Imagine you're a judge, deciding a First Amendment case involving speech that you know many (though not all) people find offensive or even hateful. You conclude, however, that the challenged restriction on the speech is unconstitutional, and now you're writing your opinion. In terms of how you frame the speech in question, you have two main choices:
(1) You can emphasize the "good" of the speech -- why it may be valuable, or challenging, or part of an important American tradition; or
(2) You can emphasize the "bad" of the speech -- how it is legitimately offensive, or injurious, or hateful.
Which is the more effective rhetorical move in persuading an audience which, for the most part, views the speech as some form of horrible?
The former seems like the obvious play: by explaining why this speech serves a valuable function in society and by locating it in a grand social tradition, one validates the decision to give it constitutional protection. We appeal to the audience's better angels, and basically tell them that while they don't have to like the speech, it's not unadulterated awfulness.
But the latter has something to say for it too: by emphasizing the negative aspects of the speech, one shores up the notion that it truly is being protected because that's the position obligated by the law. A casual reader of the first type of opinion might suspect that the speech is being protected because the judge is sympathetic to it (look how she waxes lyrical about its virtues!). The reader of the second opinion, by contrast, can be pretty sure that the judge has no extrinsic sympathy for the speech, which then suggests that the ruling really must be compelled by the law (otherwise why would the judge do it?). This, of course, is the hypothesis I explore in my Sadomasochistic Judging article.