What, exactly, is a conspiracy theory? How do you define the term?

I have to admit, I had never really given these questions much thought before. Conspiracy theories seem to be almost self-evident, usually through the context in which they are presented. If you read something written by a serious believer, it's pretty obvious. If you see conspiracy theorists portrayed in a movie or television show, once again it's usually obvious -- even when they're not literally wearing tin-foil hats. But when studying the history of conspiracy theorizing, coming up with a clear definition of the term is an absolutely necessity, since it will dictate which data is included and which is omitted in the study.

When reading the new book American Conspiracy Theories (Oxford University Press -- also available at Amazon.com) and while interviewing its two authors, I found myself returning again and again to my own personal definition of what constitutes a conspiracy theory. If any given conspiracy theory is later proven beyond a shadow of a doubt to be true, would you still call it a conspiracy theory -- either before or after the actual conspiracy has been shown to exist?

The authors of American Conspiracy Theories are Joseph E. Uscinski and Joseph M. Parent, both political science professors at the University of Miami. They devote the initial chapters of the book to explaining their own definition of what constitutes a conspiracy theory for the purposes of their study, and then present a historical overview of American conspiracy theorizing for roughly the past century. The three sources they used to compile their data are interesting: they fielded a survey in an attempt to understand conspiratorial thinking (and who is affected by it), they studied a very recent slice of conspiracy theories posted on the internet to see the current state of such theorizing, and they conducted an exhaustive search of over 120 years of printed letters to the editor of the New York Times. This last source is particularly brilliant, because it reveals the extent of conspiracy theories all the way back to the 1890s (a decade which was, incidentally, a hotbed of conspiratorial theorizing).

The book is not a history of American conspiracy theories, however. It doesn't purport to be an encyclopedia of the past 12 decades, listing which theories appeared when. Instead it is an academic attempt to seek out the relative prevalence of such conspiracy theories in American life, and to explain which types of people are drawn to such ideas. The book does present many fascinating examples from history and the writing itself is amusing and lighthearted at times (it would probably be impossible to write about conspiracy theories without occasionally noting how ridiculous some of them are, especially when seen through a historical lens).

Of course, when defining conspiracy theories, there is always an element of bias, whether political or personal. The authors do footnote their own possible biases as: "This study was written by two unmarried, 30-something, overeducated white males. Uscinski advises the local College Republican and Libertarian groups and Parent interned for Democratic politicians, but neither identifies with either major party or votes consistently for one." Writing about conspiracy theories from the 1890s or 1930s is relatively easy and safe to do, but the closer the timeline gets to the present, the more delicate it becomes.

When conducting the following interview (via email) with the two authors, I had my own set of biases, I fully admit. While reading the book and taking notes in preparation for the interview, I focused on certain subjects while a reviewer with a different political viewpoint might have focused on entirely different items.

But I did enjoy the book, as I am somewhat of a conspiracy theory dilettante -- I enjoy a good conspiracy theory (from the right, left, or outer space) mostly for the entertainment value it brings. Thinking "what if..." and suspending disbelief is enjoyable in the same way watching a scary movie is -- gets the blood pumping and the adrenaline flowing, and allows for a certain amount of escapism from reality.

Overall, though, the deepest thing I took away from reading American Conspiracy Theories was how hard it is to actually define the term. Especially while conducting the interview, I was forced to re-examine and redefine what I would personally qualify as a conspiracy theory. In other words, the book made me think -- especially about my own biases and perceptions. From me, that's a pretty strong recommendation for any book. I would definitely recommend the book to anyone interested not so much in conspiracy theories as entertainment, but in why people believe in them and why the fact of who does so doesn't change much over time -- or in any sort of partisan fashion.

[Note: The following interview has been edited for length. Preceding each answer is the name(s) of who is actually answering the question.]

You rely on three sets of data in your book: letters to the New York Times over approximately the past century, a one-year slice of the internet from more modern times, and a recent opinion survey about people's attitudes towards conspiracy theories. Given the extensive time period for the letters to the editor, why did you so severely limit the time period for the internet data? One year doesn't seem like enough time to accurately measure modern public attitudes. It seems that, at a minimum, you should have chosen one year during a Democrat's presidency and one year during a Republican's (Obama and George W. Bush, obviously, since the internet was in its infancy in Clinton's time and didn't really previously exist), especially considering the conclusions you draw about partisanship later in the book.

