Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-L] gravitational waves



On 04/19/2016 10:16 AM, William Katzman wrote:

saying: sorry I’m not going to try to give you a sense of what
we’ve found unless you can participate in a class of activities with
me, is a poor response to public interest.

That's a straw-man argument.

Please let's not pretend that the only choices are (a) horrendous
detail or (b) nothing. That's a false dichotomy.

Furthermore, please let's not pretend that the only choices are
(a) horrendous detail, (b) nothing, or (c) trampoline model.
That's a false trichotomy.

when I explain to a 6 year old versus a physics graduate I use
different means of explanation.

That's entirely reasonable as far as it goes, but it's not
very informative. The devil is in the details of how to
explain things to each customer.

Suppose a tabula_rasa 6 year old comes to the piano teacher
and says "Teach me how to play the Hammerklavier Sonata".
https://www.youtube.com/watch?v=zucBfXpCA6s
Rather than saying simply "No" the wise teacher will say "How
about I teach you how to play Mary Had a Little Lamb instead.
Then we can move on to more complicated things, one step at
a time."

All models are imperfect. Approximations are absolutely necessary.
However this does not give us a license to use the first wacko
model that comes to mind. There are lots of different models to
choose from. Choosing one that is appropriate in this-or-that
given situation requires judgment and skill.

IMHO the trampoline model is not suitable for the aforementioned
6 year old. It's too complicated. It's also not suitable for the
aforementioned grad student. It makes too few correct predictions
and waaay too many incorrect predictions. The crucial fact is:
better models are readily available.

If the 6 year old asks about gravitational waves, you could
quite reasonably say it's a wave, and it's gravitational. If
the kid wants more detail, you can give progressively fancier
answers about what a wave is. Going directly to a bogus model
of curved spacetime doesn't help anybody.

Similarly, if somebody wants to know about curvature, you can
roll out progressively more detailed /correct/ ideas. Going
directly to a bogus model of curved spacetime doesn't help
anybody.

*) One can learn about waves per se:
++ waves on a pond
++ waves in a ripple tank
++ waves on a slinky
++ waves on a Shive machine
++ et cetera
*) One can learn about accelerated reference frames:
++ starting with centrifugal field
++ including weightlessness, e.g. in the frame comoving with
the space station, or in an aircraft.
*) One can learn about geodesics
++ strings on a globe
++ masking tape
++ et cetera
reference: https://www.av8n.com/physics/geodesics.htm

One does *not* need to learn all of that at once. Going directly
from 1st grade to general relativity would be bad pedagogy.
Going directly from 1st grade to the trampoline model is wrong
physics on top of bad pedagogy.

As David Bowman has been pointing out for many years, if you
want to understand curved spacetime, you ought to start by
asking /in what direction/ is it curved. That's the price of
admission. A model that can't answer that question correctly
is almost certainly worse than nothing.

===========

One should not expect one analogy to carry the whole burden. So
we use multiple analogies and combine the results. In the educational
psychology literature, this is called the building block approach.

It is important to know in what ways each analogy is faithful to
reality, and in what ways not. Consider the blank space in the
following chart:

* ** *** ****
x xx xxx xxxx
+ ++ ++++
o oo ooo oooo
v vv vvv vvvv

Here the rows are faithful as to shape, while the columns are
faithful as to number. In this way we can communicate the idea
of "three plus signs" even though it is not expressed directly.

Trying to construct a mechanical model that represents curvature
*and* wave propagation is almost certainly a step in the wrong
direction. The cost/benefit ratio is terrible. The alternative
is to use two models: one that represents waves, and one that
represents curved spacetime. Students can combine them, just
like they combine the notions of shape and number in the chart
above.