Dark Messiah of Might and Magic is full of spiders: monstrous, poison spiders, spiders that leap at your face, spiders that build city-sized nests in which they wrap up and devour their human prey.

These spiders terrified me. Despite knowing they weren’t really there, I felt uneasy when they were around and would lose my cool when they advanced on me. I wasted a lot of potions in my panicky desperation.

It’s just a game

Dark Messiah does a lot to encourage a sensual sort of immersion. Its settings are rich and colourful, blows from swords feel weighty, and there’s a nice meaty thump when you kick an orc in the chest—often off a cliff. (Their yells are satisfying too.)

The game also has many unreal elements. Equipment is represented as icons in a grid and has explicit attributes like +3 Damage. No matter how well you can sneak, if your player character hasn’t upgraded his Stealth skill, guards will spot him long before he can get behind them. And why are guards so often hanging out by racks of spikes and cliff edges anyway?

These arbitrary and artificial things work against the sense of being there supported by its clashing swords and pretty vistas. Thinking about skill points and armor classes draws the player out of the dungeon and into her computer chair, managing stats instead of impaling goblins. Well, you might think so.

It’s real to me, dammit

Dark Messiah‘s unreal elements (stats, inventory swapping) didn’t hold me back from dreading the spiders, though I was never unaware of them. I knew I was playing a game—I was constantly reminded by the interface, the click of my mouse, the cheesiness of Xana’s dialogue—but my reaction to the spiders was nonetheless a gut-level thrill.

“We really can’t differentiate between real and unreal images,” notes Chuck Klosterman.¹ “We can describe the difference, but we can’t manage it.”²

Humans have existed for 130 000 years. The Great Train Robbery was made in 1903. For roughly 129 900 years, any moving image a human saw was actually real. It was there, right in front of you. If a man in 1850 saw a train chugging toward his face, it was actually a train. For 129 900 years, we were conditioned to understand that seeing something in motion had a specific meaning. But that understanding no longer exists; today, we constantly “see things” that aren’t actually there.

I am struck by the idea that for all my understanding and all the game’s self-evident artificiality, I was still affected by the spiders. Once they were dead, I would quickly become conscious of the game, the system. Using potions to get rid of the annoying and overlong poison effect the spiders dealt me was never so natural that it felt part of the same reality in which giant spiders were attacking me.

It’s as if the sense of being there and the knowing of toying in a system were like a figure–ground illusion: I could see both, but never at the same time.

Maybe, or maybe I need to grow some balls and do something about my arachnophobia.

Retro Game Challenge has been getting some attention in year-end roundups. In a recent episode of The Brainy Gamer Podcast, Retro Game Challenge was praised for being true to the NES-era games it recreates without being as frustrating. While RGC includes a number of familiar retro games, it does not seem to be as punishing to play as the original games or their remakes (Mega Man 9, I’m looking at you).

RGC is not just a collection of retro homages: the player plays as a player that must complete a number of “betcha can’t” challenges across a bunch of old-school games. The design of this meta-game works against much of the frustration of playing otherwise difficult and unforgiving games. It creates a variety of things to do and game types to play, and keeps play sessions short and sweet, not long or repetitive. It also gives the player the ability to help herself out of uncertain, frustrating challenges.

Variety keeps things fresh

There are a variety of challenges in the game. Game Master Arino dares the player to get 250,000 points in Star Prince, or finish first in the third race of Rally King. These challenges are explicit—the player is told what to do and, sometimes, how to do it—and there is no doubt when the player has completed them.

The challenges encourage the player to experiment or play under certain constraints,¹ something the games do not do themselves. While Rally King only asks the player to come in the top five in a race, RGC asks her to perform a number of boosts, to win while avoiding damage, or to reach a high score—without cheating.

There is also a variety of games to be played: Robot Ninja Haggle Man is an action platformer, Rally King a racing game, Guadia Quest a JRPG, etc. Though some games are repeated (Haggle Man 2, Rally King SP), the shifts in mechanics and genre keep RGC from feeling like a grind.

Play is brief

Most of the challenges can be completed in minutes. Each game has five challenges. RGC also allows the player to quickly restart games. This makes it easy for players to retry challenges without having to go through many game menus.

As Adam Saltsman has noted, long play times can discourage a player because she finds it too daunting or the rewards too spread out. RGC does not ask the player to invest hours of her time into completing its challenges, instead asking the player for a few minutes and rewarding her often.²

Help is at hand

In the same podcast, it’s mentioned that Shigeru Miyamoto meant The Legend of Zelda to be played by many players who share tips and discoveries with one another. Sharing was important, often necessary, to play through those older games: without friends or game magazines, I would never have beaten Super Mario Bros. or put up with Battletoads.

