Part of a Series: Cleaning the Overton Window
In the first post in this series I introduced the Overton Window theory of political change, and explained precisely what the theory maintains as it was first intended. In this post I will ask whether the theory is an accurate description of political change, and stress what political phenomena does the theory not describe.
A Distinction: Descriptive Versus Predictive Theory
It’s critical to remember that what Overton has provided with his theory is a description of how large-scale political change happens over time: the window shifts left or right, or contracts or expands, changing the policies considered politically feasible. Other theorists have also proposed a predictive hypothesis about how groups might move the window, with most suggesting that groups standing ‘outside’ the window are necessary for change in public opinion to occur (Daniel Loxton has made the same distinction between the descriptive and predictive versions of the theory in the past). It is reasonable to ask what data is available to justify the accuracy of both the description and the predictive hypothesis. This post deals with the theory as a description of our political reality: is there really a “window” of feasible policies, does it change over time as suggested, and what does the theory leave out of the equation?
Is the Overton Window an Accurate Description of How Political Change Occurs?
Descriptively, the theory looks fairly good at first glance. It is clearly the case that the range of politically feasible positions in given policy areas does change over time. We can all think of major political issues in which now-accepted ideas were once considered outrageous, and we can point to the individuals and groups who first advocated those ideas.
Consider Prohibition: this was a major intrusion of the government into people’s private lives – a policy far to the left on Overton’s scale – which received sufficient public support to make it feasible for politicians to enact through an amendment to the Constitution. But today, such a brazen intervention by the government into personal decisions regarding alcohol would be unthinkable, and no politician would be able to advocate prohibition of alcohol in today’s political climate and maintain public support. When is comes to government regulation of alcohol consumption he Overton Window has shifted quite far to the right (in the direction of less government control). An international comparison makes the case even clearer: in the UK the age required to purchase alcohol is 18, rather than 21 in the US. Fifteen US States ban underage consumption of alcohol, even with the permission of a guardian, whereas this is legal in the UK. And these policies seem reasonably popular with their relative publics: the US public does not support lowering the drinking age to the UK’s 18, and supports even stricter penalties for underage drinking (source). In the UK, the idea of raising the drinking age to 21 does not seem to receive majority support, and repeated calls to do so have been rejected by both English and Scottish Parliaments.
It seems clear that the Overton Window as regards government involvement in the purchase and consumption of alcohol is further to the left in the USA than in the UK – the UK public are less supportive of government intrusion into this area of life than the US public seem to be. But, if the situation were to change, it is conceivable that different policy options would open up on both sides of the Atlantic, just as the political landscape in the USA is massively different now to how it was at the time of Prohibition, enabling politicians to find support for a different range of policies at different points along the scale. It seems that the Overton Window can indeed shift left or right along the scale.
I also believe it to be true that, at different times and in different nations, a wider or narrower range of policies is politically feasible. On a wide range of issues, for instance, there is a narrower Overton Window in the US when compared with many European countries. In the UK, for instance, you can find people in Parliament who are socialists who advocate, with significant public support, statist policies which are far to the left of anything the most left-wing democrat could feasibly propose, such as the . And while our right-wing isn’t nearly as right as America’s, you can still find the odd MP promoting broadly libertarian policies. It is my view t hat , in the UK, the Overton Window is simply wider on a large number of issues, such that a broader range of policies is feasible.
So, something like the Overton Window of political feasibility certainly exists, and it does contract, widen, and shift over time. It probably does so slowly, as Overton suggests, and is shifted most often by social change movements who apply constant pressure to public opinion rather than by individual leaders. This description seems to fit the observed political environment today, makes sense of international comparisons, and changes in policy over time. Empirical support for this position is offered by Page and Shapiro (1983), who note that “public opinion is often a proximate cause of policy, affecting policy more than policy affects public opinion”.What the Overton Window, as Descriptive Theory, Does Not Address
Of course, this single theory it doesn’t explain everything about political change. First, there are many factors relating to the feasibility of a policy which are not related to public support for that specific policy. Politicians frequently get elected because of their likability, their family heritage, their party affiliation (which can develop at a time when the party’s policies were very different, and continue to inform voting decisions for a lifetime), their wealth – a host of factors other than their policy platform – and are then free to advocate for an unpopular policy at least until the next election.
The public are also notoriously contradictory about their policy preferences, with opinions on key political issues dependent, often, on how the issue is presented (a topical review of polling regarding views of gun control, for instance, can be found here). Zaller, in The Nature and Origin of Public Opinion (1992), provides an excellent rundown of research regarding how wildly public opinion on an issue can seem to change simply due to the wording of a poll. There are also issues on which the public has “no meaningful opinions” (as Burstein points out) – applying Overton’s theory to such policies would be a misstep, since it is not public opinion so much as special interests who are driving policy here.
Further, it is clear from the above analysis that the theory Overton developed has a limited scope – and here is where some prominent atheist authors begin to make some errors, in my judgment. It concerns how specific policy positions related to a given issue – an issue which can be represented in a scale measuring the level of government intervention – become more or less politically feasible. It therefore, in its original formation, has nothing whatsoever to say about how, for instance, maligned social groups seek greater acceptance in society. To attempt to apply Overton’s theory to explain how atheists are becoming progressively more accepted in American public life would be to misunderstand it as originally stated, and therefore to use it to develop a strategy to improve acceptance of atheists in society (as atheist blogger Adam Lee does in this article for The Modern Rationalist) would be a misstep. To suggest that the theory is about “extremists making centrists more reasonable by comparison” (as Greta Christina writes) is to introduce a concept to the theory which the original formulation does not posit. To suggest that, by making “the very concept of atheism a respectable part of the public debate”, the “Overton Window” has shifted (as Sean Carroll suggested in Discover Magazine) is to misread the theory. To apply it to policy areas where no scale of government intervention can be applied (policy areas like equal marriage rights for queer people, for example, as Adam Lee does here) is to risk confusing the matter.
These errors share a similar problem: they take a theory designed to describe a limited fact of political life and use the name of that specific theory to attempt to describe a much wider range of political situations. This may not cause a problem – many theories initially developed to describe a limited phenomenon can be fruitfully expanded to explain much more. But expanding the range of a theory requires argumentation which demonstrates why the new situation being described is analogous to the original situation – and none of the authors linked, as far as I can tell, attempt to justify the way they have broadened Overton’s theory. Rather, they use the term “Overton’s Window” as if it always had, and was intended to have, the meaning they ascribe to it. This is problematic because essential distinctions between disanalogous phenomena can be missed if a cogent case is not made as to why a theory designed to explain phenomenon A can be accurately applied to also explain phenomenon B.
Furthermore, the theory is about positions, not directly about persuasion. The theory, properly understood, says nothing about persuasive strategy. It is not a theory which has any bearing on debates over tone, or the use of ridicule, or framing. It cannot guide us when designing a billboard or an ad campaign relating to a particular message. To the extent that it addresses extremes at all it is about extreme policy positions (positions far out of the mainstream of current acceptability) not extreme ways of presenting policy positions (extreme rhetoric, for example). This is important, because the idea of the Overton Window is often misunderstood and used to justify harsh rhetoric and other persuasive strategies which much communications research shows to be ineffective.
This question – of the strategic implications (if any) of the descriptive theory – will be the focus of the next post in the series.