Thanks, Stephen, for pushing this. I wish I could eaxipln it in more detail, but at the moment it is difficult to follow all the events in addition to all the trivial demands of the day that academic work brings with it. But here is a thought-experiment. Think about the exactly same events, exactly same numbers, same transmission rate, same case-fatality rate, but with an avian influenza virus as its cause. I think the pandemic would have been declared a week ago in that case. I think at this point they can't yet declare a pandemic because the political pressure is extremely high (as well as the stakes), because it's not the expected virus subtype (and thus not the story that experts have been telling again and again), and because there have not yet been enough fatal cases. The technical question of classification is a part of this, but not the whole story. In public and expert discourse, the virus has already been identified as new, without making explicit in what sense it is new (obviously not in terms of subtype, so then what?). The fact that it is called a swine flu virus seems to make all the work here. But of course, you will find in swine populations both swine flu viruses as well as human viruses. What makes a flu virus a swine virus and what makes another flu virus a human virus is largely unknown. You simply find some only in swine, or in humans and in swine, or only in humans (today.) In technical terms: this is a question of specificity (of host range), but not of immunity.