If I understand correctly, Feynman implies there exist other defining properties, better than names, that are preferable when thinking about animals (or objects

If I understand correctly, Feynman implies there exist other defining properties, better than names, that are preferable when thinking about animals (or objects in general)... unless one is communicating with other people, in which case a name will do. But it's names all the way down! These substitute properties are also only names. Using articulate attention (science), we divide the world into abstract categories, giving names to classes of "things" that are in fact unique, even if the classes are fuzzy around the edges and just leaky abstractions of reality. Is thinking in terms of the swirling, teeming, interconnected mass of "stuff" that's truly at the core of Nature a useful tool for getting ahead and making progress? It seems it's just too much to grasp, with everything unique and indivisible, no categories, no repeats. Even leaky abstractions (thrush! tiger! rock! water! human!) do better. To question the utility of names is to question the utility of the power of articulate discrimination, one of evolution's critical inventions. Not sure what lies that way except inarticulate madness.