First off, Overton Window is neither a hypothesis nor a proven theory. It's an attempt to label an existing observed phenomenon.
You can think of Overton windows as a spectrum of ideas, with each idea on a spectrum having more and more support/political acceptance; till you get to the level of idea being an official government policy; and then further up the spectrum less and less support again.
Let's look at the usual example (based on the post by the person who first popularized the concept, Joshua Trevino): Education policy.
In a typical state, say Michigan, there are a variety of options regarding the State's involvement in children's education.
Each option has a certain level of popular support, which are roughly grouped into 6 bands: Unthinkable/Radical/Acceptable/Sensible/Popular/Policy. The last band is the level of acceptance which is high enough that the government can implement the idea as policy.
The point made by Overton (aside from the bands) is that usually, a very narrow range of ideas on the spectrum have political acceptance.
| Acceptability | Idea |
| Level | |
| Unthinkable | No government involvement in education. |
| Radical | All schools private with government regulation. |
| Radical | Voucher system with public schools. |
| Radical | Tuition tax credit with public schools. |
| Acceptable | Homeschooling legal. |
| Popular/Policy | Private schools restricted. |
| Acceptable | Homeschooling illegal. |
| Radical | Private schools illegal. |
| Unthinkable | Children taken from parents and raised as janissaries. |
You can also simply think of it as a graph of what % of people's support is required if we want to achieve a certain level of political acceptance:
It may help if we show a graph and a table
Here's my own graph (please note that the #s are made up - I'm not aware that Overton meant specific #s, or that mine are even remotely accurate. Apologies for somewhat unparallel lines - making graphs by hand in Paint.NET is not the best idea ever :)
In the graph, the Blue letters are the Overton Acceptability scale, and they are graphed against the % of population which should support the idea before it reaches that level acceptability. Here's that graph in a table form:
| | Acceptability | % popular | |
| | Level | support | Comment |
| U | Unthinkable | < 1% | Not 0, some wackos always like ANY idea :) |
| R | Radical | 10% | Some # |
| A | Acceptable | 25-30% | Majority of one political party supports |
| S | Sensible | 40% | Independents start liking it |
| P | Popular | 53% | A majority outside margin of error. |
| Po| Policy | 60%+ | Because filibuster |
To re-iterate, the actual percentages in the table are made up by me based on some well-reasoned logic, not actual numbers proposed by Overton (he didn't propose any) or formally proven by science. I just feel that having actual numeric estimates illustrates the idea better.
Please note that there is a related concept that sometimes gets confused with "Overton Window" - and that is "moving Overton Window".
The idea is pretty simple: if we look at the spectrum, and the idea we like is NOT within "Acceptable" range right now, we want to do something to make the Overton window move in the range, so our idea is going to be acceptable. Or in reverse, to make the window move so an idea we oppose but is currently acceptable becomes Un-acceptable. (e.g. Radical/Unthinkable).
For example, in 19th century, the idea of owning slaves was quite acceptable, and was a policy. The goal of the Abolitionist movement was to move the Overton window so that the idea became unacceptable to majority of population. Or take drug legalization. In 19th century, it was quite acceptable to take recreational/medicinal drugs. Later, the Overton window moved so that it became unthinkable/radical for drugs to be legal. Then the hippies made the idea of legal drugs acceptable, or at lease sensible in the 60s; and libertarians moved it from acceptable to at least popular, and with some states legalizing it, to Policy.
And further, another notion that gets confused, is a very specific method for moving the Overton Window (as discussed above) by openly proposing Radical/Unthinkable ideas into popular conversation; with the effect that less radical but unpopular ideas further up the spectrum become Acceptable or Sensible.
Now, I have a feeling that in your bullet point questions you were actually referring to this specific concept and not to an actual static Overton Window I defined above, based on the Skeptics link.
As such, the fact that this method works is not really a proven theory, and more of an observation of how the human decision making works. It's not even specific to politics, the same exact psychological concept works for example in business:
You need to convince your boss to use approach X to solve a problem
You favor a solution Y that's not very acceptable to the boss at the moment.
Instead of proposing Y, you propose 3 distinct solutions, "Y, the bad, and the ugly"
The boss is more likely to agree to Y, after seeing the awfulness of the other two solutions. Why?
By comparison, the negatives of solution Y seem insignificant (this is called the decoy effect).
By being able to reject the 2 bad solutions, you "sate" the boss's need to "perform managerial functions" (aka "I gotta provide some criticism here").
This isn't necessary with good bosses, and totally useless with either great (already has a correct vision/solution) or poor (already has A vision/solution, not necessarily correct, and not willing to stop micromanaging :)
As an example, take a look at the healthcare discussion in the USA. This is very instructive because it was exactly the topic discussed in the DailyKos article I linked to in the very beginning, where the authors - in 2006 - proposed that using this method of moving the Overton window is desirable for Progressives overall and used healthcare in partucular.
Before, the concepts of fully nationalized healthcare, single-payer system, requiring everyone to have health insurance, and forcing people to pay for other perople's health insurance via a tax were... not very popular shall we say. Witness Hillarycare and its failure's subsequent negative effects on Clinton and DNC.
As David Atkins advocated in the DailyKos article, the progressives, instead of running scared from the single-payer idea, should have used it to drive the changes in Overton window.
Which is exactly what happened. When Obama administration and liberal Congress of 2008 started discussing single-payer system and how NHS is great, the idea of requiring everyone to have insurance and have taxpayers foot the bill all over the sudden became "acceptable" and reached 40% popularity, even if it never reached the "popular" Overton status (current polls have Americans reject the idea 54% vs 44%, despite it having already become "policy").