[mjn]

It depends on the AST, but most languages' ASTs are quite unwieldy to work with directly. They get marked up with all kinds of semantic information useful for a compiler or interpreter quite early in the parsing process, and are often more explicit and regularized than the syntax is. For example, you can dump a Python AST with the 'ast' module, but it's not pretty. The statement "x=5" balloons into:

    Assign(targets=[Name(id='x', ctx=Store())], value=Num(n=5))

I think a comment up this thread (https://news.ycombinator.com/item?id=16386702) is right that Lisp manages to make this work because the s-expression structure of Lisp source code isn't really an AST, though it has some relationship to one.

[kazinator]

That Python doesn't look like abstract syntax; it's an intermediate form full of additional semantic objects geared toward further translation or interpretation.

A Lisp compiler might build up something like that, from a fully macro-expanded body of code. That's not usually accessible to programs.

Like the fact that n is converted to a Name() node, where it has a ctx property indicating Store() has nothing to do with syntax; it is the result of an analysis arising from how x is being used semantically. The user didn't specify any such attributes on the x just that x is assigned and that is already recorded by something like Assign(x, 5).