Quantum Non-Demolition Experiments are Impossible

To salvage the application of the Michelson interferometer for gravitational wave measurements, the LIGO team has resorted to all sorts of ways to avoid the Heisenberg Uncertainty Principle (HUB). Things like Standard Quantum Limit (SQL), Quantum Non-Demolition (QND) experiments, and "back-action" are completely meaningless. All this is the result of confusion between the wave-particle duality of quanta. To talk of the reflection of a photon from a mobile mirror, and to calculate its change in frequency as Braginsky et al do in \it{Quantum Measurement} is the wrong picture to use. Rather, it is the wave picture that is appropriate in which the wave recoils from its target, and to speak about the uncertainty in the wave number that results. Anywhere the wave function in momentum space is appreciable $\phi(k)$, it is impossible to control the momentum of a body to within $\Delta k$, where $k$ is the wave number, while anywhere the wave function $\psi(x)$ is appreciable, there will be an uncertainty in its position $\Delta x$. One cannot consider that "kicks" the position of a particle due to momentum fluctuations, and consider there are no "kicks" in the momenta due to fluctuations in the position of a particle by saying there is no "back action". In fact, back action or back reaction or whatever you want to call it is completely foreign to quantum mechanics. If force cannot be defined (Newton's second law), then surely equal action and reaction (Newton's third law) cannot. Since uncertainty in conjugate variables are determined by the minimum action, $\hbar$, amplitude and phase of an oscillator or an electromagnetic wave are not conjugate variables. The claim that two conjugate variables, $E$, (instead of amplitude) and phase $\varphi$ are related by $$\Delta E\cdot\Delta\varphi=\hbar\omega/2$$ is definitely incorrect since $\Delta\varphi=\omega\Delta t$ so that $$\Delta E\Delta t=\hbar/2$$ is the correct uncertainty relation. The claim that $\Delta E=\hbar L/2\tau\Delta x$ is "not constrained by SQL since by taking long enough time for the measurement to be made, one can obtain any desired sensibility" (Braginsky et. al.) shows a complete insensibility for the meaning of what is a quantum measurement. It is tantamount to considering the amplitude of radiation pressure, given by $\Delta A\propto 1/\omega^2$ to be an error that is not constrained by SQL since by taking a high enough angular frequency, $\omega$, it is possible to reach any desired sensibility. QND measurements [cannot] be made on: 1. observables that are conserved during the free motion of the body; 2. observables in which there is no perturbation by its conjugate variable; 3, observables whose conjugate variables are perturbed according to the uncertainty principle. There are clear distortions and misunderstanding of the nature of quantum measurements. The whole idea of converting a Michelson interferometer making position measurements into a "speed meter" because velocity is a QND variable is ludicrous. Moreover, if amplitude and phase are non-conjugate variables that do not satisfy HUP, what is the meaning of "squeezing" one of them and "stretching" the other? Finally, the introduction of correlations between conjugate HUP variables converts an equality into an inequality so that the proportionality between the uncertainty in one and the uncertainty in inverse of the other is lost.