[USCINSKI:] The internet data weren't used to measure public opinion in the way that the survey data or letters to the editor were; they were intended to measure the online information environment. A common claim about conspiracy theories is that the internet is an echo chamber that amplifies them. To find out, we compared a measure of internet usage over time to our letters to the editor data and show that the introduction of the internet does not affect the level of conspiracy talk in the letters. If the internet really drove conspiracy talk, we would have expected lots more letters to discuss conspiracy theories, but we don't find that. The second way we tested this claim is with the internet news data. Using Google's news alert function, we gathered all the stories each day over the course of a year that contained the term "conspiracy theory." We then coded these stories to see how conspiracy theories were treated. About 70% of the stories –from all sorts of news outlets and blogs -- treated conspiracy theories very negatively. This is strong evidence that the internet does not really drive conspiratorial thinking -- instead, the internet, and the media more generally, seem rather anti-conspiracy theory.

I've kept the Google Alert running, so in a few years (assuming a Republican wins in 2016) I will have data as you suggest. Unfortunately, I didn't start the Google Alert until 2012, and you can't really go back in time with it.

On a separate note, one thing that we do as a robustness check to our letters to the editor data is to look at Google Trends (this is not in the book but discussed here). Google Trends allowed us to examine people's internet search patterns over time since 2004. What we found was that people searched the internet for conspiracy theories that implicated Republicans during the George W. Bush administration, and for conspiracy theories that implicated Democrats during the Obama administration. We also found that geography was associated with internet searches: blue states searched for conspiracy theories that implicated Republicans while red states searched for conspiracy theories that implicated Democrats.

While the letters to the editor from the New York Times is indeed an inspired method of measuring public opinion about contemporaneous conspiracy theories, your book never really addresses the human element of the editors of such letters. To put this another way, doesn't the possibility exist that any given conspiracy theorist's letter was printed in the Times not because the theory itself was particularly valid or widespread among the public at large, but because of the entertainment value of the letter? If the editor had a couple of column inches to fill and decided to use a highly amusing (and well-written) letter about aliens from Pluto taking over the government, does that really say anything valid about the public's attitude towards such an idea? Or could it just be that the editor wanted to give the readers a good belly laugh?

[PARENT and USCINSKI:] We sure hope that the editors (and readers) of the New York Times have a sense of humor, and we don't doubt that sometimes that makes its way into published letters. Still, senses of humor -- and the need for laughs -- are constants, but the level and types of conspiracy letters published varies over time. If this was more about entertainment than reflecting public preferences, conspiracy theories would appear regularly or randomly. They don't. Instead, the letters are published in predictable patterns that follow changes in political power. Sadly, almost all the conspiracy theory letters we came across were not that funny and the ones that were funny were not conspiratorial.

That's an interesting distinction. Were obviously "funny" -- that is, obviously printed for amusement value -- letters not counted in your data? How did you handle such letters? Would the aliens-from-Pluto letter have qualified?

[USCINSKI:] We didn't remove any letters because we thought they were funny. If they fit our definition, they were included. We did find some letters that we thought were bizarre -- letters about Bigfoot and letters about leaders from other planets. But those letters indicated paranormal beliefs, and not conspiratorial beliefs. Therefore they were not included.

I had never considered it much before reading your book, but it seems to me one of the biggest questions in studying the subject is how you determine what exactly to call a "conspiracy theory" and what to ignore. In particular: if a conspiracy theory later is proven to be true (or even mostly true), then how did you classify it? In the following list, which (if any) would you have counted as conspiracy theory letters?

In each case -- at the time written and published -- these were nothing more than wide-eyed conspiracy theories. No proof backed them up. They were completely unsubstantiated and they all sounded impossible to believe, but they were all later proven to have more truth behind them than conspiracy theory. So how did you handle such letters, when determining whether they were conspiracy theory letters worth counting in your data, or not? Was later vindication of any conspiracy theory taken into account?

[PARENT and USCINSKI:] No. The line we drew on coding conspiracy theories was a pretty standard one: whether it was an explanation of a past, present, or future event that relied primarily on a small group, working in secret, against the common good, which ran counter to the consensus of impartial experts. All conspiracy theories could be true, and sometimes it turns out that they are.

