Tuesday, December 9, 2014

6 major language anti-patterns

I’ve noticed over the years a few common misconceptions people (especially me, until I learnt better) have with languages they’re not familiar with. Like computer programing antipatterns, these bad ideas can often be grouped and given a funny name.

Imagine a sailor on the Spanish main who accidentally dropped a waterproof bundle of his possessions over the side of the brig and expected to find the same bundle years later at that exact same spot, bobbing up and down, without having moved in the interim. That would be bizarre, you might say, while also observing that it’s quite improbable such a thing would ever happen given that a sailor, of all people, would be expected to appreciate the power of wind and ocean currents. Never mind that for now.

If it was just one sailor you might think that they were affected by some mysterious ague and forget about it. If it came up again and again though you’d probably want to put a name on this pattern of behaviour, something like 'St. Tibald's Fever', just to make it easy to refer to it now that a few people were afflicted with it. You might even call it an ‘anti-pattern’.

Anti-patterns are a thing in computer programming and web development, my particular field. They are wrong ways of doing or thinking about things, commonly done. They often have vivid, descriptive names like 'lava flow', 'cargo cult programming', and 'golden hammer' that make you want to find out more. When you do you realise (if you're in IT) that not only do you recognise the problem, you're actually 'lost in the weeds' yourself, or, worse, in a 'death march', with ‘God objects’ and 'poltergeists' everywhere.

When we programmers get together, we find it useful to refer to these bad practices by name like this to spare our friends having to listen to a five-minute tirade about the messy code at work, where you can't find anything and things are just put anywhere willy nilly, and why is that file there, and this interface here, and ... "Oh you mean it's 'spaghetti code'?"

It seems to me that this convention would be useful in the similar domain of language learning. If the same mistakes, broadly, are made over and over by people, then a nomenclature system can be useful to help identify them, or at least to shortcut the process of talking and writing about them. In case this already exists and I'm coming late to the party, I apologize. In case it doesn't, splendid, let's continue. In either case, here are my proposed anti-patterns.

Howard Clifford running off the Tacoma Narrows Bridge during collapse, Tacoma, Washington
Howard Clifford running off the Tacoma Narrows Bridge during collapse, Tacoma, Washington, Flickr

1. Argumentum ad Exoticum, or "Those Crazy Chinese"

Faced with a deeply unfamiliar set of characters, like Chinese ones, people begin to form some strange ideas that borrow heavily from craziness. Just because the building blocks are a different type to our own doesn’t mean that most of the basic concepts of communication, like words for example, don't exist. Yet this is exactly what I've heard from someone I know, someone interested in languages, namely that Chinese doesn’t have words, and I’ve argued against this if only just to keep the net amount of sanity of the universe from dwindling more than it has to. Not on my watch and all that. "I just read on Wikipedia (an attempted Appeal to Authority right there) that they don't have words, just characters. Just because we have words doesn't mean every language has to have words you know."

I asked my friend to think of a bazooka in his big fat head, and imagine two Chinese people discussing whether they should purchase a new bazooka. What in God’s name did he think they use when they’re talking to each other about shoulder-launched munitions if not some word thing corresponding to the word thing we’d use in our language to mean ‘bazooka’? Do they whistle instead? Or mime? Or all jump up and down at the same time and cause a tsunami? Chinese has words in the same way we have words: arbitrary assignments of sounds and signs to represent things in the real world, often made up recursively of more basic sounds and signs.

Or do they? The strangest thing was that even though I had been studying Chinese for a while, I began to doubt my own certainty that Chinese had words, such was the vehemence of my friend’s objection. Because we trade in words in English, had I simply assumed that things like 孩子 (child) and 你 (you) were “words”? They don’t look like words, really...maybe they’re not. I was feeling the cold hand of exoticness on my shoulder.

I shrugged it off. Of course they effing well have effing words! They even have a handsome word for word: 词. I have since decided that I should have more faith in my own studies, and have sadly given up on my friend as being willfully obstinate and abstruse when it comes to unfamiliar concepts. Chinese will stretch you as a language learner if you’ve only done Indo-European tongues before. Don’t be put off by appeals to exoticness from a starting point of ignorance, and above all use your common sense.

