This experimental arrangement is familiar to school students who use light, rather than electrons, projected onto the two slits to demonstrate interference fringes on the detecting screen. However, Einstein demonstrated in his 1905 paper on the photoelectric effect that light comes in chunks called quanta (renamed “photons” in 1926), and it wasn’t long before experimenters were able to reduce the intensity of the light falling on the flat plate to such an extent that only one photon would reach the screen at any time. With modern equipment you can actually see the effect of individual photons hitting the detecting screen, one at a time.

The illustration shows the detecting screen after it has been peppered with a few decades of photons (it could equally well be electrons, as in Einstein’s original thought experiment, but we shall talk of photons for the moment). The position of any individual photon landing on the screen seems to be quite random, although, as the experiment proceeds and the number of photons arriving at the screen increases, a vertical-banded pattern emerges, and, if the experiment is carried on long enough, the familiar fringes come into view. If the light source is, say, a laser, then the spacing of the fringes is determined by the colour of the laser light, i.e., the frequency of the emitted photons. (Max Planck and Einstein established that the energy of a quantum of light – the smallest packet of light energy, later to be called a photon – is its frequency times a constant: Planck’s constant, h.)

This experiment can’t be explained classically. The appearance of the banded pattern strongly suggests interference, but, if light comes in isolated photons, then, classically, the photon can’t go through both slits, and so there can be no interference. On the other hand, if the photon somehow really does split into two and so goes through both slits (so that it can interfere with itself), then the energy of the original photon would have to be divided between the two daughter photons, hence lowering both their frequencies. However, those reduced frequencies would result in the bands being spaced out more, but that doesn’t happen. Yet if, despite the overwhelming evidence for light being a quantized, photon phenomenon, we nevertheless treat it as a wave, then how do we explain the observed detection of individual photons on the detecting screen – how can the light wave “collapse” instantaneously onto the single point of detection without it collapsing at other locations as well – as Einstein pointed out in 1927 for his hemispherical screen? Also, as Einstein had already noted as far back as 1909, how could all the energy of the spread-out light beam suddenly be concentrated into one point of detection? (Einstein, A. Über die Entwicklung unserer Anschauungen über das Wesen und die Konstitution der Strahlung [On the development of our views concerning the nature and constitution of radiation]. Physikalische Zeitschrift 10, 817– 825 (1909).

These questions certainly illustrate the mystery of quantum mechanics that Feynman was referring to. However, rather than try to interpret the double-slit experiment here, I’m simply going to use it in the following webpages as a tool to help me make the case that the universe is fundamentally indeterministic. In passing, notice that this same experiment can be, and has been, performed using electrons as well as photons. Electrons, and, indeed, all particles, have a quantum wave associated with them, as we shall see next, although their wavelengths are generally much smaller than those of visible light. From now on, when we are talking about the double-slit experiment, we shall be thinking of particles rather than photons, because there are properties of particles that will come in useful when we develop our arguments later on.

Click the button for the next step in the journey!