But one of things that we avoid doing in the work is assessing which conspiracy theories are true and which are false. For the purposes of the book, the truth or falsity of particular conspiracy theories is not what's most important. What we care about is how people are engaging in conspiratorial thought and actions, and the underlying reasons for doing so. Our survey data suggests that belief in conspiracy theories is driven in large part by predispositions -- specifically a predisposition to view the world through a conspiratorial lens -- but political, cultural, religious and other predispositions are also important. Two people could be given the same information about a conspiracy theory, yet come to very different conclusions about it.

Suppose two Republicans, one with strong predispositions towards conspiratorial thinking and the other without strong conspiratorial predispositions, are given information about Obama's faked birth certificate. The Republican who views the world through thick conspiratorial lenses will likely believe the evidence and become a "Birther." The Republican without strong conspiratorial predispositions will likely not buy into that same evidence. Just the same, Democrats, because of their partisan affiliation, will not view evidence purporting to show that Obama faked his birth certificate as credible, and they will therefore not subscribe to "Birther" theories. And for the Democrat, it doesn't matter how strong the Democrat's conspiratorial predispositions are; she would have little incentive to see a conspiracy. But in any case, it isn't so much the evidence that drives individual beliefs, it's predispositions.

Or imagine a person who believes in countless conspiracy theories. Some will eventually turn out to be at least partially true. Does this mean that the person is not a conspiracy theorist, but just a correct thinker? Probably not -- broken clocks are right twice a day. What matters is not so much that the person was right here or there, but how they got there. If predispositions marry people to conclusion with little consideration of conventional logic and evidence, then being right or wrong is not what's at issue.

In Chapter 6 of your book, you attempt to measure conspiracy theories about foreign threats, and you break down your total 114-year timeline (1897-2010) into two groups: when America is at war (cold or hot) with great powers, and when America is not. However, your inclusion of the Cold War seems to unbalance the data, due to its length. From 1897 to 1938, America was at war for six years and not at war for the other 36 years. However, when you add World War II to the Cold War, this means America was at war (by your definition) from 1939 to 1991 -- a period of 53 uninterrupted years. Doesn't this make it a little hard to draw any definite conclusions about foreign conspiracy theories?

[PARENT and USCINSKI:] That's an interesting concern. The problem is we're hostages to theory and evidence. There are good theoretical reasons for thinking there was a correlation there and empirically that turned out to be the case. On the theory side, hot wars and cold wars present real danger and threat. The Cold War could have ended in mass annihilation, just as WWII could have ended with a Nazi victory. These types of wars are clearly different from our incursions into Korea, Vietnam, Somalia, Grenada, Panama, Iraq, Afghanistan, etc. These countries posed minimal threat. No one was worried that any of these countries could rival American power or that possession of these countries would seriously shift the distribution of power.

Further, the granular details support the theory. At the peak of Cold War tension, many of the letters were about foreign threat. As tensions eased, letters on foreign threat abated. Since the end of the Cold War, there's a lot less concern about foreign conspirators. In terms of Hot and Cold Wars taking up half of our timeframe, that's actually a strength in the analysis. It gives us enough observations on which to draw meaningful conclusions.

Again and again in your modern examples, you present theories from the right which are largely unverifiable or unprovable, and then present examples from the left -- which are indeed verified by concrete facts -- to somehow balance them. Perhaps this is my own bias talking, but this seems an awful lot like what the left accuses some journalists of -- presenting a "false equivalence" between the two sides, so as to not make the Republicans seem so bad.

[USCINSKI:] So the left thinks the media does not make the right look bad enough, and the right thinks the media does not make the left look bad enough. Why should this shock anyone? Both sides think the other is "bad," and they both want the truth to be known. The reactions to our book follow a similar pattern: people on both sides are upset that we don't make the other side sound like a bunch of conspiracy nuts. Unfortunately for everyone, we have to side with the best available data, which shows Republicans and Democrats as equally conspiratorial.

[PARENT:] We all have our biases, which is why it's so important to subject them to critical scrutiny. What evidence would change my views? We started with the least controversial definition of conspiracy theory and used the best methods we could to transparently collect data. When our results came back -- and they fit with Dan Kahan's at Yale, and Larry Bartels's at Vanderbilt, which find that partisans on both sides about equally good at assessing inconvenient facts -- they angered people on all sides. Paul Krugman dismissed us as "crazy centrists". But the attacks were basically anecdotal. Many people miss the forest because they prefer to see the cherry-picked tree.

