At the time it wasn't obvious. In those days information traveled much more slowly, with many more intermediaries than today and was, therefore, much less reliable.
I think that's the wrong framing, at least for an organization like the NY Times.
For one, there's a difference between "a lie" and "an inaccuracy" even though most of the time (but not always!) lies are a subtype of inaccuracy. If you make a similar conflation in software development, you'd say that a team member who pushes buggy code to your repo sabotaged your project instead of saying he made a mistake.
Also there's a difference between an organization and its members. Those individuals you mentioned may have lied, but that doesn't mean the organization lied. If a disgruntled member of your team does try to sabotage your project, that doesn't mean your team is sabotaging the project.
Duranty did his reporting literally 90 years ago, which is several generations in the past. That case is of limited use when reasoning about the organization today.
Miller was criticized by colleagues and was eventually forced out. She seems to me kind of like a software engineer who came up with some grand design that had fundamental problems, but couldn't let the design go. That's mistaken, arguably incompetent, but not sabotage.
More recently, they allowed themselves to be used as the uncritical mouthpiece of the American federal government when they were starting an illegal war predicated on lies: https://en.wikipedia.org/wiki/Judith_Miller
If the NYTimes isn't pure, then surely none are. Genocide denial is particularly egregious, but at least they eventually came clean. I still read the NYTimes more than any other paper, but we must always remember to keep an open mind and to never be too trusting of what any newspaper says.