Water, perhaps humanity's most notable liquid, plays a key role in a myriad of physical mechanisms. For example, it is commonly known that the unique medium makes oily droplets aggregate with each other (i.e., oil does not mix with water). Nevertheless, the peculiarities of such a rudimentary process have been a contentious issue for decades: computational predictions propose a weak oscillatory force, yet experimental measurements suggest a strong monotonic force. In this work, I offer a resolution for this discrepancy by invoking an informatic property termed the relative entropy. In particular, I develop a comprehensive framework via the relative entropy for modeling purposes, and I in turn show that the apparent discrepancy is in fact a multiscale phenomena: oily droplets tend to aggregate via a weak oscillatory force if they are small and via a strong monotonic force if they are large, with the crossover occurring on the order of a nanometer. As the water-driven aggregation of oil is inherently linked with biological assembly, this dissertation has significant ramifications in various pharmaceutical and medical applications.