Totality and Order

I said that "order" was no more than a memorable array and that "disorder" was an indiscernible array. A problem with that idea, of course, is that it seems to deny the role of relationship in presenting order. For example, a totality is related to and constituted of its parts. But if the parts/totality schema is seen as merely a memorable array of parts, how do I work toward a totality?

Two considerations in response to that objection:
1) Relationship, in this case in the form of an addition of parts, need not necessitate any resource to a conception of "order". We simply add the parts.

2) Is a totality constituted of its parts? It does not seem possible to present that idea without circularity, viz: "how many parts make a totality? 'All' of them?'. The problem can be put another way: when do we stop adding parts - when we get a totality? But then how do we know when we have a totality? Thus, there seems to be a problem with the idea of relationship in this case.
.