How accurate is the horseshoe theory?
The horseshoe theory in political science asserts that rather than the far left and the far right being at opposite and opposing ends of a linear political continuum, they in fact closely resemble one another, much like the ends of a horseshoe.
A few more examples to clarify:
- extremist Muslims vandalize bikini ad for socially conservative reasons.
- radical feminists doing the exact same thing for socially liberal reasons.
- anti-racists go so far that they end up being racist.
What do political scientists generally think of the horseshoe theory?
I think your examples show similarities of behavior, but not necessarily similarities of intent/philosophy.
Yes, but not in the simplistic left/right way it's presented.
Some of those charts illustrate and explain the reason why horseshoe theory fits pretty well (this one is from Political Compass organization):
If you notice, the left and right both have an independent second dimension - liberty vs. authoritarianism or totalitarianism (in other words, how much power an individual has vs. the state). If you draw a horseshoe - either upside down, or upside up, it would fit.
Both the authoritarian left and right are authoritarian at the core, and frequently have more to do with each other - especially concerning methods and tactics and view of human nature - than the libertarians on either right or left.
This is why you hear right-libertarians in USA complain about G.W.Bush and "big government republicans" - while far from "far right", the latter are just as happy to grow the power of government, they simply wish to use it for their preferred ends and not the ends of the Democrats.
To expand on the similarity of approaches:
The overriding human-nature philosophy of an authoritarian is that most people aren't fit to govern themselves, and only select few (with the right ideas and the right skills) are to be entrusted with power.
The preferred strategy is to pass intrusive laws to enforce desired approaches and outcomes; and harsh enforcement on those violating said laws.
Heavy use of propaganda and icons.
Governing approach frequently characterized by lack of interest in efficacy of approaches (as random examples from both extremes in USA, abstinence-only sex ed on the right and pouring endless money into schoolrooms with zero improvement on the other, though this isn't always confined to the extremes).
Viewing the political opposition as the enemy, not the opposition.
Finding a specific group to vilify (frequently, but not universally, Jews work out pretty well, both for left and right)
At the far extremes, willingness to use violence and terrorist tactics to achieve goals.
Also, slide into petty crime to finance themselves (some far-right militant groups in USA financed themselves through crime - ignoring the deep irony that so did far left anarchists 100 years ago, and so did Koba Jugashvili)
At times this horseshoe convergence become worthy of an Onion article. In 1969, a Marxist leftist group plotted to bomb a Jewish community center commemorating 31 years since Kristallnacht.
Quite interestingly, people that are "extreme far right" according to the left-wing (that is, fascists), don't consider themselves to be on the right (or left) at all. Given that I'm not an expert on fascism, let's ask an expert, shall we?
Benito Mussolini in 1919 described fascism as a movement that would strike "against the backwardness of the right and the destructiveness of the left".
Hitler named his Nazi party "NSDAP", which means National-Socialst German Worker's Party.
 - of course, as always in politics, nothing is quite as black and white. One of the most authoritarian lefties ever, V.I. Lenin, famously pontificated on the topic of "We know that an unskilled labourer or a cook cannot immediately get on with the job of state administration" - but followed up with admonition to teach said labourers to govern. Whether he was genuine about the latter or not, we all know how that evolved in the state he built - only the Party elite had real power
 - In a non-extreme way, we have people like NYC ex-mayor Bloomberg, who basically seamlessly switched from D to R so he could run for mayor on a totalitarian platform of not allowing people to drink large-sized sodas. Or every single Party Bigwig in USSR suddenly becoming a bigshot "democratic" leader in post-Soviet space. In a more extreme way, see the latter bullet #3 about fascism
 - ... and most people chanting "Death to Jews" in Europe in 2014 were Socialist party voters, not Marie Le Pen ones.
The interesting thing about drawing a horseshoe on that graph is that it *can* rotate relative to the traditional linear spectrum. So at one point in time the Libertarian Left might be considered "far left", with the Libertarian Right being "far right", but at some later point the horseshoe may have rotated clockwise 90°, so that the extreme ends are both "Left" but differ on the Authoritarian/Libertarian axis, and the "Right" represents the more mainstream view. Or vice versa with a counter-clockwise turn. The current horseshoe is not the permanent ordering.
Also, I'm not sure if extremism should be a third axis (opposed to accepting? Inclusiveness?), or whether that's represented by the edges of the graph.
