Cosmology That ClicksBonus 7 / Standing on the "operator's side" (trilogy, part 3, the close)

Can you turn the energy knob as far as you like? — No, it has two ends. And those ends are where it gets interesting

The Fineness Knob
Has Two Ends The "how finely you look" knob that grows \(\alpha\). Turn it down and it hits a floor at 1/137; turn it up and it slams into an edge where the whole framework itself ends.
The knob's two ends coincided exactly with where today's physics runs out.

What you'll need: Episode 5's "bandwidth," Episode 6, Bonuses 4 & 6 Floor at 1/137, ceiling at the Planck edge

In Bonus 6 we saw that \(\alpha\) has a way of moving — it "runs with energy (with how finely you look)." So then — can you turn that knob as far as you like? The answer is "no, it has two ends." And what's more, those two ends aren't just limits: they line up precisely with the place where today's framework of physics itself runs out. What we said in Episode 5, that "the range you can see has an edge (bandwidth)," and what we said in Bonus 4, that "equivalence breaks down at the Planck scale" — all of it reappears here as the two ends of this single knob. This is the close of the trilogy.

01The lower end — hitting the floor at 1/137

First, turn the knob all the way down (look coarsely). As you lower the energy, \(\alpha\) stops at \(1/137\). It won't drop below that. The reason is exactly the screening picture from Episode 6, Part 1 — once you get completely outside the \(e^+e^-\) cloak that wraps the electron, there's no more cloak left to add that could hide the charge. That fully-hidden, weakest-looking value of the charge is \(1/137\).

What the floor really is

\(1/137\) is "the fully-screened, coarsest value" = the knob's lower limit (the floor).
So when Episode 2 said "everyone who measures it gets \(1/137\)," what that meant, precisely, was this floor value. It doesn't move with your choice of units, of course — but also you can't make it any coarser than this, so everyone sees the same value.

02The upper end — it grows, but there's a triple wall

Now turn the knob up (look more finely). It keeps growing past \(1/128\). But it's not "the same rule, all the way up, forever." There's a three-tiered wall at the top.

Wall 1
The rule for running changes partway upAs you raise the energy, more particles join in the screening (muons, quarks, the W...). Each time you cross the "mass gate" of a new charged particle, the cloak thickens and the slope (the way it runs) changes. It's not one smooth curve — it's a staircase.
Wall 2
~100 GeV
The electroweak scale — the variable α itself stops being rightAbove around here, electromagnetism is no longer a "fundamental force." It's just a blend of the weak force SU(2) and hypercharge U(1); what you ought to be running isn't α but those two couplings. Continuing to talk in terms of α at all becomes a stretch. → On to grand unification in Episode 6, Part 2.
Wall 3
~10¹⁹ GeV
The Planck scale — the very concept of "fineness" breaksQuantum gravity kicks in, and the premise that "spacetime can be resolved smoothly, arbitrarily finely" collapses. This is exactly the region we called "the edge of the bandwidth" in Episode 5, the region where we wrote in Bonus 4 that equivalence breaks down. The very "tick marks" you turn the knob against cease to exist.
Bonus — with QED alone, it's the "Landau pole," infinity If you let the calculation run away using pure electromagnetism (QED) alone, then at an astronomically high energy a point appears where \(\alpha\to\infty\) = the Landau pole. This is a sign that "QED is not a finished theory on its own — it's only an approximation (an effective theory)." But that height is far above the Planck scale, and in reality Walls 2 and 3 have long since taken effect well before you get there, so this is an academic footnote.
Figure: 1/α against how finely you look (energy, horizontal axis logarithmic). The left end is the floor at 1/137; run to the right for 1/128, then on to the electroweak, grand-unification, and Planck edges. Use the slider to check which region you're in.
◇ ◇ ◇

03The knob's two ends were the edges of the framework

Here's the close. The knob's lower end (\(1/137\)) is the limit where "no matter how much coarser you look, nothing more is hidden." The upper end is the limit where "electromagnetism stops being electromagnetism / spacetime stops being smooth." At both ends, it stops right where the quantity \(\alpha\) itself can no longer carry any meaning. In other words —

The heart of this installment

The two ends of the "fineness knob" coincide with the places where today's framework of physics runs out.
The value (α) runs, but the range over which you can run it has a physical edge. The very width the knob can turn through is the "bandwidth" of the universe visible to us.

In Episode 6, Part 1 we reframed the question as "what's deep isn't the value, it's the rule for how it runs." This time we can see one step beyond that — that rule for running itself changes at Wall 1, ends together with the whole framework of \(\alpha\) at Wall 2, and loses its very foundation of "fineness" at Wall 3. Even the rule that looked invariant eventually has an edge. This was precisely the \(\alpha\) version of Episode 5's "the range you can see (the bandwidth) has an edge." The trilogy (5: the freedom of units, 6: the two axes, 7: the knob's two ends) lays out on a single page "the three 'ways of moving' — time, units, and energy," and closes by seeing that every one of them has an edge.

04Reading guide — for those who want to know this story more deeply

"Aren't there hardly any books or people who explain this area?" — that feeling is half right. There are great books on the individual pieces. But something that binds running, changing constants, the freedom of units, and equivalence with expansion into a single story barely exists. It falls in the cracks between fields, on no one's turf. Piece by piece, here are the closest things I can point to.

