I recall your point, but what you provided doesn't demonstrate that 2 + 2 will not always equal 4. The only possible way it could is if you accept that + has no meaning and can be substituted for anything, but this is demonstrably false. It denotes a specific relationship. If you were to change + to t without maintaining that relationship, it becomes something entirely different and has no bearing on the original expression.

What about the praxeological action axiom? Can that be falsified somehow?