Can there be a disconnect between the ontology and the RDF data in an RDF database?

Is it possible that the relationships of the data in a RDF graphstore are different from that defined in an ontology?

For example, let's say the ontology says for a class Person, it has the following attributes,

Person hasName "someName"
Person hasBirthDate "someDate" etc.

Now let's say you have defined a predicate called hasGivenName and hasLastName somewhere else in the ontology but not associated with class Person.

Now can the data be stored as follows?

PersonX hasGivenName "someName"
PersonX hasLastName "someLastName"

If so then what is the point of the ontology?

How are large scale triples created, mapped and stored if they do not conform to a schema or ontology?

Let's say dbpedia, how does it create the graph ? Is there an ontology or schema used when the triples are created and updated in the database or is it just freeform mapping ?