He is insisting on the 'dogma', but things have moved on and a lot more is known now, just as it has happened for the famous Koch's Postulate - I don't think that in Koch's days much was known about the intricacies of the immune system and the very complex cellular interactions; I am not surprised that it is now obsolete, but he is mentioning it just to score a point. Not very scientific.
As I see it, Koch's postulates are intimately associated with germ theory, where the idea is that contagious diseases are spread by organisms. For present puposes, although viruses aren't in a strict sense organisms, let's for convenience include them under this umbrella. For the idea to work, there has to be some kind of contact of infectious agents with target organisms, either direct or indirect.
I don't think that the "postulates", as they are termed, are obsolete. For heaven's sake, if viruses are thought to spread disease, then to conclusively prove that is the case, one needs to be able to isolate them and show that they and they alone cause specific diseases.
The smaller an organism is, the more difficult it becomes to isolate and prove it to be the cause of a specific disease. One can do it with some organisms, particularly multicellular or large enough unicellular ones, because they can be isolated and manipulated. But viruses are so very small that one can't directly observe their life cycles. Open any biology textbook (or look at
this web page) and you will see illustrations or descriptions of their life cycle, and for most of my life I have simply accepted this without questioning it.
Cowan may be right or wrong, but he has a point. If Koch's postulates are "obsolete", then why do virologists insist on using the term "isolation"? Seems to me that they recognise that for for proof they have to show that viruses are disease vectors.
Fact is, viruses are extremely hard to isolate. I have my doubts that anyone can produce a test tube containing
whole virus particles,
and nothing else (no organismal fluids such as may be found in nasal mucus, for example) suspended in some kind of neutral medium (e.g. pure water). I stand to be corrected on that, so if you or anyone here thinks otherwise, I'd welcome a link or two to evidence.
As far as I have been able to determine, viral life-cycles as commonly conceived of are
illustrations of what is thought to happen, but no one has ever actually observed this. Rather, snapshots (electronmicrographs) have been taken, and inferences have been made, but it's not like following the life-cycles of more visible disease vectors like, say, malarial parasites.
Virologists may resort to employing indicators for them, such as the presence in tests of putative viral nucleic acid/coat proteins. They have to be sure that such things are actually in the virus and do not have some other source of origin. That is very difficult if the medium containing them is organismal in origin and likely to be contaminated with cellular proteins/nucleic acids, even celluar debris.
Additionally, the discovery of endosomes does rather obfuscate the issue, because they look quite similar to at least some viruses. I mean, what is the evolutionary origin of viruses? I can only think that they originated in cells, because they contain nucleic acid, which as far as I'm aware only cells can produce. How would a virus ever have produced nucleic acid on its own? After all, it's just a protein coat enveloping a nucleic acid core. It contains none of the machinery needed to generate itself; indeed, it can't reproduce unless it utilises the machinery of cells.
The role of viruses in horizontal gene transfer is increasingly being recognised, which does raise the issue of whether viruses aren't in fact endogenously developed mechanisms of cells of one organism used for nucleic acid transfer to cells of another organism of the same (or even different, I suppose) species.
We might conceivably have the concept of infectious diseases not quite right. "Viruses" could be entities generated by cells that have, among other purposes, that of combatting, rather than causing, disease. At least some contagious diseases might have other, as yet unknown, causes than viruses; but the latter might commonly be associated with them. They might, who knows, be like the firefighters at a fire who are assumed to have caused it rather than be there to help put it out.
That there might not be such a thing as contagious disease wouldn't be my position. In some cases, organismal infectious causes have been pretty much definitively established. But it's a different thing to say that there's a definite subset of contagious diseases caused wholly or in part by viruses. In my eyes, that's a hypothesis that might or might not be true.
I don't see what's wrong with merely considering whether such an opinion might have some merit, and this could be worth investigating. To close the books and consider the issue settled seems somewhat myopic, and maybe even irresponsible. The medical and allied professions have a long history of closing books only to discover they have to reopen them after having fought tooth and nail to keep them closed!