Ender’s Ants

Just saw Ben Kingsley playing the Maori warrior Mazer Rackham in the movie Ender’s Game (2013). Kingsley nails the kiwi accent like no other non-Kiwi actor I have seen in decades or perhaps forever. Felt moved to blog a couple of unrelated notes.

Formic Queen

Depiction of the alien Formic Queen.

Military Strategy when Fighting Aliens

The scenario: you have to fight a race of aliens so ancient and unknown that you really have no real insight whatsover into their psychology. In Ender’s Game the Formics are creatures insectoid-like, similar to giant ants.  Naturally enough their psychology would be very weird and hard to understand without close up anthropological/sociological/entomological study.

The human’s training Ender and his colleagues are running massively complex computer simulations of a possible battle with the Formics.  Do you see the big flaw?

If the human’s programming the simulations do not understand the alien psychology, then the training simulations are not going to be very useful when the real war breaks out. They might get lucky. But the premise in Ender’s Game is that the Formics battle formation patterns and movements look chaotic and unpredicatable, apart from the high energy signature emanating from the Queen.

What should they do?

Why can’t script-writers get these story details right?

A Solution — To Be Random

Maybe it ruins the later plot, but the thing to do when faced with indecision about an opponents strategy is to act  randomly yourself.

What I would love to see in the movie is a scene where Ender Wiggins is taken aside and tutored on what exactly the purpose is for the simulation games. The idea should be clearly explained to Wiggins that the simulation games are only useful for training combat readiness, troop cooperation, and execution skill. It cannot teach strategy, because the games will not be acting like the real enemy, the Formics. The simulation games were design by human’s so they will not likely have many scenarios built-in that will realistically simulate fighting actual Formic armies.

So it should be explained to Ender that when facing the real battle he will need to almost forget about any simulation game strategies, and instead think and adapt in real-time. He should be told to act more unpredictably, and to not guess about Formic battle patterns based in the game simulations, but rather to do so based upon what he gleans in the moment of battle.

This is only a brief adjustment to the movie plot, it would take a one minute scene of dialogue between Ender and General Graff, but it would make the movie so much better for a geek like me to watch and enjoy and forgive all the other terrible Hollywoodisms the film commits.

In the end it turns out this way, since Ender is sufficiently unorthodox and imaginative that he ends up doing something that was unthinkable for the adult military commanders.  So I’m not moaning and bitching.  I just think it is more awesome when a script delivers high-brow intellectualization of the characters motives and thinking.  Somehow it makes a fantasy story more immersive and believable when the dialogue and plot have a lot of detailed intellectualizations and binding logic and scentific principles applied rigrously.  For the vast majority of scifi flicks one has to give up scientific realism at so many points in a plot that injecting some really hard science inwhere possible is a very cool thing to do for the thinking audiences out here.

Totally Unpredictable but Not Perfectly Random

As a general meta-strategic point, this is your real strategy when you truly have close to zero insight into your enemy’s psychology and motive.  You would want to be completely unpredictable at times when it pays o have your opponent guessing your next actions, and yet not behave hopelessly chaotic. (Although note that Ender’s Game works as a gripping story partly because Ender Wiggins does end up understanding one of the Formic civilisation’s motives, which is their desperate need to find a world with abundant water.)

So how does one achieve unpredictability without resorting to random incoherence?

This is the bit I write for all my mathematics and science students. Readers who do not know the answer can follow along.  So the idea is that you will draw up a table (or some kid of database) which presents all the possibly non-stupid strategies available.  (An example of a stupid strategy might be the one where you murder your own army, or kill your entire species to allow the aliens to win.  So you make sure none of those types of options sneak into your database! You also would probably want to purge any options that are just silly, like  the ones where the aliens are attacking and you do absolutely nothing not even try to communicate with them or try to show them you are peaceful.)

For the sake of utter simplicity let’s sketch this on a graph, suppose your database of options boils down to choosing a number from a scale, say strategies 1 to 100.  Each strategy is weighted by a frequency which is your and your war counsel’s expert opinions on the probability or likely chance of success of each strategy.

Decison options can be weighted then chosen randomly but biased towards the more favourable options.

Example tableau of strategies you can choose from unpredictably using biases in favour of the more likely successful options.

You see the point is that these strategies are not equally favoured.  So you use a random number to choose one strategy at a particular time when a decision has to be made.  But the random number selects among these strategies in a biased way so that strategy 25 above is half as likely to be chosen by your random number generator as is strategy 30.  The strategies like option 75 here are more likely to be chosen. You get the idea yes?

