Poles Apart

Orson Welles

There's a famous story - one of the most famous - from the history of physics. It involves an extremely serendipitous discovery which opened the door to a huge portion of the history of the universe previously hidden to our eyes, and won the discoverers plaudits, the Nobel Prize, money, fame, girls (OK, I might have made that bit up) and a place in history among the greats of physics. Incidental to the story, though critical to our purposes today, is something known as a 'dielectric'.

Let's tell the story, even though we've skimmed it in other places, because it's both instructive and entertaining - unless you're a bit of a Dicke.

The Holmdel Horn Antenna in Holmdel, New Jersey, was
constructed in 1959 for Bell Telephone Laboratories as part of a NASA project in passive satellite communications. Project Echo, as it was called, was much more barn-door in engineering terms than the foregoing might suggest. It involved floating plastic balloons powder-coated in aluminium and the reflecting radio signals off them. Ordinarily, over long distance where Earth's curvature is a factor, short-wave signals are bounced off the ionosphere, but this process is subject to scattering effects such as Compton scattering, which results in loss of fidelity at receiver. Using balloons reduces this effect and allows signals to be directed. 

The antenna itself is beautifully simple, if a little odd-looking. It isn't so incredibly different from more conventional dish antennae, despite how it looks. in fact, it's a partially-enclosed section of a conventional dish antenna, and this has some advantages given proper context. Let's start with how a conventional parabolic dish antenna works.

Here's an edge-on cross-section. As you can easily see, the shape - a parabola - means anything reflected off the dish from on axis - the direction the dish is pointing - is focussed into one spot, the focal point, which is where you'd mount sensors to collect everything. The effect of this is to amplify signals, by collecting as many photons as possible and sending them all into one spot. In fact, if you have a reflecting optical telescope, it works in fundamentally the same way, except the focal point, rather than a sensor, has a mirror angled toward the eyepiece.

It's also significantly the way sonic reflections work to amplify sound in cupola domes, such as in the famous Whispering Gallery in St Paul's Cathedral.

To see how the Holmdel works, we need to zoom in on the left-hand side and fill in the details.

Here you should be able to see the shape of the Holmdel Antenna and how it relates to a conventional parabolic radio antenna. It's essentially a wedge cut out of a parabola with walls and a narrowed opening. The walls and the narrow opening have a very important effect on the operation of the antenna. We can go back to the Whispering Gallery to see why.

Anybody who's been to a place like the Whispering Gallery will know you don't have to be at the focal point for signals to be amplified, hence the name of the gallery. It's a huge space but, because of the shape of the dome, you can whisper on one side of the gallery and it can be heard clearly on the other side. In fact, many of those 'off-axis' sounds will inevitably find their way to the focal point, and this presents a problem when looking for really faint signals, as they constitute noise.

The Holmdel, then, mitigates this off-axis signal by having built up sides and a narrowed opening. This makes it very good for faint signals and directionality.

When Penzias and Wilson arrived at Holmdel, their work was in R&D for Bell, specifically in the development of hyper-sensitive microwave receivers. In 1964, they hooked up their latest design to the Holmdel antenna and switched it on. And Houston, as the line goes...

There was a noise. It was really persistent. No matter which direction they pointed it, how warm or cold it was, how much moisture was in the air. The signal seemed to be everywhere which, in tech terms, is usually a good sign the noise is coming from inside the setup somewhere. So, they did all the things a tech does in such a situation, a process familiar to musicians and producers the world over; they started checking their leads. The ran diagnostics, tested every cable, checked all their grounding, did everything they could think of, up to and including sweeping out the hero of our tale today, a collection of what they described in their paper as a 'white dielectric material' or, to put it in the vernacular, pigeon shit.

So what's a dielectric, then? To find out, we need one little factlet about light. 

Here's a representation of a photon. The \(x\) axis is the direction of the photon's travel. Off this axis, you can see there are waves running orthogonal (at right angles) to both the \(x\) axis and to each other. Those waves represent the magnetic field B on the \(z\) axis and the electric field E on the \(y\) axis. As long as they remain orthogonal to each other, the electric and magnetic fields can rotate about the \(x\) axis, and the angle they're rotated to is what we refer to as the polarisation. By convention, when we refer to a specific angle of polarisation, the angle we refer to is that of the electric field.

So what a dielectric is, then, is a material which, when you push photons through it, it changes their polarity to a specific angle. As it turns out, one such material is bird shit.

In the event, after asking for some advice to deal with this noise problem, they were directed to another physicist, Robert Dicke. Dicke had been working for years to construct a detector he could use to look for the cosmic microwave background (CMBR), the photons  reflected off the last free electrons as they became bound and the universe became transparent to light, some 380,000 years or so after the Planck time. Dicke had, in fact, been struggling to find a way to exclude all the sources of radiation that would necessarily swamp it, in much the same way the design of the Holmdel Antenna had been to exclude off-axis sources. When he spoke to Penzias and heard his tale of woe, he was crestfallen, because he knew he'd been scooped. 

This isn't the only time polarisation has been a source of issues, of course. There's another famous example featured in some of my earliest writings here, in the pre-Planck cosmology series, where we talked about a discovery announced by the BICEP-2 team seeming to strongly favour one of our leading pre-Planck cosmologies over another, when they trumpeted to the world they'd found a particular kind of polarisation in the CMBR. In the plot, the lines indicate the polarisation of the light. As you can see, there's a peculiar sort of twisting, broadly following the differences in temperature. Note these differences in temperature amount to a couple of ten-thousandths of a degree between the hottest and coldest, and the average temperature is a balmy 3 Kelvin or, to put it in everyday terms, -270°C, or about 3 degrees above absolute zero.

As it turned out, they were a little premature in their announcement. In particular, they'd overlooked one potential source of this kind of polarisation in their observations, namely all the intervening detritus in the universe. All the clouds of gas and dust, and all the other shit which, in essence, acted as a dielectric. 

Once all the shit was corrected for, the signal disappeared below the error bars, the champagne went back in the fridge and, after a little shuffling of feet, everybody went back to their research.

There is a moral to this, of course. It's become increasingly apparent the world we live in is one in which there really aren't such things as polarising issues any more. In fact, every issue is polarising. As we noted in an earlier post, we seem to be living in a world in which voicing a preference between clippers and scissors for trimming toenails gets one consigned to a box with complete associated worldview and assigned opinion on everything from the solution for Zeno's Paradox to the aesthetic merits of the Cheeky Girls.

We live in a world in which these filters are imposed on us unwittingly by ourselves and our choices. Watch one Youtube video on a topic you've never shown interest in before and watch your feed fill with recommendations for more. Watch a few more and start seeing your feed filled with some of the same faces saying many of the same things. Watch still more and you start to think like they do; to trust them; to repeat their slogans and think in their tropes. 

You don't even realise you're doing it. In psychology, this is known as 'mirroring', and there's good reason to suppose this and related behaviours underlie things like Stockholm syndrome. Trust me, I know what I'm looking for, and I've still caught myself on occasion falling into some echo chamber or other.

The propagandists of the past taught us many lessons, but probably the most important is the slogan. Repeat a lie often enough and it becomes the truth, they say. In fact, this is the case for both truth and lies. The slogan, though, is something you can really hang things on, because you only have to repeat it a few times before it gets picked up and, before you know it, you have a meme.

All these things, slogans, memes, algorithms, propaganda, these are the dielectrics of modern discourse. They are the filters which polarise.

And, if there's one thing we've learned here today, it's where you see polarisation, there's probably some filtering going on. By corollary, the chances are not insignificant that the content of those filters is nothing but shit.