The experience of sharing expertise is recreated in RGC. All of RGC‘s games have manuals. The player’s in-game buddy drops hints after failed attempts. He also has a shelf of GameFan magazines that the player can browse through to find ways of overcoming challenges and improving her play.

These things help the player help herself. Through the amusing, player-driven help RGC keeps the player from becoming lost and frustrated, without putting her through tutorials or adding anachronistic button prompts to the games.

Dead Space‘s in-game interface has gotten a lot of attention. No doubt, the holographic displays look slick, and the way they’re presented in the game world does a lot to make them feel like a part of it, less artificial.¹ But it’s the fact that the interface does not interrupt the game that makes it worth mention. The fear and vulnerability central to Dead Space isn’t ever trumped by the needs of the UI—and critics noticed. Rarely has the unobtrusiveness of an interface been so acclaimed.

Not pausing the game while the player fiddles with her inventory is not new (it’s a design choice used similarly before—System Shock 2—and since—Demon’s Souls), but Dead Space pulls it off particularly well. Thoughtful design compensates, in part, for the lack of pauses and the limited time the player has to keep track of her stuff. By sorting, organizing, and keeping the inventory simple, the interface reduces the amount of attention the player needs to spend on it, leaving her to explore, shoot off limbs, and be scared.

See-through and sorted to be useful

In Dead Space, the player is often assaulted or surprised by monsters. The in-game interface takes up three quarters of the screen, but is semi-transparent so the player can still see advancing zombies while she heals, reads, or orients herself.

Once the zombies hit, the player will need to heal herself or refill her air tanks, but may not have the time or ability to run away from a fight to a safe place. There are buttons that allow certain items to be used without going through the inventory (using small medkits or reloading weapons), but sometimes the player needs to use an item in her inventory quickly, in the action.

To help her out, items in the inventory are sorted by their usefulness in frantic play situations: medkits are first, then air canisters, then stasis packs (fuel for the slow-mo ability), then ammunition. Within each type, items are sorted by size: larger medkits come first. When the player is running low on HP or oxygen, the items to alleviate her distress are the ones the fewest button presses away. She can restore as much HP as possible with as few presses as possible.

Small, separated, and simple

Even when not under pressure, the player’s tasks are not needlessly complicated. The interface remains simple and interruptible, not ruining or allowing escape from any surprises.

Items occupy a single inventory slot, no matter their size in the world. There is never a need to rearrange things in order to fit in another item—no inventory Tetris. The player either has space or she doesn’t.

Fortunately for her, things such as weapons, power nodes, or quest-specific items are stored and presented separately from cheaper, fungible things like medkits. This ensures she has space for key items, reduces the chance that she will discard expensive weapons accidentally, and keeps them out of the way of the stuff she is likely to use or drop from the inventory.²

Without the need to move items around, and without needing to take much care to avoid messing things up, the inventory interface is pared down to the bare necessities. The player can, and need, only ever do two things with an item: use it or drop it. (When at a store, she can buy a new item; when at storage, move or sell one she already has.)

The holographic UI in Dead Space makes things fast and easy, intruding as little as possible on the game’s intense atmosphere. It looks cool, fits the fiction, and it supports the player’s tasks and the aesthetic goals of the game: mainly, being scary.

Of the many key activities in Mass Effect—shooting, driving, bumpin’ uglies—talking to others was the most satisfying. Though not something with much marketing appeal, dialogue is one of Mass Effect‘s most effective game mechanics. It is easy to use, produces natural-sounding dialogue, and adds a sense of discovery to conversations. It was what I enjoyed most, and, despite a few faults, something I would like to see in other games.

The dialogue wheel is easy to use

The dialogue wheel is an efficient means of presenting and selecting dialogue options. It is a variation of a pie menu: options are arranged in equal distance around the player’s cursor. Any one option is as close to the cursor as any other; the player need only pick a direction to highlight an option. This maps options to a gamepad’s analog stick directly: the direction in which the player takes a conversation is the one she takes with her thumbs on the stick.

In traditional list-based menus, some options require the player to move the cursor farther than others. Lists also place options along the same dimension, which makes it more likely that the player will overshoot the option she wants, selecting the wrong one or spending time correcting herself.

