Inferring the behavior and function of ancient organisms is hard. Some paleontologists would say that it cannot be done because such hypotheses can never be testable, whereas others would say that this is surely a prime task for paleontology—to seek to bring ancient organisms back to life.
These issues have long troubled paleontologists. The founder of comparative anatomy, Georges Cuvier (1769–1832), insisted on the common pattern of the skeleton of living and fossil vertebrates and that anatomy could be reconstructed with confidence from incomplete fossil remains. Further, he argued that the skeleton of a living or extinct animal held unequivocal clues about function and behavior. Cuvier saw his mission to establish rules for comparative anatomy that would allow paleontologists to make certain statement with clarity and confidence [1], a key principle today, what one might call “evidence-based reconstruction” (for example, sharp teeth indicate a diet of meat rather than plants, or mammalian characters in the teeth indicate that the unknown animal was endothermic and nourished its young from mammary glands) as opposed to speculation (“this dinosaur was purple because I guess it was”).
Fossils can provide great evidence about functions. such as the hard skeleton of a fossil arthpod reveals the number and shape of limbs. the nature of each joint in each of said limb. and perhaps also mouth parts and other structures related to locomotion and feeding.
There were fewer people at that time and even though they had their religion and myths there were still a lot of unanswered question that people like philosophers would try to find answered for.
