I recently ran across this critical view of “bureaucratic language” in journalism. According to the writer interviewed:
“[Bureaucratic language] removes the actors from their actions, thereby obscuring attributions of responsibility and leaving the reader with little information about cause and effect. These absences are particularly problematic when it comes to reporting on complex political and social issues. If one of the roles of journalism is to promote accountability, then our reporting must not shield responsible actors through our syntactic structure.”
The example in the article is that of an “officer-involved shooting” – which is a particularly tortured way to avoid saying directly that an officer shot a civilian. The construction is not technically the passive voice, but has the same effect, erasing the person who’s actually doing something, in this case, shooting someone. There are other colorful examples of how powerful actors are obscured through the “debasement of language" in George Orwell’s 1946 essay “Politics and the English Language."
For me, one way to avoid this weaselly wording is a grammatical principle that I initially encountered a few years ago in the University of Chicago's "Little Red Schoolhouse" and the related book Style: Toward Clarity and Grace, by Joseph Williams. The principle is now the basis for much of my line editing.
As I'm reading a draft manuscript, I identify its main “characters,” the handful of things (human or nonhuman, concrete or abstract) that the text is intended to be "about." In academic articles, these are often the keywords.
Then I go back through the document line by line and make suggested edits so that most (not all!) of the sentences have those characters as their grammatical subjects. Of course not all of the sentences need to be structured that way, as there are other details or tangents that the author likely wants to cover, plus varying the cadence a bit is typically a good idea.
One area where I commonly make a lot of these edits is in literature reviews, where the authors of various studies are often written in as the sentence subjects, even though they're not really characters. The characters are probably better conceptualized as those scientists' questions and results. Occasionally another author might be considered a character, if, say, an academic article is directly replicating another study's methods or is trying to extend a theory constructed by a particular scientist who is worth mentioning at length. But for a study that is referenced and described in just a few lines, there's no need to name the author(s), let alone put them in the all-important subject position.
This approach is mostly about improving the clarity and readability of a text, but it can also help draw attention to the actors who might otherwise (want to) stay out of sight.