In J.P. Murre's "Reduction of the proof of the non-rationality of a non-singular cubic threefold to a result of mumford", in the proof of Theorem 3.11 he remarks that "the Jacobian of a curve is ’irreducible’ as a principally polarized abelian variety (i.e. does not split up in a product of principally polarized abelian varieties)."
I was not able to find any reference to an explanation or proof of this result. Can anyone point one out to me, or explain why this result is true? Much appreciated!