The very "heart" of this trilogy (free, arXiv)
Paper M. Duff, L. Okun, G. Veneziano, "Trialogue on the number of fundamental constants" (2002)
Three authors argue over whether the number of fundamental dimensionful constants is 3, 2, or 0. Duff's "0 (all dimensionful ones are just conventions of units; only the dimensionless are physics)" is exactly the backbone of Bonus 5. arXiv: physics/0110060.
Popular books (as reading)
Popular book John D. Barrow, The Constants of Nature (2002)
α, 137, whether constants are really constant, all the way to quasar observations of a varying α — a whole book on it. The frontrunner among popular books on this subject.
Popular book João Magueijo, Faster Than the Speed of Light (2003)
A popular book by the very proposer of variable speed of light (VSL) theory. The living voice of the "insider" from Bonus 3.
Popular book R. Feynman, QED: The Strange Theory of Light and Matter (1985)
The classic that called 137 "a magic number written by the hand of God." Vacuum polarization (screening) = the source material for Episode 6, Part 1, at its most accessible.
If you want to check it seriously (technical reviews)
Review J.-P. Uzan, "The fundamental constants and their variation" (Rev. Mod. Phys. 2003) / "Varying constants, gravitation and cosmology" (Living Reviews 2011)
The definitive treatment of the observational constraints and theory of varying constants. The atomic-clock, Oklo, and quasar numbers can be sourced here.
Explainer Accessible explanations of running
Popular books stop just short of 1/137, and textbooks jump straight to the renormalization group — a "thin layer" in between. For the middle ground, Matt Strassler's blog, David Tong's free QFT lecture notes, and Wilczek's essays are the bridges.
Why "so few people explain it" These are all siloed. The philosophy of constants (Duff) doesn't discuss running; running (textbooks) isn't tied to varying constants or to expansion; varying constants (Barrow/Uzan) doesn't put "α running with energy" and "α changing with time" in the same figure for contrast. What this bonus series is doing is laying them out on one page as "three ways of moving — units, time, and energy." Not because it's hard, but because it falls in the cracks — that's why few people wrote it up together. And that's where this series lives.
Practice problems
  1. Why can't α get smaller (1/α larger) than \(1/137\) even as you turn the knob down?
    Show answer
    Because once you get completely outside the screening cloak that wraps the electron, there's no more cloak left to hide the charge. 1/137 is "the fully-hidden, coarsest value" = the floor.
  2. In one line: why does "continuing to talk in terms of α" become a stretch above about 100 GeV?
    Show answer
    Because above the electroweak scale, electromagnetism stops being a fundamental force and becomes a blend of SU(2) (the weak force) and U(1) (hypercharge). What you ought to run is those two couplings, and α stops being a good variable.
  3. The knob's upper end (the Planck scale) corresponds to which phrase from Episode 5?
    Show answer
    "The edge of the bandwidth." The visible range (the bandwidth) has an edge, and there the premise that "you can resolve it smoothly, arbitrarily finely" = the very tick marks of the knob, break down.
An honest line

The energy scale and the positions of the walls in the figure are schematic, meant to give you a feel for the orders of magnitude. The actual running is logarithmic and depends on the kinds of charged particles, and the electroweak "blending" (the Weinberg angle) and the grand-unification convergence have theoretical details. The discreteness of spacetime and the breakdown of "fineness" at the Planck scale are unsolved problems in physics, not confirmed by experiment.

The book titles and years in the reading guide are representative; the wording may differ across translated editions or printings. The arXiv numbers (physics/0110060, etc.) let you read the full text for free.

Trilogy summaryAll three "ways of moving" had edges

The "fineness knob" has two ends. Down is \(1/137\) (the floor of complete screening); up is Wall 1, where the way it runs changes, Wall 2, the electroweak wall where α stops being a variable, and Wall 3, the Planck wall where the concept of fineness breaks. Both ends coincided with the place where today's framework of physics runs out. The value runs, but the range over which you can run it has a physical edge — this is the α version of Episode 5's "the bandwidth has an edge."

Looking back over the trilogy: in 5, "if you move α, you can only fix three; the one that moves is a choice of bookkeeping (gauge)"; in 6, "for α to move there's a time axis and an energy axis, and the latter has no gauge freedom"; in 7, "even that energy axis has two ends, and they coincide with the edges of the framework." Units, time, energy — every one of the three ways of moving leaves only the dimensionless α as physics, and every one of them has an edge. Bonus 2's "the absolute value is bookkeeping, only the ratio is physics" showed the same face across all three axes.

This document is Bonus 7 of the "Cosmology That Clicks" series, a piece of reading for physics-loving high-schoolers and undergraduates. It is established physics that, in the low-energy limit, \(\alpha^{-1}\approx137.036\) (the Thomson limit, zero momentum transfer) saturates, and that at high energy it runs up to \(\alpha^{-1}\approx128\) (the \(Z\) scale). That the running (the β function) changes at the mass thresholds of charged particles, that above the electroweak scale (~100 GeV) electromagnetism is embedded in U(1)×SU(2) and α ceases to be a fundamental coupling, and that at the Planck scale (~10¹⁹ GeV) the picture of continuous spacetime can break down due to quantum gravity, are all standard understandings (the details of the Planck scale are unsolved). The Landau pole of QED alone lies at energies far beyond the Planck scale and is meaningless in practice. References: Duff–Okun–Veneziano (2002, arXiv physics/0110060), J. Barrow, The Constants of Nature (2002), J. Magueijo, Faster Than the Speed of Light (2003), R. Feynman, QED (1985), J.-P. Uzan (Rev. Mod. Phys. 2003 / Living Reviews 2011). — To print, use your browser's "Print" and "Save as PDF" (in the print version the slider and answers are frozen and hidden).

Print / PDF: ⌘+P (Ctrl+P on Windows). On screen, use the slider to move the knob from the floor to the walls. "Show answer" opens each solution.