I shouldn't feed the trolls, but just for the record:
This is extremely and wildly not true. The most basic part of doing literature review is following original sources and everyone I know does this. You have to, because reviewers pick this stuff up.
Actually, what I said is true. As a reviewer I DO pick this stuff up. And manuscripts with inadequate citations are rejected. Many submissions come in lacking any citation to a source (say, from 25 years ago). They will instead cite one of their buddies who parroted the primary 2-7 years ago. If it is a new interpretation or whatever, of course the more recent (primary source) who did so should be referenced.
Also, you're fooling yourself if you think that just because something was done 30 years ago, there's no point in citing more recent sources. A lot of more recent work is nothing more than just repeating old ideas but with slight modifications that nevertheless reveal new insights.
Whoever FIRST reported a specific observation, measurement, or hypothesis should be cited. Credit where it's due. All science is built upon previous work, so intelligent researchers cite the appropriate sources.
Finally, when writing a paper, there is no need to cite everything that has been done right back to ancient Greece. The audience of a scientific paper is assumed to be the scientific community which is already familiar with the body of work.
Ancient Greece? Thanks for the straw-man. RULE OF THUMB: If it is in books, there is no requirement to cite the originator.
If, OTOH, you are extending the theory of so-and-so, you had better cite the primaries (or the most recent in the Ref. chain who modified it). Otherwise your manuscript WILL be rejected, by me or by anyone other referee.
Last, if you're writing for submission to Science or Nature, targeting a broader audience than your specific field, it is imperative that you cite proper sources. I have a feeling that you've never published in either place, in any other prestigious journal, or probably never in any archival, peer-reviewed journal.
If you ever get to the point in your career of being asked to referee, you will see the huge volume of 'minimal-effort' submissions that must be screened-out to maintain the quality of a given journal. Copying and failing to cite is the hallmark of bad (rejected) journal submissions.