Conversations flow

Mass Effect‘s designers also made it simple for the player to decide on an option. Firstly, the options are short and quick to read. Secondly, the position of dialogue options is meaningful: options along the right half of the wheel advance the conversation, along the left, they are more exploratory; nice-guy, neutral, and aggressive options are along top, middle, and bottom respectively. The player need not read all the options, only the options relevant to her. This helps conversations to move at a natural pace because the player does not spend much time reading and deciding.

On top of that, the dialogue wheel appears a moment before the final line of dialogue is spoken. The player can begin deciding among dialogue options while the current piece of dialogue is ending, making it less likely that there will be dead time between utterances. The dialogue seems to be written in a way such that information relevant to the player’s objectives comes early on; what she hears when the wheel appears is not important, so she can be a little distracted by the new dialogue options.

The combination of ease of use, well-placed and well-timed presentation of options, and good writing make conversations seem much more natural than they do in other games while reducing the burden of reading and picking options.

Dialogue is about discovery

Natural dialogue that is easy to participate in is quite an achievement, but what I found fun about it was the sense of discovery. The terse dialogue options in Mass Effect are quick to read, but only suggestive. It isn’t until the player selects one that she can know exactly what it was.

For the most part, dialogue in other adventure games can only be discovered by the player if it is spoken by non-player characters: the player knows what her character is going to say because she picked the lines already. In effect, her half of the dialogue is repetition.

There some exceptions: Sam & Max Hit the Road and Strong Bad’s Cool Game for Attractive People use icons to represent dialogue options. These, as with the short options in Mass Effect, suggest a response instead of showing it outright. When the player selects one, she has some idea of what her character will say, but not exactly what.

Not knowing exactly what will be said adds freshness to the player character’s half of the dialogue, and creates a sense of exploration.

Unfortunately, this method risks betraying the player’s intentions: ambiguous or poorly written dialogue options can cause the player character to say something the player didn’t want. The position of dialogue options in Mass Effect adds information that can help a player know whether an option is along the Paragon or Renegade path, but that does not necessarily map to the character’s tone or discretion. Another problem occurs when options do not seem different enough from one another, making decisions more difficult or feel as if the writers and designers are forcing the player to follow a particular path. Such problems are less likely to occur when dialogue options are presented verbatim.

The Megahunks & Superbabes fiasco

About a year ago, I prepared a talk on the use of analogies to understand complex systems to give as a part of the Un-Distinguished Lecture Series (better known as UDLS). It’s a topic I find interesting, but I worried that it would be boring to a bunch of grad students dropping in on Friday before going out to drink.

In order to spice it up a bit and, yes, try to secure a larger audience, I announced a talk on a different topic: sexy men and women. Before giving the presentation on analogies and complexity, I would give a joke presentation full of garish slides, inane talk, and unfortunate typos. It would lighten the mood, I thought. The audience and I would laugh at the amateur quality and vapidity of it all. I put together a few slides, and mailed out the following abstract:

there are lots of people that are super hot for men and for women but do we know which is hotter the man or the woman? im going to show a lot of pictures of sexy people and we can decide which are the most sexy or the two. SAFE FOR KIDS!!! (your sick for thinking i’d make a pornshow) ;P c u there

The trick worked: somewhere between twenty and thirty people showed up in time for the talk (as punctual as CS students are when there’s no free food), a good turnout. I did my best to live up to my stupid slides, speaking with the judgemental attitude of TMZ in the style of JEFF K. The three slides in my joke presentation on megahunks and superbabes follow.

The audience laughed and called out jokes. A few pointed out the incorrect use of “whose,” that I had stretched some of the images, and tried to guess who was with Jeff Goldblum in that photo.

I followed these slides with a blank one, over which I said something along the lines of “seriously though, here’s what I’m here to talk about” and then got into the real presentation: Thoughts on the Use of Analogies in Understanding and Solving Complex Problems.

That was when I lost the audience. Talking though the first few slides, I noticed that they’d become much quieter. Emt, interrupting my explanation of how colour is to spectrum as tone is to scale, asked “Analogies? Are you serious?”

“Uh,” I hesitated. “Yeah.” I had hoped to pull the rug out from under them, but I had sucker punched them instead.

I got through it, but I’ve not lived down my mistake. Since, other UDLS lecturers have joked about it by showing slides on unrelated topics with long titles in the middle of their presentations, threatening to “pull a Lucas.”

I’ve thought about that bungled joke a while. My joke presentation flopped, but why? I’ve tried to diagnose what went wrong, and find ways to avoid making the same mistakes. What follows are a few of the things I’ve come up with.

They’re not hard and fast rules, nor are they applicable in every case. Some should be obvious. Nonetheless, I do think they can be useful, and not only to fake out your audience with cheesecake.

