When does an internal table have columns that you can address with column names in a dictionary table type?

An internal table can be addressed with column names in a dictionary table type when its row type is defined as a structure. A structure in ABAP can have multiple components, and each of these components can be assigned a field name. This allows for the internal table to reference its columns using these field names, providing a more descriptive and intuitive way for developers to interact with the data.

When the row type is a structure, it encapsulates a logical grouping of fields, allowing the internal table to maintain a clear schema. As a result, you can use the field names directly in your coding, which enhances readability and maintainability.

In contrast, if the row type were a data element or a database table, or defined as a table type without any specific structure, the internal table would not provide the same level of descriptive naming for its columns. Therefore, it is the structure that grants the necessary capability for addressing the columns with meaningful names, thus solidifying the correctness of this choice.