextual” – which roughly means that they would depend not just on the system but also the observer – and most of us think this really does put an end to physics being described by any fixed “properties” of the system itself.
And what is a “property” of something anyhow?
I think what most physicists consider to be changed when they become aware of how a system has influenced some measurement apparatus is just their relationship to that system rather than the system itself. In particular, if they have a good theoretical model for how the system evolves, then they may be able to predict the results of future similar “measurement” processes (so long as the system does not interact with anything else in the meantime). The repeatability (or at least predictability) of such outcomes is analogous to that of measurement of some property of a classical system, but quantum systems don’t seem to have fixed “properties” in the same sense, so we generally refer to the results of measurements as “observables” rather than “properties”.
And there may be “complementary” observables about which they necessarily have less information after that first measurement (and for which a subsequent measurement of those complementary observables leaves us back in the dark about what to expect for future measurements of the original one).
But the question of how we came to accept this lack of fixed classical “properties” is a good one.
It’s a long and interesting story that I can’t possibly do justice to here. But it starts with some observations by Werner Heisenberg in the 1920s about the way looking at a small object, so as to determine its position very precisely, involves shining a very short wavelength light on it; and according to Einstein’s interpretation of the Planck’s idea of electromagnetic energy being restricted to discrete jumps (which get bigger as the wavelength gets shorter) interaction with shorter wavelength light causes a larger change in momentum that is not completely predictable because of the size of the lens needed to reliably collect a photon. It soon became clear that this “observation effect” was independent of any particular observation process and was actually inherent in all the attempts that had been made so far (eg wave mechanics, matrix mechanics, Hilbert space etc (which are all the same really)) to come up with theories of matter that match our observations (of atomic spectra, what happens when streams of small particles are sent through pairs of slits in a barrier, and so on).
So by the 1930s, most physicists already accepted that the classical idea of fixed “properties” would have to be abandoned. But some (such as Einstein, Bohm, and others) considered this premature and suggested looking for “hidden” fixed properties that could restore certainty if only they were known, and would explain the uncertainties of quantum mechanics as just the result of a lack of knowledge of their actual values.
Various attempts have been made to predict the results of quantum mechanics as averages over unknown values of such hidden variables. But none of them provide any way of actually measuring the hidden variables, and so they are more like a theoretical tool (just like the wave function itself) rather than anything “real”.
Also, it turned out that all of the ways people discovered for actually getting the same results as quantum mechanics from a hidden variables theory (such as the ”pilot wave” theory of de Broglie and Bohm) involved instantaneous interaction between widely separated parts of the system, and so if the hidden variables were “real” this would allow things like faster than light communication (and if the speed of light is independent of that of the observer then this would lead also to things like time travel and the grandfather paradox).
For many years people wondered if perhaps it was just due to not being clever enough and whether perhaps someone might eventually discover a “local” hidden variables theory which didn’t have the ftl and time travel problem. But over time that has come to seem less and less likely,
In 1964 John Bell noted that, even for a very simple system, in any “local” hidden variables theory there would have to be certain relationships between the probabilities which are violated in quantum mechanics. Since then, various experiments have been done with ever more precision to test the quantum predictions (leading to a Nobel prize in 2022 for Alain Aspect, John Clauser, and Anton Zeilinger). So it has become more and more certain that no hidden variables theory can avoid the ftl effect and the paradoxes that would arise if the hidden variables could ever be observed.
Perhaps even more important is the fact that, according to another theorem (proved by Bell in 1966 and extended by Kochen&Specker in 1967), in order to match the quantum predictions the hidden variables would have to be “contextual” – which roughly means that they would depend not just on the system but also the observer – and most of us think this really does put an end to physics being described by any fixed “properties” of the system itself.