Lactose intolerance after weaning is normal in all mammalian species, perhaps to reduce energy consumption producing unnecessary lactase enzymes (as CliffordK suggests). It is believed that this was the normal condition in all
early human populations, prior to the development of agriculture.
Lactose persistence is genetically dominant, and can be enabled by changing as little as a single DNA letter (a SNP: Single Nucleotide Polymorphism) in a promoter region for the Lactase enzyme; several independent mutations are known: http://en.wikipedia.org/wiki/Lactase_persistence#Genetics
From the interview:
within about a thousand years a gene evolves which allows people to tolerate milk so we've become lactose persistentRe Terminology
: You often hear this loosely expressed like "a gene evolves which allows people to digest
milk" (or "for bacteria to digest citrate" in the case of the Lenski experiment). Both of these would involve the evolution of a large and complex enzyme (1927 amino acid units/5781 nucleotides, in the case of the lactase protein-coding region: http://en.wikipedia.org/wiki/Lactase#Structure_and_biosynthesis
However, what we have here is a mutation which disables a regulatory function for an existing enzyme. This disabling can occur due to a point mutation, which is far more likely to occur in a short amount of time, compared to the evolution of a new enzyme.
I guess this confusion is propagated in the general media because when the reporter hears "a new gene", they cannot conceive (and hence cannot convey) the huge difference between "a new enzyme" and "a tweaked promoter for an existing enzyme", so the audience just assumes the former.
It's important for science communicators to be clear about such differences.