Get your rewards right

I went from ogling Megan Fox to a remarkably un-sexy brain teaser from the GRE test prep. I feigned a fun, juicy presentation but delivered a dry, thoughtful one that had nothing to do with hot bods. It was a mistake to do so. It is better to risk scaring the audience away at the beginning, then rewarding them for their patience, than to tease them and not deliver.

Perhaps I should have started with a dull presentation, later switching to something more engaging. I could have made drab and heavy slides, full of complicated diagrams of the geometry of beauty, then moved on to more colourful things, like the photos of hurricanes and crowds in the slides on complexity.

Take advantage of their good will

Thing is, it’s okay to start juicy and then dry up—a little. The slides could have hooked them, softened them up for some meaningful content. As it was, the fake presentation grabbed everyone’s interest, and then the second one did nothing with it. There was no connection to be made from sexy people to complexity. It is important to take advantage of the good will of the audience.

I could have realized hunks and babes would be more exciting than analogies and complex systems, and kept those slides, used them to draw the attention of the audience before dropping science on beauty and sexual attraction. After I had everyone’s eyes and ears, I would ask “Why do we find this supermodel so attractive?”¹

Be clear before being clever

I thought the goofy, poorly made slides would tell them that the presentation was a joke. I mean, the images are warped, the spelling off, and it’s in Comic Sans, fer chrissake! They were meant to make the audience sceptical of the presentation.

They weren’t clear. Not everyone takes typos as cues to become critical and wary, as I do. Instead, the mistakes and imperfections signalled to the audience that they could kid and relax. The real joke—that the whole presentation was fake—was hidden. It is better to err on the side of readability rather than of subtlety.

The joke might have been clearer had I used something more esoteric, less appealing, and been more forceful and bizarre. I could have made the topic ridiculous, presenting close-up images of insect larvae and asking the audience which they thought was most attractive. “Which of these slimy young things would you like to kiss?”

Be sure they’re in on it

The audience did not know it was a parody, and I did not do enough to help them see it as one. I had failed to get the audience on my side. They had fun with “Megahunks & Superbabes” itself, not at its expense.

I assumed they would come in expecting to see a serious presentation, but sent out a silly abstract and started with funny slides. I failed to see that the audience did not, and could not, know that I intended to talk about analogy. I’m the only one that can read my mind. Presentations should be considered from the perspective of the audience, divorced from the experience of the presenter.²

I’m certainly not the only one that’s failed to get his audience in on the joke. Ghandi made the same mistake. He describes it in his autobiography:

When my turn for speaking came, I stood up to make a speech. I had with great care though out one which would consist of a very few sentences. But I could not proceed beyond the first sentence. I had read that of Addison that he began his maiden speech in the House of Commons, repeating “I conceive” three times, and when he could proceed no further, a wag stood up and said, “The gentleman conceived thrice but brought forth nothing.” I had thought of making a humourous speech taking this anecdote as the text. I therefore began with it and stuck there. My memory entirely failed me and in attempting a humourous speech I made myself ridiculous. “I thank you, gentlemen, for having kindly responded to my invitation,” I said abruptly, and sat down.

Ghandi and I should have known better. Not only could we have communicated our intent more clearly (Ghandi by impersonating Addison), but we could have taken a moment to think about the audience. Would they know about Addison’s stumble? Would it be funny to them?

Be confident

All things considered, I would have been better off without “Megahunks & Superbabes”. I was told later by a by a few audience members that the presentation about analogies and complex problems was pretty good, despite starting off on the wrong foot. Instead of resorting to cheesy humour and sex appeal, I should have been confident enough in my presentation to go out and give it as it was.

The large radio tower in the RKO logo which appears at the very beginning of King Kong is a curious thing. It is a piece of an old medium, radio, in a new one, film. The logo seems awkward, a sign of a time in which movies were “moving pictures,” only beginning to gain the legitimacy of radio and print, to develop a vocabulary.

There are a number of elements of new media that are left over from their earlier forms: the prompt to “insert coin(s)” in home console versions of arcade games, electronic documents broken into distinct pages, to be stored in file folders on a computer desktop.

You might say these are analogous to vestigial traits.¹ They are remainders of past forms, still present but made useless by gradual change. The RKO radio tower is to film as the grasp reflex is to newborn babies.

This analogy is misleading. The truth is that carryovers from older to newer media are not always dead weight. In many cases old media serve as a fulcrum on which to advance the goals of new media.

