A table is not simply the set of a number of particular legs, or a particular stump and a particular board, altho it is made up of them. Three or more legs at one place and a board somewhere else do not yet constitute a real table; maybe they can or could constitute one. To actually be a table, more is needed and this more of every table is, or is found in, its attributive predicament. 'Mereologists' may talk about the different parts of the table and even wonder whether they are essential or not, while at the same time defining this table as a mere extensionality. (Mereological essentialists claim that all or certain parts are always essential to their whole.) When running into difficulties they may distinguish 'popular parts' for the ordinary or loose sense of part from 'philosophical parts' when speaking of "parts" in a strict sense. In a 'creative' bout of supernaturalist elation a realm of 'entia successiva' (things which are such that at any moment of their existence something other than themselves serves as their stand in and does duty for them), 'entia per alio' (things which derive all their attributes from other things which do duty for them) and 'entia per se' ('real things') may be hypostatized. But ordinary talk about wholes and parts cannot be that inadequate, even tho one of the issues, that of things remaining the same (or identical) thru time, is a delicate one.

Theories which cannot tackle the ordinary language of wholes, parts and attributes without taking refuge in a thicket of ethereal entities just don't ring true. This is not to say that they would be wrong or uninteresting in every respect. For example, it does turn out very fruitful to draw a distinction between a strict and a loose sense of the words part and having, also in our nonmereological framework. The difference is that it is a very straightforward distinction here which can be easily defined in set-theoretical terms. For a clear understanding we must in the first instance reject in our case the axiom that if X is a part of Y and Y is a part of Z, then X is also a part of Z. If parts are genuine wholes themselves (and recognized as 'individuals'), then a part X of Y is, strictly speaking, no part of Z, even if Y is part of Z. (The axiom itself is the expression of a mereological conception of wholes and parts.) A part of a part of Z is something else than a part of Z, just as a parent of a parent of P is not (necessarily) a parent of P, and just as an utterance about an utterance about a fact F is not (necessarily) an utterance about F itself. At the same time it cannot be denied that one often calls a part of a part of A also "a part of A", or an utterance about an utterance about B also "an utterance about B" (and a friend of a friend of C also "a friend of C"?) in a loose sense. Especially when the division into parts, and also subjects, is vague or quite arbitrary the loose usage may suffice for the purpose of the conversation. But whether the term part is employed in a strict or in a loose sense, Y is never a part of X, if X is a part of Y (as another axiom reads). This is because in both cases part only refers to proper parts, not to wholes which would be 'their own part'.

The appeal of the distinction between the strict and the loose meanings of part and having is that it provides us with an alternative to extensional mereology (having refused to accept total mereology in any case). The distinction allows us to do away with the very complicated 'fusions' to which not only the parts of the whole belong but also the parts of parts and collections of parts, or of parts of parts, and so on. This is not a very practical, accurate picture because, for example, somebody does not have a hand and five fingers in addition to that hand; that person('s body) has a hand and that hand has in turn five fingers (granted that the number is correct). It follows from this that also that person has those five fingers, but then in a looser sense of having.

The part-whole configuration and the transition from a strict to an ever looser sense of having can be well demonstrated particularly with regard to persons. It is therefore high time now to see how these peculiar entities which no adequate ontology may ignore or neglect can be portrayed by means of our own conceptual apparatus.