Math is not "truth". Math is a tool for examining and labeling. You cannot take math, in and of itself, and prove itself, on the nature of itself. Numbers are not real, they're a system of comparison.
Lots of numbers are real: http://en.wikipedia.org/wiki/Real_number
And seriously, from the information we know now, there are pretty decent odds that your assertion is false.
http://www.sciencedaily.com/releases/2013/07/130715070138.htm
If you study the philosophy of mathematics, you could even convince yourself that mathematical truths would exist independently of humans or even the entire universe.
) OP: true, you know what lunch you had, and you were close to the actual lunch. This does not have much to do with you being able to know what lunch you had if you were unable to surround yourself with any number of false statements, beliefs or views. The quote by Kafka seems to me to very prominently refer to the "truth" being something (as mentioned by myself later in the first page of the thread) in the realm of the mind, so to refer to basic external phenomena and knowing they are true is not against that premise (which is exactly why i myself mentioned pi, due to it being a mental calculation).
Wouldn't that sort of require that everything is infinite? (can a finite world have an infinite number of reflections about it, stemming from a finite number of beings - human or alien)? And that is only as long as the world is finite, or infinite but in some sort of relation to the sentient beings in it (ie canceling idealism). If it is not, then i suppose that the false statements are anyway smaller than those in theoretical existence (from a non-sentient being's point of view).