"Friday" Seminar: Great Expectations

Image via Wikipedia

Dear Readers,

  I am sorry about the glitch that leads to this post's late release.  I place the blame squarely on bad software created by the Augury Department.  I do not think, as their Chairman does, that the graduate student responsible should be sent back to a time-normal existence, but I imagine that Provost Notaras will talk some sense into her.

  At any rate, it is now your Sunday, and we have managed to fix the problem, so here is the Seminar that I hoped to post on Friday.  It is, as per usual, drawn from my lecture materials for CHRN/AUG 100, my introductory Anachronism class here at the University.

  Today I want to talk to you about the idea of likely and unlikely events, and how we can use them to get at the "big idea" of my last few seminars.  That idea surrounds the various "tricks" that we as Anachronists must use to put together the raw text that we get from Augury.  Remember that because data comes from distinct choice paths, it can be nigh impossible to correlate two datasets and say that they come from the same time stream, or that one follows the other temporally.

  Unless you use these tricks.  Last time I talked about how to use linguistics.  This time I'm going to talk about probability.

  First I need to define my terms.  What is an unlikely event?  Likelihoods are a little strange when you are in the study of all possibilities.  Essentially everything happens somewhere or sometime.  But when you compare a wide variety of time streams, you learn that there are events in some times and places that are very unlikely and found rarely in other streams.  As an example, consider the death of Atilla the Hun in your own time stream.  He died while passed out drunk, drowned in his own blood, due to a nosebleed.  Poetic, certainly, but not a very likely thing.  If you changed that one nosebleed, Atilla might have created a dynasty to rule all of Europe.  In many time streams, he did.

  For a likely event, let's consider the fall of the Empire that spawned our lovely University.  The Eastern Roman ("Byzantine") Empire fell because of a series of events that took place over the course of four to six centuries.  No single, random event killed our Empire in your time.  It took many very disconnected but combinatorial events to bring it down.    You can't change one single thing and erase the fall, and so the event is called likely.

  Now, how does this help us?  One way to learn what time stream a piece of information is from is to look at characteristic events in the history of that time stream that are alluded to in the data.  If an event is likely, then it happens in many time streams and it is not very useful to differentiate it from other time streams.  Consider the request, "Please find the man with the head and bring him to me."  Not very specific, and so I doubt you will be able to find the right man.  However, likely events can be conspicuous in their absence.  If I find a source dated 1920 and it is discussing current foreign policy from Constantinople, then I can say that source comes from one of a very small set of time streams.  The relevant example would be the request, "Please find the man without a head and bring him to me."  That man would be hard to miss.

  Similarly, unlikely events are useful when they do happen.  Atilla drowns in his own blood in a few time streams, but in many more he dies in other ways.  In fact, his death is a great marker because it can happen in so many different places and different ways!  It's an unusual feature.  Sort of like the command, "I need you to find me the man with the monkey tail."  Of course, there are a great many time streams where Atilla died peacefully in his sleep after a long life.  Those time streams ask you to find the man without a tail, which is again not helpful.  So the unlikely events are useful when they are mentioned, and are not so useful in their absence.

  As you can see, this gives us another way to group sources and connect them.  That kind of painstaking work launches many an Anachronist's career during graduate school, and I have to thank my students for what they do to provide me with connected narratives.  Their work lets me present you with the coherent work that you see here.

  Always,

  Dr. John Skylar
  Chairman
  Department of Anachronism
  University of Constantinople