Tropes as bridges

Consider the spinning newspaper trope (first used in Citizen Kane—an RKO radio picture). Presenting a brief banner headline, a film communicates that an important event has occurred, shows that it has impact, that it is known publicly, and that some time (enough to publish the paper) has passed. Taking advantage of the audience’s understanding of how newspapers work, this technique manages to get across a great deal of information in an efficient manner.

In this understanding, the radio tower before a film isn’t without function, but is instead a bridge between the old and new. It is as the language used to describe internet media—email, web-site—something familiar to explain something unfamiliar.

The spinning newspaper remains effective despite the fading relevance of and familiarity with newspapers nowadays. Ditto the comedic record needle scratch²: how many viewers laugh at it, but don’t know where and how the sound is produced?

(Of course, things continue to change. These tropes are not going to be effective forever.)

Tropes as interference

Ideas from old media can be useful, but can also interfere with the new. Their effectiveness and convenience can limit media as well as they can leverage it.

Take, for example, film and video games. Video games often use letterboxed screens (once only seen when watching widescreen film on television screens) to denote sequences in which players are meant to watch the game as they would a film, rather than participate in it.

For the same reasons, letterboxed cutscenes are frustrating. They are a sign that the game is no longer an interactive experience, but a passive one; that the designers were not able or willing to look beyond their understanding of film.

The sequence in which Sephiroth kills Aeris works in the same way as does the twist in The Sixth Sense. It’s more film than game. Like radio on TV (which The Score seems to think is a good idea), it disregards core, distinguishing aspects of the newer medium to shoehorn in an older one.

The spinning paper trope does not make a newspaper of the film in the way Final Fantasy makes a film of a game. The spinning paper requires viewers to read only a headline, not an entire article. The trope makes use of the newspaper form without mimicking it. In this way, it is a successful borrowing.

Artful use of old media is possible. The video editor included with Grand Theft Auto IV, with which players replay and edit stunts they’ve performed in-game, is an example of how, as Steve Gaynor has said, “taking cinematic techniques and applying them to games that isn’t ‘let’s just put movies into games.’”

Just as old media can be useful, a reservoir of convention and time-tested mechanics, they can be distracting and limiting. The methods and tropes borrowed need to be questioned, sometimes tempered and adapted, sometimes discarded.

To me, the key difference between first-person games on PCs and consoles are the control schemes, how they map from intentions (move forward) to player actions (press W), and how the buttons, sticks, and etc. interact with the player’s body.

For example, keyboard control requires coordination of multiple fingers for diagonal motion (e.g. hold W and A), whereas analog controls have a more direct mapping of intention to action (the movement vector is the angle and direction of the stick from the centre).

Speed is also different. In Thief, the player using a keyboard has to toggle between discrete sneaking and running modes: there are only two speeds. On the console, the player has direct, analog control of their character’s speed: it’s up to the player how fast or how slow to move.

The keyboard also has a greater management cost: the player has to remember stuff like the keys to toggle modes, the mode she’s in, and has to coordinate two or three simultaneous actions across two or three fingers.

I think in these two examples, dual analog controls are less abstract and more closely related to what the player is trying to achieve in the game world than are keyboard controls. This might be why some players find it easier to pick up and learn console games.

That said, analog sticks are controlled by thumbs—they’re literally “all thumbs.” With the mouse, motion comes from wrist and arm movements. With the mouse, gross movements are easier to make (arm) and fine ones too (wrist, slow movement). Thumbs aren’t as capable as arms and wrists, and that makes the tradeoff between accuracy and speed harder to adapt during play.

The mouse is also not bound by an arbitrary magnitude as are analog sticks (which the player can only push so far). Instead, mouses are limited (at most) by the physical range of the player’s arm (or the mouse wire): I can move it almost as far and as fast as I can. This means that, in theory, I can whip around much faster in Quake, and snipe more effectively in Team Fortress, with a mouse than with analog sticks.

But, in practice, I suck at Halo and Team Fortress, mouse or gamepad.

On coming home, I found the essay and read it over. I’ll admit, its conclusions are not earth-shattering—basically “it’s complicated”—or inarguable, but it was enough to spark and fuel some pretty interesting discussion. So, I’ve reproduced it here. If you agree or disagree, take offense or whatever, feel free to post a comment. Please keep in mind that I’m no historical linguist or essayist, the citations are patchy, I like commas, the sun was shining in my eyes, etc.

