No Ghost in the Machine

Abstract

The conscious mind may turn out to be a virtual reality simulation that is largely illusory.

End Notes

I have argued here that direction and intelligence in nature follow from adaptation in a complex dynamic world based on the foundations of classical physics, without the need for anything either quantum or supernatural. Elsewhere, I have argued that the foundations of quantum mechanics have been profoundly misunderstood, and that quantum mechanics properly provides a unified foundation for all of physics. The comments below describe some implications of this for the nature of time and for the future of computing.

A. Quantum Waves and the Nature of Time

Time is central to physics, biology, psychology, and computation. But is it the relativistic spacetime of Einstein, or the subjective time of our internal chronometers? I have suggested [12], and in an earlier FQXi essay [9], that time is defined on the microscopic level by quantum waves, and everything else follows from that.

In the Newtonian clockwork universe, time is abstract and universal. In Einstein’s universe, time becomes part of abstract spacetime, which is non-universal and inhomogeneous. Einstein focused on light waves, any one of which has a frequency f and a wavelength λ, and a universal speed of light c = fλ. But taken collectively, light waves (electromagnetic or EM waves) can have any value of f; there is no characteristic frequency for EM waves. But consider a quantum wave (or de Broglie wave) for an electron of mass m_e. This has a minimum frequency f = m_ec²/h, which represents a characteristic frequency f_c (h = Planck’s constant). Similarly, an electron has a characteristic length, the Compton wavelength λ_c = c/f_c = h/m_ec. So an electron is a natural clock and ruler which defines the local calibration of time and space. These change in a gravitational field, giving rise to the trajectories of general relativity. These trajectories may be computed using classical formalisms, without reference to abstract spacetime metrics.

Any system of physical units has three independent units (plus the fundamental electric charge e); for the standard metric system (SI) these are MKS, for meter, kilogram, and second. In the universal system described above, the natural units are f_c, λ_c, and h. Quantized spin provides the basis for h, and is Lorentz-invariant. Other parameters are defined in terms of these: c = f_cλ_c and m_e = hf_c/c².

Within this picture, an electron is a real extended wave in real space, with quantized total spin. There is no intrinsic uncertainty, and no entanglement. Composites such as a proton or an atom are not quantum waves at all; they are bags of confined quantum waves, which inherit quantization from their constituents. Such a picture provides a simple, unified system with local reality all the way from atoms to galaxies and beyond.

Time is not a physical dimension; rather, it is simply a parameter that governs motion and change. Our most accurate clocks are atomic clocks, which inherit their properties from the fundamental electron clock. This defines the time of classical physics, which in turn defines our macroscopic clocks, as well as our biological and psychological clocks. Not all of these are quite so accurate, but they derive from the same physics. The direction of time follows simply from increasing microscopic entropy and macroscopic adaptation.

B. Quantum Computing is not the Future of Computing

There has recently been much attention to the potential of quantum computing, which promises the ability to perform computational tasks that are virtually impossible for any classical computer, regardless of the architecture. It has even been suggested that modern classical supercomputers will soon become obsolete due to “quantum supremacy”. This essay proposes that classical computers with non-classical (neuromorphic) architectures may enable true artificial intelligence and even consciousness—why not quantum computers for this role? The reason is that I do not believe that the promised quantum computing is possible, on fundamental grounds (see [11]).

The key point is that the promised exponential enhancement in performance is based on the presence of quantum superposition and entanglement among N coupled quantum bits (or qubits), as implied by the orthodox Hilbert-space mathematical formalism for quantum states. I have questioned whether this mathematical formalism is correct [8, 10], and suggested some new experiments that can address this question. One can see the essential role of entanglement from the following simple argument. In any classical computing architecture, one enhances performance by the use of parallelism in hardware bits. If the hardware consists of N parallel bit slices, it can operate (ideally) N times as fast. In sharp contrast, for a quantum computing system, if there are N entangled qubits, this enables an effective computational parallelism by a factor of 2 ^N, which increases exponentially. For example, if N = 300, 2 ^N is greater than the number of atoms in the known universe. Clearly, no classical supercomputer with merely billions of bit slices can compete. The ability to obtain exponentially scaled equivalent performance from linear growth in hardware provides the entire motivation for quantum computing.

Such a claim of exponential performance is fantastic, in both senses of the word. Carl Sagan once said [20] that “Extraordinary claims require extraordinary evidence”. In fact, no such evidence exists, and people should be extremely skeptical. The primary reason why this has been accepted is that the orthodox theory of quantum mechanics, obscure and confusing though it may be, has been defended by most of the smartest minds in physics for decades. But early in the 20th century, both Albert Einstein and Erwin Schrödinger questioned the foundations of quantum mechanics. Recently, Weinberg [25] has also questioned these foundations.

Given the billions of dollars that are now being invested in quantum computing research by governments and corporations, I expect that this question will be settled within 20 years. I am not suggesting that quantum computing research is useless; on the contrary, it provides important insights into the isolation of nanoscale systems from environmental noise, which will be essential in the continued evolution of “classical” computer technology toward the atomic scale. But if I am correct, this will radically disrupt the orthodox understanding of quantum mechanics, leading to the adoption of a new quantum paradigm, with major long-term implications for the future of physics.