The concept of sex is not dependent upon morphology (although within broad taxonomic units they are highly correlated). Rather it is dependent upon chromosomal configuration. While abnormal configurations of sex chromosomes can lead to viable offspring in humans (e.g. Klinefelter syndrome XXY) almost all normal humans are XX or XY, and each of these is highly correlated with specific anatomic configurations and adult phenotypic characteristics (mostly driven by the SRY gene on the Y chromosome btw). In other species, sexual dimorphism may vary in specifics, as noted in the article you cite. However, the dimorphism is still present in those species.

It's a fallacy to propose that sexual dimorphism must have universal characteristics. Biology is much more complicated than that (almost nothing is universal, not even metabolic pathways if you consider, e.g. archaea). If you're going to assert that sexual dimorphism doesn't exist because its manifestation is not universal, then by the same logic cellular respiration doesn't exist because the metabolic mechanisms are not universal.

In short, this is a red herring. The wide variation in manifestation of sexual dimorphism has nothing to do with the dependence of human social norms on human sexual dimorphism.