Could you expand a bit on how parameterized queries are not sufficient for defense against SQL injection (assuming, of course, that developers use the escaping and do not concatenate unescaped data into queries)?
As for them being required -- you can obviously escape queries yourself, but the normative reference for escaping is the target database itself, and reproducing escaping locally in the client brings with it the likelihood of introducing an error in the custom implementation.
Not every "input" to a "query" (using these terms loosely) can be bound as a variable. Simple example: ASC and DESC. There are trickier examples that are still common.
It is a bad idea for applications to implement quoting regimes, and it is a bad idea for frameworks to try to create one-size-fits-all quoting regimes like PHP used to. That doesn't mean it's a bad idea for a framework's e.g. MySQL support to provide the capability of sanitizing MySQL inputs under a common database API.
The problem with that is that almost nobody uses the official database APIs. The official APIs are usually C libraries (e.g. libmysqlclient) but pretty much everybody uses third party wrappers (e.g. Perl DBI, the mysql/mysql2 gem for Ruby, the PHP default MySQL bindings, etc). Few people program against the database in C or C++. It was and is up to the third party API providers to provide easy sanitization APIs, I don't see how MySQL could have changed that situation by providing such APIs themselves.
As for them being required -- you can obviously escape queries yourself, but the normative reference for escaping is the target database itself, and reproducing escaping locally in the client brings with it the likelihood of introducing an error in the custom implementation.