Language and thought control

In his most popular work, 1984, George Orwell introduces the language of Newspeak: “a medium of expression for the world-view and mental habits proper to the devotees of Ingsoc [the totalitarian governing power]” that will “make all other modes of thought impossible… literally unthinkable, at least so far as thought is dependent on words.”¹ The language is based on English, but with significant changes, including a more regular system reliant on prefixes and affixes, an emphasis on shorter, easily pronounceable words, and a drastic reduction of vocabulary. Orwell seems to think this last feature most important in controlling speakers’ thought, “each reduction [is] a gain, since the smaller the area of choice, the smaller the temptation to take thought.” By shrinking the choice of words and simplifying their construction, Newspeak aims to shift the locus of control over language from the higher brain sectors to the larynx, away from any unorthodox or seditious thought, making them inexpressible. By 2050, Newspeak is to be the one and only spoken language in the lands ruled by Ingsoc.

Despite being a work of fiction, the threat that language can be co-opted or replaced by those in power is considered as real outside of 1984 as in. Mamet insists that “names are powerful,” that “the assignment of nicknames, the application of jargon is an understood tool for the manipulation of behaviour.”² Noting the increase of unnatural, government-made terminology in the United States since late 2001—weapons of mass destruction: “overlong, clunky, and obviously confected”—he warns against a shift “from the conscious into the automatic,” worried of a linguistic take-over very much like the one in Orwell’s dystopic England.

Are fears of Newspeak justified?

The fear of political control through language is based on the implicit assumption that “the language spoken by the individual determines the way in which that person thinks”³, a concept known as linguistic determinism. If true, this concept, distilled from the work of Edward Sapir and Benjamin Lee Whorf, would certainly be cause for alarm; Newspeak, once adopted fully, would limit thought and expression eternally.

However, linguistic determinism does not stand up to scrutiny. If thoughts were determined by language, how is it that new words come about? How could “nigger” (“nigga”) and “queer,” both potent derogative words, have changed over the past few decades into friendly terms of address or exclamations of pride? In truth, the relationship between language and thought is not unidirectional, but two-way, social reality both shaping and being shaped by its language. “Sociolinguistic conventions have a dual relation to power: on the one hand they incorporate differences of power, on the other hand they arise out of—and give rise to—particular relations of power”⁴.

The examples of “nigger,” “queer,” and “sexist”⁵, are doubly important: they are shifts in language use introduced by minorities, groups deprived of much political influence, often working against established powers. While examples such as these do not eliminate language as an influence on thought (and, indeed, many of the same social causes that gave rise to sexist aim, wrongly or rightly, to reform language and, in doing so, common conceptions of gender⁶), they do diminish it. Constraining language does not necessarily constrain thought, thus fears or fantasies of absolute control and stagnation of ideology by means of language alone are less likely than Mamet or Orwell propose.

Language and empire

Language alone may not determine thought, but it is a vehicle for and a body of ideology, a way of exercising power⁷. Those in positions of control—over nation-states, corporations, armies—as well as many of those over whom they hold sway, act accordingly: idioms are imposed, language changed through coercion, use warped on the fulcrum of political power. Phillipson quotes a Spanish report written for the Queen in 1492 that proposes the use of language as “a tool for conquest abroad,” noting that “language has always been the consort of empire, and forever shall remain its mate”⁸. The difficulty (likely, impossibility) of engineering thought through the manipulation of signifiers (words) may not be overcome, but spreading a whole system of signs—a language—as a means to indoctrinate or undermine another group of people has been attempted time and time again. Could Newspeak ever become the only language in a large empire? Is political power effective in enacting and maintaining a language? Just how great a factor is power in language change?

Japanese occupation

Consider Miyawaki’s study⁹ of the harsh Japanese colonial language policies in Taiwan, Korea, Micronesia, and occupied territories in China and in Southeast-Asia during the first half of the twentieth century. These policies were rigid and explicitly aimed at eradicating non-Japanese cultural and linguistic influences, as well as impressing Japanese values on the colonized people. In Taiwan, this began in the mid-1890’s by legally

stating that the fundamental objectives of common school education be the provision of moral education and practical skills to Taiwanese children, thereby cultivating in them attitudes of Japanese nationalism and also leading them to be well versed in ‘Kokugo’ [the national language i.e. Japanese]…

More drastic revisions such as the abolition of the native language (Chinese) teaching and the integration of the educational system and curriculum with those of homeland Japan were made in 1937 and in 1942 respectively.

