A View from the Back of the Envelope top

Why does the sun shine?
Deep understanding
"If you can't explain it to a nine-year-old..."

Because it is hot, and hot things shine.

Hot stuff glows. Not everything which glows is hot, fireflies for instance, but everything hot glows. And glows different colors depending on how hot. If the Sun were a little bit hotter, it would be blueish. A little bit cooler, redish. Some other stars are. The Sun itself will be one day. Sunspots are dark because they are cooler areas. When you turn on a light bulb, you are making the wire inside as hot as the Sun! Thats why it's the same color.


Why did I write this page? Children ask simple questions which are profound and difficult to answer. And are rarely answered well. Someone suggested the difficulty was kids' limited ability to handle complexity in answers. My own impression is complexity in answers usually reflects only the explainer's limited time and limited understanding of the phenomena. ("If you can't explain it to a nine-year-old..."). It was suggested "Why does the sun shine?" could not be answered well. After all, how could one explain fusion to a second grader? This page (well, actually the email below) was my response.

Email excerpt
When a young student asks "What makes the Sun shine?", perhaps "it is very hot, and hot things glow"? I'm not sure why fusion is the common answer. Asked about a light bulb, one seems more likely to focus on the hot tungsten filament, than on the power grid or plant. Fusion is quite localized, and the Sun's vast interior stores something vaguely like 50 million years of core energy output, something like 1% of its cumulative generation. Lets see... if the core doing fusion were a 100 W aquarium heater, maintain say a 100 degrees delta, thats 1e15 s, 1e17 J, 1e15 J/C, at 4e3 J / C kg(H2O), say 1e11 kg, at 1e3 kg/m^3, 1e8 m^3, so a 100 meter big fish tank of a sun. If someone asked the source of heat, the aquarium heater might be good answer. But if they asked why there was condensation on the ceiling? If one toggled the Sun's fusion on and off with alternate million years, I wonder if any change would be visible for quite some time.

Besides, there seem more simple stories about heat and color than about fusion. Fusion has "hydrogen nuclei smashing together so hard they stick", but that's rather intangible, and the fudge factor is a bit high. My very fuzzy recollection is the real process is quite hairy. One might segue to "the universe is all hydrogen", then on to heavy element synthesis and "we are made of star stuff", but that path seems too abstract for younger kids. "Hot stuff glows" lets you do Planck. "Not everything which glows is hot, fireflies for instance, but everything hot glows." "And glows different colors depending on how hot." "If the Sun were a little bit hotter, it would be blueish. A little bit cooler, redish." "Some other stars are. The Sun itself will be one day." "Sunspots are dark because they are cooler areas." "When you turn on a light bulb, you are making the wire inside as hot as the Sun! Thats why it's the same color." (This last invites problems with fluorescents). If you have a dimmer switch and clear bulb, you can see the coil of wire inside heating up, first dark red, then orange, then white. Also with pins on gas ranges. The red-orange transition can also be seen in electric ranges, toasters, and video of lava. Candles do blue, white and orange, and red in the ember, but require more explanation. "You are glowing too. But you are much, much cooler than the sun. So you don't glow white. Or orange, or red. You glow cooler than red, so cool your eyes can't see it, though rattlesnakes have an extra "eye" which can." Color, light generation and perception. It seems a more fruitful explanation.

I don't know how to describe the Sun's visual surface (photosphere), as one would the tungsten filament. Perhaps something like brightly glowing air (density is actually 10^2 down from SP), a couple of hundred km deep, with each cubic meter glowing like a few 100 W bulbs (1e8 W/m^2 / 1e5 m). So a room of it would be a glaringly bright. Eh. (And the last part is slightly broken.)

As you say, younger students require answers of limited complexity. I wonder though if that "complexity budget" is being spent well. Does the complexity of current answers stem mostly from inherent complexity of concept? Or is much a compositional artifact, stemming from limitations on time and teachers' understanding?

We are sometimes lucky enough to encounter exceptional teaching. A teacher who's insight into something allows them to lay bare its core, making simple and understandable what was formerly complex and impenetrable. The someone lecturing on something-or-other which simply must not be missed. But such experiences are the exception. Usually we just slog along. Why?

Perhaps most things are not subject to such insightful simplification. Or a suitable person has not gotten to them yet. Or it has been done, but is not generally available, is not at hand.

If we slog along because insightful simplification normally cannot be done, then we are stuck. But we can do something about the other two factors, so if they are significant, we can make progress.

How often is insightful simplification possible? I don't know. I have not seen it discussed, though it seems a researchable question. I've certainly heard many individual claims of inherent complexity. But how often are such evaluations accurate? My own rule of thumb is to distrust such judgments if the person is operating outside of their own immediate subfield. And even then I encounter problems. It seems the question is currently hard.

So I am reluctant to grant that the bottleneck is fully characterized by students' skill in handling explanations. There seem difficulties upstream as well - with our creating and delivering the explanations.

A View from the Back of the Envelope
Comments encouraged. - Mitchell N Charity <mcharity@lcs.mit.edu>

 I'm not sure this page is worthwhile.
 Sun and filament are not quite the same temp - 6kK vs 3kK.  A factor of 2.

  Spectrun diagram.
  Decide if this page is worthwhile.
    If so, link it into the rest of the site.
  Collect web answers.
  Do a full up explanation (1 day? 2?).

  1998.Jun.11  Created this page.
  1998.May.15  The original email.