Why can't a top level class be static in Java?

All top-level classes are, by definition, static.

What the static boils down to is that an instance of the class can stand on its own. Or, the other way around: a non-static inner class (= instance inner class) cannot exist without an instance of the outer class. Since a top-level class does not have an outer class, it can't be anything but static.

Because all top-level classes are static, having the static keyword in a top-level class definition is pointless.

Some code to play around with:

public class Foo {

    public class Bar {
         // Non-static innner class
    }

    public static class Baz {
         // Static inner class
    }
}

public class Example {
    public static void main(String[] args) {
        new Foo(); // this is ok
        new Foo.Baz(); // this is ok
        new Foo.Bar(); // does not compile!

        Foo f = new Foo();
        Foo.Bar bar = f.new Bar(); //this works, but don't do this
    }
}

I put the "but don't do this" in there because it's really ugly code design. Instance inner classes should not be visible outside the outer class. They should only be used from within the outer class.

Simply put, a top-level type declaration cannot be static, because the Java Language Specification (JLS) doesn't say that it can be. The JLS says this explicitly about the static keyword as a modifier of top-level classes:

The modifier static pertains only to member classes (§8.5.1), not to top level or local or anonymous classes.

However, the accepted answer - which has many upvotes - says that this is because top-level classes are implicitly static "by definition", so the static modifier would be unnecessary. That is wrong.

The word "static" appears in the JLS in quite a few places, but never to refer to top-level type declarations. Here is an exhaustive list of things that can be "static":

• Static fields, also called static variables, including static constant variables
• Static methods
• Static member type declarations
• "Static members", which are the three constructs above
• Static initializers
• Single-static-import declarations and static-import-on-demand declarations, which are top-level declarations, but not type declarations. Here, "static" refers to the names which are imported, not the import declarations themselves.
• The language is statically typed, and expressions should have statically known types so that their safety is "statically guaranteed".
• The way that names, including field accesses, are bound at compile-time is referred to as static resolution, or static binding.
• A lexical context can be a static context.
• The invocation mode of a method invocation expression or method reference expression can be static.
• The class name used in one example implies that checked exceptions declared in a throws clause are statically thrown.
• Part of the memory used by the JVM is referred to as static storage, and the same section refers to "static" linkage in the C programming language.
• The preface to the JLS mentions static analysis tools.

There are no uses of the word "static" in the JLS to refer to top-level type declarations; so as well as not being explicitly static, they are not (and cannot be) "implicitly" static, by definition.

static can be added nested classes of an interface, even though this is the default.

I believe static cannot be added to top level classes because initially there were no nested classes and you couldn't add static to any class.

Later nested class were added and static could be added to nested classes, however there is a tendency not to change the syntax any more than needed so it wasn't added to top level classes. (as there was no need/benefit)