In thinking about Harris’ take on science and morality it is worth comparing two types of models, as above. The first is a typical weather prediction model, complex, but with all the basic assumptions of science applied, the sciences of atmospherics, fluid dynamics, etc…(btw, chaos theory enters and makes full prediction problematic…)
And then look at the model in WHEE (the third edition actually had a better presentation, the fourth edition version is somewhat sidelined). We see that values are instrinsic to the model which collapses from a causal system to a discretized series of intervals where innovations, mostly of value phenomena in emergence (art literature), change the causal sequences into something more general (and along with the possibility of free agency).
So it is probbably true that science can deal with values because we see that in nature and ‘evolution of this new kind’. We see large scale systems that are probably not the result of designers (we can’t be sure) that can process values: literature, art, philosophy, religion, etc…How does it do it? We don’t know because we don’t fully observe the system in question, but we can see that it has moved far beyond conventional science.