Null References and Missing Objects

When beginners first meet object variables, they often describe them in everyday language: a variable is a container, and an object is put inside it. The phrase sounds reasonable, and many people use it without thinking. In Java, however, that picture is not accurate, and the difference matters once you start writing programs that work with several objects at once.

This article starts from the wrong belief, tests it with a prediction, and then builds the correct model from the ground up. By the end you should be able to say exactly what a variable holds and exactly where the object is.

The belief to examine is this: an object is stored in a variable, so the variable holds the object. Learners who hold this belief tend to imagine the variable as a small compartment with the whole object sitting inside it. Before reading the explanation, you were asked to commit to an answer about where the Student object is. If you pictured the object inside std, the container belief is doing the predicting, and the next sections test whether it survives scrutiny.

Think of objects as floating around independently in the computer's memory rather than being attached to the variables that use them. A dedicated region of memory, called the heap, is where objects live. The objects themselves are placed there, and the variables that need them are separate from them.

The diagram separates two things that the container belief merges. The variable std is in one place, and the object is in the heap. The only link between them is the reference, which the variable carries. The object is not inside the variable at all.

So what does a variable hold? Instead of the object, it holds the information needed to find the object. The variable carries a way of getting to the object, not the object's contents. When the program later uses std, the computer reads the reference stored in std and follows it to the object. The variable is never the object. It is the route the computer uses to reach the object.

The source is careful about language, and it is worth adopting the same habit. Saying the object is the value of the variable std is not quite true, and saying the object is stored in the variable std is simply false. The more accurate statement is that the variable std refers to or points to the object.

Shorthand such as std is an object hides the real structure. Read such a phrase as meaning that std is a variable that refers to an object. Keeping this reading in mind prevents the misconception from creeping back in as you write more code.

Because a variable holds a reference rather than the object, assigning one object variable to another copies the reference. Both variables then refer to the same object. Nothing about the object is duplicated.

After the second line, first and second both refer to the same Student object in the heap. The reference was copied; the object was not. A change made to that object through either variable is visible through the other, because there is only one object.

The reference model also explains missing objects. A variable of object type holds a reference, and a reference is a way of reaching an object. When an object variable refers to no object, it is said to hold a null reference. The variable still exists and still has a type, but there is no object in the heap for it to lead to.