@DVK Do you have any evidence that two dimensions accurately describes people's political behaviour? DW-Nominate scores suggest that one dimension is sufficient to describe the voting records of Congressmen in particular, but it may be possible that two or more is a better indicator of people's political opinion in general. Still, it would be nice to see evidence for this..
@Avi - this MAY have something to do with the fact that there are virtually no non-authoritarian congressmen? :) Also, there's a difference between a political opinion and an option to vote for a finite set of bills that come to a vote in Congress.
@DVK Yeah, I think it's quite likely that fewer dimensions are required to model congressional voting behaviour than people's political opinions, but I haven't seen evidence that any specific number of dimensions provides a good model for people's general political opinions.
@Avi - to paraphrase the "faster than the bear", the 2D model doesn't have to be good. It just has to be better than absolutely useless 1D model
@DVK For all I know it isn't, or it isn't any better than the 1D model than a 3D is than a 2D model, and so on inductively. That's why I am requesting some form of evidence.
@Avi - nothing can be worse than 1D model, in which Stalin is closer to Daniel Patrick Moynihan than he is to Hitler (and Hitler is closer to Scott Brown than to Stalin).
@DVK unless it's a 1-dimensional horseshoe as proposed by OP. I could easily argue for any n-dimensional model by that logic, as it provides more information than an (n - 1) dimensional model, but at a certain point we're not gaining any useful information. How do I know that point is two-dimensions? Maybe three are necessary?
@Avi - the **only** assertion my answer is making is that 2D (of almost any kind) is better than 1D. It's quite possible that 3+D is even more accurate - however, that is not necessary to provide **an** answer (although a more accurate ND model may result in an even better answer for all I know)
@DVK I think it is necessary to provide an answer. "Not as accurate as a model with more dimensions" is trivially true to the point of meaninglessness if you think you can establish it deductively. It doesn't tell us how accurate OP's model actually is or what political scientists think about it.
@Avi - which part of the linked Wikipedia article leaves **any** doubt that political scientists think that 1D model is inadequate?
@SQB - sorry, you need citation **for a citation**? (The bullet quote in #3 was **a direct quote** from Mussolini. You just don't get more primary-source than that).
@SQB - if you meant political leaning of French Muslims from my 3rd footnote:- I don't have fresh polling data at the moment, but here's 2012: http://www.businessinsider.com/muslims-hollande-france-sarkozy-2012-5
@ user4012 that's indeed what I meant, but I hadn't thought of anti-Semitic Socialist Party voting Muslims.
This answer argues that instead of a planar horseshoe, it's a more like a not-quite torus -- *i.e.* a *3d* horseshoe. OK, but that's still a horseshoe. I like the real-world political examples though...
@agc - I feel that I gave enough examples, but if you would like an example for any specific bullet point, please ping me here or in chat and I'll try to add one that won't immeduately get a bunch of complaints.
I like this answer a lot, except for the bizarre (and really unnecessary) claims in point 3. First, the idea that the Nazis were actually leftist because they have "socialist" in their name is absurd and debunked. More importantly, Left-Right self-identification is pretty irrelevant to your argument and distracts from the other strong points. To use that Mussolini quote as evidence for the Horseshoe theory, requires proving that he *truly* believed that his party was neither left or right and that it's not just propaganda. The answer would be stronger without dealing with self-identification.
@Avi DW-NOMINATE is for congress only, and congress are not good representatives of the ideology of the public, due to the polarization caused by FPTP voting method. https://www.ssrn.com/abstract=2260083 https://i.imgur.com/OH03zEL.png There are actually many simultaneous dimensions in political space, conceivably as many as there are political issues. But for elections and parties, we simplify and group some together. German parties require 4 dimensions to describe their ideologies, for instance: https://www.sciencedirect.com/science/article/pii/S0261379415000980
The Nationalist Socialist German Workers party was about as socialist as the Democratic Peoples Republic of Korea is democratic
@divibisan It's not only branding, the NSDAP had a real "leftist" wing for half of its history - that is, the first half, which is of course far less known than the second, unambiguously far-right one.
Mussolini in *"La Dottrina del Fascismo"*: "Granted that the nineteenth century was the century of Socialism, of Liberalism, and of Democracy, it does not necessarily follow that the twentieth century must also be a century of Socialism, Liberalism, and Democracy. Political doctrines pass, but humanity remains, and it may rather be expected that this will be a century of authority, **a century of the right, a century of Fascism**" : in 1932 at least, Mussolini unequivocally did consider fascism as being right-wing. https://it.wikipedia.org/wiki/La_dottrina_del_fascismo