If nature is described by string theory, and if the compactification r
adius is large (as suggested by the unification of couplings), then th
e theory is in a regime best described by the low energy limit of M-th
eory. We discuss some phenomenological aspects of this view. The scale
at which conventional quantum field theory breaks down is of order th
e unification scale and consequently (approximate) discrete symmetries
are essential to prevent proton decay. There are one or more light ax
ions, one of which solves the strong CP problem. Modular cosmology is
still problematic but much more complex than in perturbative string va
cua. We also consider a range of more theoretical issues, focusing par
ticularly on the question of stabilizing the moduli. We give a simple,
weak coupling derivation of Witten's expression for the dependence of
the coupling constants on the eleven-dimensional radius. We discuss t
he criteria for the validity of the long wavelength analysis and find
that the ''real world'' seems to sit just where this analysis is break
ing down. On the other hand, residual constraints from N = 2 supersymm
etry make it difficult to see how the moduli can be stabilized while a
t the same time yielding a large hierarchy.