The intent of such policies was stated more forcefully during the Pacific War. Japan pressed for the use of Japanese not only in schools, but at home; they hoped to “diffuse Japanese, gradually limit the use of European languages and eventually abolish them” in Southeast-Asia, “to stamp out European/American thoughts, and establish an Oriental-minded culture.” Positive assimilation, policies that were to nurture Japanese culture in the colonies, quickly became negative, punitive and brutal. Miyawaki finds that many native-born students under Japanese rule recall being publicly humiliated, sometimes beaten, for speaking any tongue other than Japanese.

Yet, for all the influence and coercion, both positive and negative, exercised by the Japanese colonizers, the proportion of Japanese speakers to non- in the former colonies today is small to none. Miyawaki notes a variety of small linguistic changes, notably borrowing, pidginization, and a bilingual minority, present to this day. These effects, however, are shallower and more localized than what one would have imagined a large, modern empire would be capable of over fifty years. Miyawaki concludes that colonial power is only one among many influences that can affect “language ecology.” Much as with language and thought, the relationship between power and language is less direct than anticipated (in Japanese policies). Although the Japanese policies did alter “the society, culture and psychology of the ruled,” they did not determine it, and did not succeed in imposing a foreign language on a conquered people over the long term.

Linguists urge readers to consider a myriad of factors and constraints that may cause language change. In their introduction to Language and Power, Kramarae, Schulz, and O’Barr present a variety of opinions on how language and power may interrelate, so many different avenues of study and interpretation that they seem to through their hands in the air, claiming that an “adequate understanding… may be several sociolinguistic years away.”¹⁰

Ancient Akkadian

Ostler gives much weight to the influences outside of direct political control in his historical analysis of language change.¹¹ He is quick to dismantle J. R. Firth’s assertion that “world powers make world languages,” pointing out that the Germanic rulers of Europe that succeeded the Romans were only a slight influence on the Romance languages still in use to this day, and further, that the Romans themselves were incapable of imposing Latin on their subjects in the east, where Greek remained the common tongue through the hundreds of years of Roman rule. Ostler finds explanations for lasting language change reliant primarily on political power, “based on military conquest or commercial dominance,” lacking; even “total conquest, military and spiritual, is not always enough to effect a language change.”

Ostler gives the example of Akkadian, the primary language of the impressive Assyro-Babylonian empire. Akkadian was “preeminently a language of power and influence,” a literary standard, the single language of an empire lasting almost two thousand years. The influence of the empire helped spread its language, the uptake of the Akkadian in lands outside of Babylonian control carried largely by prestige (and others’ eagerness for the relatively new technology of writing), until it became a well-established lingua franca among the many people and powers of the time. Yet, the language was overwhelmed by Aramaic, a language spoken “mainly by nomads,” a community radically different from, and hostile to, the Babylonians, with “no cultural advantage… highly unlikely to set up a rival civilization.”

Stranger still, this change in language came at the zenith of the empire’s power. Having conquered much of the area after decades of successful war, a policy of separating conquered people was instated, intending to unify the populations by “cutting them off from their traditions” while acculturating them to Assyro-Babylonian culture—importantly, its language. This policy of division and assimilation eventually displaced some 4.5 million people over three centuries, an act of immense power. Unfortunately, this policy did not have the intended effect. It backfired, encouraging the spread of Aramaic and undermining Akkadian as a common language.

Since the Aramaeans were the largest group being scattered in this way, when other western Semites, such as Israelites or Phoenicians, found themselves transplanted, they could tend to find themselves speaking more and more like their new neighbours.¹²

Aramaic quickly became the dominant language, and remained so, during the following centuries of the empire and long after its collapse. Both Assyro-Babylonian and Japanese policies (some 2500 years apart) shared similar goals and failures despite being backed by powerful, long-established cultures and military forces. The triumph of Aramaic over Akkadian is an extreme demonstration of the weakness of political influence on language.

Determined not by power alone

The wealth and might of a political power does not determine the spread and stay of the language it speaks, and may well, as in the case of Aramaic and Akkadian, exist separately from it. Through means insidious, duplicitous, or explicit, political powers may mean to affect language and, through it, how people think. However common the belief that language molds thought, it does not appear to, certainly not the the degree feared by paranoiacs or wished for by propagandists. The influences of language on thought, of power on language, are complicated, two-way, and subject to many other forces. Political power alone cannot guarantee the abolition or spread of a language, nor can it always succeed in altering it.

(If you’ve not already seen WALL·E, you may not want to continue reading this post: it may spoil certain surprises, and won’t make much sense.)

The two robots in Pixar’s WALL·E represent, roughly, two kinds of personal computers: the PC (in the old sense: a desktop computer running Windows or maybe, GNU/Linux) and the Apple Macintosh (running Mac OS).