There are good reasons why, of course. Righteousness is comforting and fortifying; it's also blinding and intolerant. There's an element of religious warfare to this. Note how such accusations galvanize one side while delegitimizing the other. Roughly half the population is fools or knaves. Republican elites are awesome snake charmers while Democrats are the People of Sweet Reasonableness (or vice versa.) The fate of the country is in the balance. These are politically expedient arguments to make, but they are not scientifically supported. With the current evidence, it would take some creative accounting to show that one party or the other has been systematically less susceptible to facts over long periods of time.

Since you have divergent political beliefs yourselves, I have to ask: did each of you choose the examples from "your own side" or did you pick the examples for the opposition?

[PARENT:] No, that never came up. Oddly enough though, we did disagree about which examples to use from the natural sciences once.

[USCINSKI:] Having worked on the topic for five years, I have observed a few prevailing constants.

(1) The quality of evidence is subjective. This applies to conspiracy theories as well as to other ideas. Different people, given the same evidence, can come to very different conclusions. This is because individual's predispositions drive how they interpret evidence. People ignore evidence that contradicts their beliefs, and focus instead on evidence that supports their beliefs.

(2) Partisans think other side is conspiring. Over and over again, no matter how we slice it, data always shows that Democrats think Republicans are up to no good, and Republicans think Democrats are up to no good. People's political competitors tend to be the mustache-twirling string-pullers.

(3) Both sides agree that conspiracy theories exist on their side, but that they are confined to the fringe. In accusing the other side of being a bunch of conspiracy kooks, each side must admit that there are at least some conspiracy theories propagating on their own side. It would be hard for Republicans not to take responsibility for the majority of Birther beliefs, and hard for Democrats to not take responsibility for the majority of Truther beliefs. But both sides think those beliefs are confined to a small fringe... (they are not confined to a small fringe).

(4) Both sides of the political divide think the other side is a bunch of conspiracy kooks. Each side thinks their own side is tied to facts and truth, while the other side is a bunch of science-rejecting conspiracy mongers. Republicans think this of Democrats, and Democrats think this of Republicans. Each side believes they came to their conclusions based on sound reasoning and unbiased evidence, but that the other side came to their conclusions through ignorance, manipulation, or willful denial. Our research however is unequivocal on this point -- it shows that Republicans and Democrats are equally susceptible to conspiratorial thinking, and hold conspiratorial beliefs about equally. Neither side is kookier than the other -- and we are sorry if this offends everyone's sensibilities.

(5) Everyone believes their conspiracy theories are conspiracy fact, and not just conspiracy theory. People don't like to admit that they believe in conspiracy theories; partially because there is a negative connotation to it, but also because people view their conspiracy theories as conspiracy fact. Why would someone believe in a conspiracy theory if they did not believe it was true? Instead, it's the other guy's conspiracy theories that are not true.

And finally, I have to ask: Do the two of you have your own favorite conspiracy theories? Most outlandish, perhaps? Or most creative? After spending so much time digging through so many conspiracy theories, you must have a few that were personally memorable when you came across them for the first time.

[PARENT:] Hell, yes. My first article argued that Machiavelli's Prince includes a coded conspiracy theory for Lorenzo de Medici to butcher the Pope and the College of Cardinals. For some reason, Dan Brown hasn't turned that one into a bestseller yet. There are some great ones that came out of our dataset, and we incorporated most of them in the book, but what really stands out to me is how unimaginative and repetitive most conspiracy theories are. Not that I could do any better, but so many conspiracy theories have a drab, disenchanting "paint by number" feel to them. Life should really be weirder.

[USCINSKI:] That the C.I.A. created lesbianism. If it's true, why would they have done that? What was the goal of such a scheme? It makes the mind reel.

[PARENT:] OK, I stand corrected: that one is amply weird. But no self-respecting conspirator would stop with lesbianism. Whoever controls what people desire controls the world. Picture the secret government facility in rural Virginia that houses the C.I.A. Fetish Research Lab, and the intense professionals laboring there around the clock. I'm not sure that makes the mind reel so much as it makes a great plot for Dan Brown's next book....

-- Chris Weigant

Cross-posted at The Huffington Post

Follow Chris on Twitter: @ChrisWeigant