Given an algebraic curve in the $n$-dimensional algebraic torus, there exists a piecewise linear object in real $n$-dimensional space (called its tropicalization) which is supposed to capture the "leading-order" behavior of the curve. One of the goals of tropical geometry is to understand when this process can be reversed; that is, when does a tropical curve "lift" to an honest algebraic one? In this talk, I will say something about the corresponding problem in symplectic geometry—when a tropical curve "lifts" to a Lagrangian submanifold—and how these two problems are related through the mirror correspondence.