Clearly, the gold standard in citing printed material exists, and that gold standard includes a page # for where the citation is taken from (especially large books).
... provide a specific pointer to the page or section you are using as a reference. A reference to a "general impression" of a book or a website is not a good reference, in my opinion. Quoted Meta.Skeptics.SE answer was skewed towards online material, but same citation standards exist everywhere for printed, e.g.:
Give only the information needed to identify a source. Usually the author's last name and a page reference suffice. (MLA citation style)
Presumably (i'm willing to be proven wrong in this assumption), that page # in the citation standard is there so that a citation can be easily verifiable by getting the printed material being cited and being able to find whether it matches the cited data by opening correct pages (which takes under a minute) as opposed to skimming/reading a whole book which may take hours.
However, if a citation provides the printed material ID yet omits the page number, it's still possible to verify the cite's correctness - merely less easy to do so (devil's advocate point: for hard to obtain printed material, the main cost of verifying is obtaining access - which can take days and money - and NOT reading the whole book).
In light of these considerations, what is our official position on the acceptability of references that note the printed material with large page count being cited correctly, BUT omit the page #?