But the key is that the first read, the morning one, involved skimming articles for their gist and reading the ones that interested me. The afternoon session possibly involved reading more, due to boredom.
The real problem the newspapers have is that most of the articles they produce fail the skim test. I don't want to read more. I get the gist. I'm not interested. If they're good I click through and read them on the original site... often then clicking to other things I find there. Win for the Google link to them. If it's a hot topic I'll often read many articles from many pubs on the same topic. But in a world with so many news sources the truth is they don't stand out, I am not going to check a thousand sources very day, or even ten. So I'm not going to wade through to find the things I'm interested in. If they don't want their headings and summaries in Google I don't care about their product.
And I'm a Google user. I pay Google's bills. I don't want them raising my rates in other areas to pay publishers for links to content that I may skim over and never want to read in depth. Why should my costs for Google - in terms of what they can deliver me for all that surplus attention of mine they harvest over a year - go up to subsidize failing newspapers producing product for which the market rate doesn't support their costs?
So taking all the philosophical statements out of the way, what is the econimical case that the newspapers are making? If Google actually showed a bigger part or the whole articles, they'd obviously be taking revenue from newspapers, but it's not the case.