I am quoting the abstract of this paper here, without comment pro or con, for interested readers who may wish to read the paper for themselves. Feel free to debate in the combox.
Abstract: The concept of a physical object has figured prominently in the history of philosophy, and is probably more important now than it has ever been before. Yet the question What are physical objects?, i.e., What is the correct analysis of the concept of a physical object?, has received surprisingly little attention. The purpose of this paper is to address this question. I consider several attempts at answering the question, and give my reasons for preferring one of them over its rivals. The account of physical objects that I recommend – the Spatial Location Account – defines physical objects as objects with spatial locations. The intuitive idea behind the Spatial Location Account is this. Objects from all of the different ontological categories –physical objects; non-physical objects like souls, if there are any; propositions; universals; etc. – have this much in common: they all exist in time. But not all of them exist in space. The ones that exist in time and space, i.e., the ones that have spatial locations, are the ones that count as physical objects.