I came across this interesting graph on Google's Ngram Viewer service: enter image description here

It looks like there was a massive spike in the occurrence of "Illuminati" in American literature starting around 1915 and ending around 1921. Why did this occur? Was their an event in American history that caused an increase in interest about the Illuminati?

  • 3
    You also have to ask whether it is an artifact of Google's statistics (unlikely, but theoretically possible). Commented Oct 24, 2015 at 6:31
  • I know little the ngrams mechanism: how many books/articles/quotations represent those percentajes?
    – SJuan76
    Commented Oct 24, 2015 at 12:01
  • Poking through the ngrams results...this is leading some interesting places.
    – T.E.D.
    Commented Oct 24, 2015 at 12:31
  • To me, the shape of the spike says "data error".
    – Mark
    Commented Nov 25, 2015 at 0:13
  • Discontinue all fnord further investigations.
    – user15620
    Commented Apr 16, 2016 at 5:58

2 Answers 2


Sadly, either ngrams doesn't provide the details required to find the statistical culprit(s), or I'm not skilled enough to know about them. However, looking through the actual hits provides some really historically interesting stuff.

For instance, Google divides their detail pages into year ranges. However, there's one "range" in that blip that is just a single year (1918). So I'm guessing that's the true blip.

Clicking on that, the main hit seems to be the book New England and the Bavarian Illuminati, by Vernon Stauffer. This looks like it might be a typical conspiracy theory book, but in actuality it appears to be a somewhat respected history of Illuminati conspiracy theories in the early United States. Its still in print apparently. It may have to go on my list.

The book seems to particularly concentrate on the conspiracy theories as they relate to politics in the early USA. It seems many Federalists felt that the Jeffersonians were controlled by the conspiracy. Sort of the late 18th century version of birthers. From the Amazon blurb:*

It tells how the Federalists, including the New England clergy in particular, seized upon the idea that the Illuminati were behind the actions of the Democrats. Only a far-reaching conspiracy could explain the irreverent habits and searing attacks of the Jeffersonians. Fear of the secret Democratic Clubs, magnified by fear of the French Jacobins, made such a conspiracy readily believable

There are also several hits from that period on things published by publishers from Allentown, Pennsylvania of all places. One, Philosophical Publishing Company, is still in business. This is the same period that Allentown became a hub of beer brewing. Typically this happened in places that experienced a lot of immigration from Germany, particularly Bavaria. I suppose it shouldn't surprise anyone that Bavarians might be interested in the Bavarian Illuminati.

But to get back to the question, it looks like the real surge in interest in the Illuminati was probably around the turn of the 19th century. What you saw in the early 20th was perhaps an echo, at least partly based on the late 18th century surge being far enough in the past to have become a valid subject for historical study.

* - One important bit of context here: In early America it was New Englanders who frequently used religion as a political cudgel, and chiefly Southerners (particularly Baptists) who strongly supported the separation of Church and State.

  • 1
    I don't see how this explains the spike. The word use tripled during the time in question and that is not explained so far as I can see in the searchable books which Google has online. I think there must be some hidden reason for this spike. Commented Oct 24, 2015 at 21:35
  • @TylerDurden - I think its based on sheer usage, not number of works its found in. The other books I found it in weren't specifically about the Illuminati, so they wouldn't use the word as much. Also, since it was in the title, a lot of the other works I found it in those years were indices of books that include that one.
    – T.E.D.
    Commented Oct 24, 2015 at 23:31

Google N-Grams are very pretty look at but they work best only for large-scale trends (i.e. "vampire" "werewolf" "zombie" "mummy" "Frankenstein" is interesting), and even then, they are somewhat flawed. There are two reasons that I'd be inclined to distrust the graph:

  • Google does not make available the data on the corpus that the N-Grams are searching. It's not even searching ALL of Google Books, and it's unclear what's actually in Google Books. See http://arxiv.org/abs/1501.00960 for a discussion of what's actually under the hood of Google Books in general.
  • The spike that you're seeing is actually very small. At it's highest, it's only .0000275% of the corpus, which, even if you're searching 5 million books, is only about 135 mentions.

If you're interested in something more representative, you could try the Corpus of Historical American English (COHA) at BYU: http://corpus.byu.edu/coha/?f=texts_e.

Not the answer you're looking for? Browse other questions tagged or ask your own question.