I'm kinda liking the idea of scientists willing to put their heads on the block for us. Cheers, Armstrong.
Journalistically, I understand the need to show both sides of the argument and perhaps stir up interest in one's article by ending on an odd/controversial note. It's also completely reasonable that Armstrong said this (I'm going to assume he was quoted accurately), if just a little unfortunate under the circumstances. But here's the thing - Armstrong said this, in my view, because he sees complexity, sees the exceptions, is keen to be truthful and accurate and comprehensive and not overly simplistic in what he says. This is how I like my scientists, and it's how I like them to talk to the media, even if some readers find that confusing. PACE authors on the other hand, tend to give a lovely, simple story to the media. Have you ever read a report where one of them has said that biological factors (other than those resulting from deconditioning, grrr) may play a role in some cases? Has one of them ever said that some cases have no psychosomatic component? Never mind most/all cases. (Rhetorical.)
I do think that in general it would be helpful if scientists were careful, or rather, media-savvy, about what they say when it comes to these psych questions, so that less discerning readers don’t get confused, but I’m also aware that they’re not in control of how what they say is reported. It's possible that the journalist asked him "Are you saying that psychological factors or brain chemistry do not play a role in a single case of CFS? Not a single one? Not one?" and that he answered (hypothetically) "No, they could be a factor in some cases." It's also possible that he's referring to the kind of psychopathology seen in some patients with neurological diseases, e.g. anxiety in some people with Parkinson's, which seems to be part of the Parkinson's/disease process rather than a reaction to it. Since we don't know how he meant this, I'm not reading anything into it, particularly given his very clear statement that it is not psychosomatic. He couldn't have been clearer on that.
As long as they keep doing the work, finding key pathophysiological factors that will lead to effective treatments, I’m happy. When we improve with immunomodulators/other treatments, that will speak for itself. ‘Cos rituximab isn’t known for its faulty-illness-belief-modulating activity.