But you are also being obtuse. A DRM'd copy is not the same as a non DRM'd copy, even if the underlying work is the same. If they were, there would be no reason to make the distinction. Really you're claim to the contrary is very Sophistic. I don't believe for a second that you don't see the difference.
The main issue being brought up here is that copyright protection is bundled with a responsibility to relinquish control of the work at the end of the copy protection period. So Johnny_Brahms is saying that if rights holders aren't holding up their side of the deal, then they shouldn't benefit from copy protection. This puts the cost of DRM back on the rights holder, where to date, the cost has been largely externalized.
Under this scheme, rights holders still may opt for DRM, if they expect that DRM will protect the work better than copyright, but they will actually have to do a cost benefit analysis on the matter. Under the current scheme, rights holders have all of the optionality. That is, they take all of the upside, and leave the downside for the public. The aim of this scheme is to force rights holders to put some skin into the DRM game.[1] Which seems appropriate to me, since there are significant costs associated with DRM.
[1] That is, the public pays the costs of access, as well as the cost of loss of access once the DRM authentication chain can no longer be completed (usually because the rights holder loses interest in the work and so stops maintaining the authentication system). We also observe that a lot of weak DRM schemes exist. This is only because copyright exists as a safety net for a broken DRM system. There are actual legal benefits to slapping a trivially circumvented (widely known to be broken) DRM scheme onto a work. But the upside to doing so is very small, and the downside is widely distributed across all legitimate current and future users (plus would be users that are thwarted from legitimate use by the DRM) of the work. If you take away the copyright safety net, then that bit of moral hazard would likely immediately disappear since the costs of adding bad DRM would be jointly borne by the public and the rights holder, rather than just by the public.
[edit: gave the discussion tree root (Johnny_Brahms) credit for the proposal rather than the article, since it was described as a previously held position.