Crazy Eddie, I think we now agree about the general biological principle of race. The mainstream science of intelligence is another science that is under attack from the public and from scientists outside the relevant field, again for the sake of fighting racism. Intelligence can and does mean many things, but, in psychology, as among the public, intelligence tends to mean the ability to learn and to solve problems. The technical definition of "intelligence" is "whatever intelligence tests measure," so an understanding of the academic definition of intelligence requires looking at the patterns common to intelligence tests. You listed "emotional empathy" and "social processing" as talents, and they are important, but they are not typically regarded as intelligence, and they are not on intelligence tests. I am not claiming that intelligence is all that matters for success in life, and we must be clear about our language, not just lump in anything and everything with "intelligence." The other talents--"memorization," "abstract thinking," and "visual-spatial reasoning"--are all closely associated with intelligence; it is not just memorization that is closely associated. The principle behind IQ is that all tests typically regarded as intelligence tests correlate positively with each other, meaning: if you score highly on one intelligence test (i.e. the SAT) then you are likely to score highly on another intelligence test (i.e. the GRE). Since they all correlate positively with each other, there would be a factor common to all them that can be statistically isolated. This factor is assigned to be Spearman's "g" for general intelligence. IQ tests are designed to maximize the relationship with g. So, an IQ score would best estimate the probable scores on all intelligence tests. Whatever definition of intelligence you may prefer, Spearman's g has undeniable value for its external correlations. The IQs of identical twins reared apart correlate with each other by 74% percent, whereas the IQs of unrelated children reared together correlate by only 26%. Variations in IQ also correlate with variations in brain size by 40%, variations in college grade-point average by 50%, variations in income by 50%, and they have a strong association with careers that we typically associate with intelligence (scientists, engineers, lawyers, medical doctors and business executives). You can define intelligence any damn way you want, but, when I say "intelligence," you should know that I mean it the same way intelligence researchers mean it: Spearman's g or IQ.
Jason Richwine is another interesting example. Harvard students actually wanted to academically censor themselves to prevent such theses as Jason Richwine's ("Let us be clear that we believe in academic freedom as it is crucial to the functioning of a university. However... such plain racism cannot and must not be tolerated."). I did not bring up Jason Richwine, but feel free to explain further your thoughts about him.