This tableau will of course shift dynamically as your enemy or opponents shift their strategies.

The point is that no one can predict your decision, you have the ultimate element of surprise which is that you yourself do not even know which decision you will make.  you only have the odds of the various possibilities.  So you can say, “I am more likely to adopt strategy 75 than I am to issue the decisions for strategy 92.”

This is the answer to how you can be totally unpredictable and yet smart and not foolishly random.  You use random numbers to select among possible options in a biased manner.  You bias the outcome towards the options that have higher chance of success.  But there is still a small chance you will choose a low chance of success option.

But when would your tableau ever prevent you with a low success probability option?  Surely these would be purged from the database?  Well, not always.  This would happen, for instance, if the game or battle was at a crucial stage and all the information suggested th enemy was in a stage of trying to predict your next move.  In such circumstances it can pay off to choose a low probability of success option because your enemy will be unlikely to guess this is what you will do.  This sort of option pays off only when the enemy will suffer if they do not take steps to defend themselves against your unlikely decision.

A good example of such a scenario might be the suicide bomber option, or a chess pawn option, where you sacrifice your resources to make it appear like you are giving away an advantage, which hurts your side, but f the enemy does not see your reason for such a sacrifice they will expose themselves to a more lethal attack on your next move.  In chess you might even let your opponent capture your most powerful piece, your Queen, because perhaps if they do choose to capture our Queen you can in three or four decisive moves checkmate them.  They were blind to the secret of your sacrifice.

OK this is where my students have had their lesson and other readers can rejoin.

Would Ender Really Have Tried to “Talk”?

“But OK, that’s fine at first blush, but real military strategy has to more incisive, a good commander has to be willing to be accountable and therefore cannot just roll a dice”, you might think.  That’s true of course and there are many subtleties to fighting a war, and I am totally ignorant of most of them.  You cannot hope, for example, to wage a successful war using a textbook on game theory.

[Spoiler alerts coming!  Skip this next paragraph if you like.]

Here I must confess that although I like high quality hard scifi, I chose to watch the film before reading Scott-Card’s novel.  I enjoy movies this way.  It’s horrible to sit through a movie and constantly complain about the missed adaptation opportunities and unfaithfulness of the script.  So I ended up enjoying Ender.  I must also confess I started writing this post before watching the final battle where Ender commits genocide  [spoiler alert!] while believing himself to be playing merely a final simulation.

Ender’s Game raises this wonderful philosophical issue.  Many in fact.  [spoiler alert!] One of which is, can there ever be a threat to an entire civilisation great enough to morally justify destroying preemptively an entire planet and all life on it?

My point is that the decision to try to communicate with the alien Formic’s was a noble and moral decision that arguably Scott-Card [spoiler alert!]  wants us to believe would have happened had Ender been allowed to face the final confrontation knowing it was not a game.  (Or maybe the General would have then pulled the plug on Wiggins and put someone else in command at the last second?)

But if you were in Ender Wiggin’s shoes and were lazy and just simply choosing from a bunch of preprogrammed stratgey options this diplomatic option might not ever get selected.  So in addition to trying to behave unpredictably so that your opponent cannot easily plan to defend against your moves, one thing you might always want to do is to think freely and outside the scope of what the authorities allow.

Dammit all!  [spoiler alert!] You know I cried when I found out Ender’s last game was for real.  I really internalised the movie and let the raw emotions overcome my senses.  Even if the acting in the film was a bit atrocious, one can let in and filter just the essential emotions and intellectual content.  It make a movie worht watching when you do this artistic suspension of disbelief.   I thought of the ral world horror committed by the second Bush administration in Iraq, and the lying generals and the hapless Colin Powell who seemed to be selecting decisions from a badly written textbook full of errors.  But mostly I thought of poor fcitonal Ender Wiggins and the sickening thoughts that must have ben racing through his mind.

Of all things in the movie, perhaps the most unreal was that his character actually still has the emotional strength to argue his case philosophically with General Graff and Mazer, only minutes after the horrific realisation.  How does one avoid such things?  Such traps for the unwary?

Play all your games ethically!   Even the trashy VDO games, even if they have the most awesome hires graphics.  Refuse even to begin playing the immoral ones that have no option for morality.

Try diplomacy more!  Even if it could wipe out your entire planet!

That would be my thought for the day.


*        *        *


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s