Read on Twitter
It can be difficult to explain how normal faults shape mountains (as opposed to just basins), so my niece, my sister @Artprof4 and I made one in a tank today (short 
).
).
We 
a fault between 2 plates made of flexible foam and glued magnets on both ends so the fault would only slide (and not open)
a fault between 2 plates made of flexible foam and glued magnets on both ends so the fault would only slide (and not open)
We then placed the whole thing in water, forced an offset on the fault, and took some pictures
Here are 3 snapshots showing increasing fault offsets. Left block moves 
, right block moves 
. Far from the fault they just move 
and 
. This forces the plates to bend upward, against gravity - and downward, against buoyancy.
, right block moves . Far from the fault they just move and . This forces the plates to bend upward, against gravity - and downward, against buoyancy.
Because of this flexure, elevation doesn’t drop everywhere in our foam rift, and a majestic 4-mm high fault-bounded range is formed.
For a detailed overview of this process, see for example Thompson & Parsons 2016, whose Fig. 3C very much inspired us https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2015JB012240
Of course in the real world, the mountain is made up of brittle upper crust floating atop weak (viscous) lower crust. Also, it gets incised by rivers / glaciers as the basin gets filled with sediments, like in the Lemhi Range in Idaho (a personal favorite, fig. by @_geoLuca )
(a personal favorite, fig. by @_geoLuca )
Next stop: core complexes in a tank?
