Very much this. The assumption is that papers only cite good research, but is something is really off, I have personally cited papers saying that the people that wrote it have no clue (with evidence of course). I have very rarely seen it done by other authors though, but that may be due to my field (CS).
The other thing is that if you do good research and explore interesting side-aspects, you are never getting a permanent academic position. Those go exclusively to people with a lot of publications (which is a bad sign in itself...). The system promotes bad scientists into positions where they can do and supervise more bad science. It is really a complete mess. And I do understand why so many industrial CS people have an utter disdain for published research, most of it is just so terribly bad it is staggering. To make matters worse, much of these terribly bad publications look good on the surface as that is required to get them accepted. But I have found outright fraudulent publications at Tier-1 conferences, misleading ones and ones that claimed findings without any proof whatsoever. I also know several people that should have their PhD removed, because they did not have the results they claimed they had. They were just clever enough to publish in a venue where the reviewers were impressed by the names on the paper or the writing, but failed to spot the often subtle but critical errors. (No, anonymous review does not help. People that want to benefit from the names of their advisors just publish a technical report that is the same as the paper and make sure Google finds it. Many reviewers even at first-rate conferences are too lazy to do a real review and instead first check whether they can identify the authors and just decides on the names if they are successful.)
