English source code is a universal language.

I’ve never seen a need for localization beyond domain terminology. And I think it would be a huge detrimental.

To implement it would be unnecessary significant complexity. Effort better spent elsewhere. And for programmers it’d be confusing. Think code snippets, mixing content, and the need for reserved word expansion or exclusive parsing scopes that would be even more complex and confusing.