EVE has a glossy white shell that resembles an iPod or MacBook. As Sancho has mentioned, EVE was designed in part by a designer at Apple. While the superficial similarities are easy to find, you can find others if you’re willing to stretch a little.

Much like an Apple computer, EVE looks elegant and packs more power than you’d first expect. EVE is quite dedicated at performing the task she is designed to do. She goes to an expensive maintenance area, all white walls and frosted glass, to be repaired by experts. It’s really obvious when EVE gets scratched or smudged, and it takes effort to keep her looking clean (MO, the cleaning bot).

WALL·E, the PC, is made of worn and noisy machine parts. He works with what’s laying around, upgrading his eyes and treads, as well as adding non-standard enhancements—a lunchbox. WALL·E is not easy to repair, his boot-up sequence is slow, and data recovery can be an suspenseful ordeal, as EVE discovers late in the film. He seems even to invite bugs to crawl around him.

While WALL·E is not as modern and shiny as EVE, he’s more flexible and the only one capable of playing video games.

Alright, perhaps I’ve stretched the analogy too far. Nonetheless, it does seem difficult to deny that there is some unusually tight cross-branding going on in WALL·E. There are a number of nods to Apple in the film, some more conspicuous—WALL·E watches movies on a video iPod—than others. (This may excuse one of the failings of my analogy: that WALL·E plays the Mac start-up sound when he has charged his solar battery.)

I don’t find this kind of product placement particularly cute. Yes, EVE is a more sympathetic and business-friendly spokesperson for Apple than the smug Mac dude, but it’s odd to see this kind of marketing in a movie that uses the ubiquity of a corporate brand as a sign of decadence and ruin.

I do choose to interpret the love between the two bots as a hopeful message for nerds and platform-zealots everywhere. Like EVE and WALL·E, Mac and PC users can learn to get along.

Update: Seems I’m not the first to see the PC/Mac parallels.

On the white board he sketched a rough plot: two curves, with amps along the x axis and voltage along the y. “Which of these would you say has the greatest power, the green or the orange?”

Green because it’s taller. Duh. Next question.

If you know what you’re doing, you’re not going to fall for that. You know that power can be found by taking the area under the curve. From the look of it, the broader orange curve will have a greater area, so you put orange down as your answer.

Now, if you’re really paying attention, you’ll notice that the scale along the y axis isn’t linear, but logarithmic. This means that, while the green curve doesn’t seem to be any more than two times taller than the orange one on paper, it’s actually representative of a much greater difference between the two. Since the green curve reaches far higher than the orange one, it covers a greater area. You answer green.

Green is the correct answer. If you’re dumb or lazy you get it and if you’re smart you get it, but if you’re in between, you know just enough to get it wrong.

Kelly’s punchline: “But I think that’s what they were going for. They wanted to get people who were smart enough to trust with guns, but dumb enough to go to Vietnam.”

We could call this kind a porridge filter: it selects those that are not too ignorant, not too knowledgeable, but just right.

How common are porridge filters in the real world? Are they more often products of design (as Kelly suspects of the above case) or accidents? What other cases could they be useful?

By the way, if I’ve messed up my physics, let me know.

Update: I came across the following in Neal Stephenson’s The Confusion.

“What is the Intelligence Test?” he demanded to know, and swept the curtain aside.

“A private joke,” said the annoyed Padraig.

But Jack saw good reasons to explain it, and so he said, “Cast your memory back to when Fortune had set us ashore in Surat—”

“I remember it every day,” said Surendranath.

“You stayed there to pursue your career. We fled inland to get away from the diverse European assassins who infested that town, and who were all looking for us. Soon enough, we came upon a Mogul road-block. Hindoos and Mohametans were allowed to pass through with only minor harassment and taking of baksheesh, but when it became known that we were Franks, they took us aside and made us sit in a tent together. One by one, each of us was taken out alone and conducted to a field nearby, and handed a musket—which was unloaded—and a powder-horn, and a pouch of balls.”

“What did you do?” Surendranath demanded.

“Gaped at it like a farmer.”

“I likewise,” said Padraig.

“So you failed the Intelligence Test?”

“I would rather say we passed it. Van Hoek did the same as we. Mr. Foot tried to load the musket, but got the procedure backwards—put the ball in first, then the powder. But Vrej Esphahnian and Monsieur Arlanc loaded the weapon and discharged it in the general direction of a Hindoo idol that the Moguls had been using for target practice.”

“They were inducted,” said Surendranath.

“As far as we know, they have been serving in the armed forces of the local king ever since that day,” Jack said.