I must take issue with the classification of "supposed" as an adjective, particularly sense 3. I do not see "supposed" as an adjective, but instead as a verb tense. I will argue my point (on the internet, haha).
When we look at language, two common uses of it pop up in the context of general semantics: descriptive use and prescriptive use.
We describe when we are trying to represent the reality we have experienced. We piece out the judgments we have have had about the reality we experience and use language to report, and denote those judgments as such. E.g., instead of saying "That horrible driver cut me off," we say "That driver drove in front of me rather abruptly, and his action really upset me."
We prescribe when we are trying to represent the reality we want to experience. Prescribing is notably different from describing, as with describing you deal with the reality you've been presented, but with prescribing you tend to ignore the reality that has unfolded and revert to expectations you have for it. When that driver drives in front of you rather abruptly, you shout from behind the wheel, "He shouldn't do that!" and continue to get mad about his contradicting an expectation you have for people that may not be one he lives by. We then turn around and "teach" other people about reality by reiterating those shoulds, so we say to our child as she learns to drive that "people shouldn't cut other people off," rather than saying "sometimes people will cut you off in traffic." Which statement prepares your child better for what she'll encounter on the roads?
Prescriptive language is populated with many different "should" statements. These are statements like "Democracy should prevail in all nations," "You shouldn't get bad grades in school," "You should get married and have kids," etc. It came to my attention in taking a German class in high school that the modal "soll," which translates in English to "should," in one German textbook was also translated to "is supposed to."
I found this very interesting. I hadn't thought of "should" that way. When I later went to college and became a practitioner of E-Prime (something I don't do some much these days, admittedly), the "hidden 'is'" inside a should statement (is supposed to) deserved consideration. At the time, I never questioned that this manifestation of "is" was an "is" by passive construction, rather than an "is of predication" which the dictionary entry above would have suggested. The truer nature of the should statement was amazingly revealed to me. A should was a secret passive construction, and by its passive construction, a "supposer" was oftentimes left out of the statement.
For instance, take the statement "Democracy should prevail in all nations." If we reworded that statement as my German textbook might recommend, it would read "Democracy is supposed to prevail in all nations." At this point, the authors of the above dictionary would stop and say that "supposed" is an adjective. But I took it as a verb form, making for a passive sentence construction, and per Bourland's advice I wanted to rid of the "is" and make a more active statement. So "Democracy is supposed to prevail in all nations" became "(Someone) supposes democracy to prevail in all nations."
This treatment becomes really interesting now. Who is it who is doing the supposing?? In this example, I fill in the blank by saying "someone." If we actually have an answer, say the President, then the statement becomes "The President supposes democracy to prevail in all nations." And then we can decide for ourselves whether we agree with the President. Without the supposer, we are often tricked into blind following. We hear a should statement as if it's been determined by some amazingly important, intelligent person worthy of the highest respect, when it just might be your friend who is trying to manipulate you into doing something he doesn't understand ("You shouldn't be an actor!!").
Prescriptive language is obviously prevalent in religion, and I'm sure that in many of the Western religions this "mysterious someone" would be deemed "God"--it is "God" who is doing the supposing. But such suggests that "God" was able to directly communicate his expectations of people. More notably, though, it seems that pretty much anyone can configure a should statement and make it seem as if a "God" is doing the prescribing. "You should get married." "Says who?" "Well, um ... says God, I guess." "No, actually, I think your parents just expected that of you, you complied, and now you're trying to expect that of me. I don't agree at all with that perspective."
It seems to me that recognizing a should as a secret passive statement that masks a supposer (or "prescriber" or "actor") would be very helpful for critical thinking. When someone shoulds you, you can respond with "Says who??" and now know why given the retooling of the statement I provided. And if you're a sufferer of "shoulditis," a particularly pernicious way of living that can provoke a lot of anxiety, a term I believe Dr. Albert Ellis coined, this analysis might be helpful in breaking the cycles of shoulding yourself by getting you to analyze a) who is actually supposing you (your parents? God? does God even exist?), and
whether you agree with the supposer's perspective if one does lurk.
If we treat "supposed" as an adjective, I don't believe we can arrive at these discoveries as clearly.