If we are to gauge, as proven, that we only need the upper extremities of the letter (from which all blessings flow, i.e. the Roman) then we must assume that there are subtle differences in their individual genetics which provide clues as to their identity (excluding the “O” and “Q”) when reading is performed. But Q is always followed by a U except in Qatar.

Just a thought... If you shortened the left serif on the capital J, or the right serif on the capital L is shortened, or if you randomized the lengths of the serifs on the capital I then anarchy would ensue. As minute as they are they provide major clues to the “divining (reading)” process. This is lost in SANS. God forbid we should play with the height of the cross bar of the H. Hi would become III. Well, sort of.

Can you imagine if we equalized the height of the X and Y. Pure bedlam!
The sky is falling!

Long live the serif!

Michael