Tables are single tables of data - and in an information map, you can specify how multiple TABLES can be joined together and you can define the relationships between the various tables. In this way, the consumer of the information map does not have to worry about the complexity of the underlying data, the physical storage and/or the database or file schema. For example, you could have one table with vendor names and vendor IDs and a second table with vendor IDs and vendor products and a third table with vendor IDs and addresses. In an information map, you can control the join that takes place so that a report person could get a report that showed vendor name, product and vendor address all on one report -- without having to worry about knowing how to join the 3 separate tables. There is no hierarchy definition or possibility of a hierarchy definition in this information map that joins these 3 tables together.
On the other hand, the term "hierarchy" refers to a concept that relates to multi-dimensional data sources -- also called OLAP cubes. An OLAP cube is a unique type of data source. Usually, the hierarchies for an OLAP cube are defined by the cube developer using either OLAP Cube Studio or other tools. This paper gives a good overview of OLAP cubes in the context of the BI Platform. http://www2.sas.com/proceedings/forum2008/182-2008.pdf