Are compositional (mereological) relations always distinct from definitional relations, or can compositional relations be considered definitional relation in some cases? Does this depend on whether we are considering nominal definitions or real definitions?

A compositional relation R1 could be "water is composed of H2 and O". A definitional relation R2 could be "water is defined as H2 and O". Could one say that R1 is just a kind of definitional relation, perhaps a real definition? What makes me hesitant to assert this, is that i think of definitions as somewhat arbitrary ideas, while compositional relations seem to be true or false with respect to reality. Also definitions seem to be symmetric relations while compositional relations are necessarily anti symmetric. Maybe this is just a semantic issue with no material implications...

My understanding is that there is some kind of taxonomy of relations, involving causal relations, compositional relations, real definitions, nominal definitions, etc. Essentially my question is about where compositional relations and definitional relations fit in this taxonomy, if composition is a subtype of definition. Any references to not to difficult texts would be appreciated (i am not a philosopher).

I have glanced at the following sources but the different kinds of relations are not compared here. Definitions

Many thanks in advance!


  • Some context would help here, what are "definitional relations", "compositional relations", etc., do you follow some source in particular? When heart is defined as that part of the body which pumps blood a "compositional" relation is used "definitionally", but I am not sure that this is what you are after.
    – Conifold
    Commented Feb 26, 2019 at 23:48
  • @Conifold Tried to clarify my question with some context.
    – user34796
    Commented Feb 27, 2019 at 4:54
  • There is a school of thought that considers composition of to be essential to certain things, so what is not composed of molecules of H2O is not water, see essentialism. Composition then is part of the "real essence", but, obviously, it need not be so. For example, a wart is, presumably, not essential to a human having it. On the other hand, there are plenty of definitions that do not involve composition, including "real" ones, e.g. Euclid's definition of a circle. So they overlap, but neither is a subtype of the other.
    – Conifold
    Commented Feb 27, 2019 at 8:42


You must log in to answer this question.