2 Yoda Sprechen Sie?

This is the idea that word order unimportant is when a rich flexional system one has, like in German. But the problem is that their flexional system, like many a flexional system, is a vestigial appendage, a largely useless hangover from a time when knowledge was in the hands of Greek- and Latin-lovin' scholars disinclined, as it were, to open source their work. You are retrofitting a putative pragmatic purpose to this useless flexion system by assuming that Germans efficiently (yes!) avail of this syntactic godsend and just place words anywhere since the endings of the words flag the subject and the object in a sentence like "the man bites the dog" (der Mann beißt den Hund) in a way that, burden-wise, has to be shouldered in English by word order.

Let’s look at it in reverse: you could say then that because the word order does largely dictate what the subject is and what the object is in your typical English sentence, then surely what few flexions that still exist in English can be arbitrarily reimagined on the fly since they are as surplus to requirements as word order is in German. Ergo, us should all just alternate between say "I spoke to her" and "Me speak to she" (and all combinations in between) because it obvious who be do the speak (me did) and who be do the listen (her did) from the word order. Go ahead and try that in English if you think word order is irrelevant in German.

3. La Carte et le Territoire

I read a book a few years ago in French by the bête noire of belles lettres Michel Houllebecq called La Carte et le Territoire, whose title I want to use here to highlight a confusion between a map and its territory. For example, just because in Spanish they use the word “gerundio” doesn’t mean that the word “gerund” maps to the same thing in English. If you think this, as I did until recently, then your map is slightly out and you need to redraw it. I got into a kerpickle on Duolingo about this very misunderstanding. I thought the map was the territory. But there are kinks; it’s not a straightforward mapping at all.

In English a gerund is more than just the form of a verb ending in “ing”. It’s only a gerund when that word becomes a noun, like “The Taming of the Shrew”. “Taming” here is a verby noun. By the way, don’t use that term (“verby noun”) around language people. This sentence, this one, on the other hand, as well as being a bit contrived, does not in fact have a gerund. In Spanish however, las cosas son distintas. Even though the word gerundio also refers to a verb in “ing” mode such as “esperando”, or “andando” it doesn’t need to be a noun to be a gerundio.

I got schooled on Duo by people who knew this when I didn’t. I naively assumed that gerundio was the point on the Spanish map corresponding to the concept of an English gerund, and that therefore they were conceptually identical. Maps are handy, but they’re not the territory.

4. Function doesn’t always follow form, or "It may well walk like a duck and quack like a duck but it’s actually a rhino"

Have you heard about those Myrmarachne jumping spiders that have evolved to look like ants? They resemble the little critters to such an extent that they insinuate themselves into the ants’ business and then turn on them, all unbeknownst like. In the animal kingdom that’s called Batesian mimicry. In Greek it’s called ‘deponent verbs’. You have this species of verb which has evolved to pass itself off as a reflexive-looking verb, complete with pronoun suffix at the end à la decidirse in Spanish. In some cases it has a passive, or reflexive meaning, as in ‘I am loved’ but all too often that’s a red herring and it‘s just a normal, active, common-or-garden verb. Έρχομαι means I come, χρειάζομαι, I need. No reflexivity or passivity there.

I got into a bit of a ding-dong with my Greek teacher Kelly when she tried to explain all deponent verbs like this to us as being somehow passive in meaning, if only in some cases in an underlying, implicit kind of sense. I pointed out that no matter how you slice it “I come” is not passive in English, and hard to see how it could be passive in Greek. Same with “I need”.

Confusion reigned, and she probably felt her authority was being undermined by an Irish smartarse, which it was. I felt the others’ annoyance from then on when I stopped the class and demanded to know more about stuff that didn’t add up (for all that, I was the only one who finished the course: the others had all dropped out by the end.) I wrote to the guy who wrote our textbook, Theodore Papaloizos, feeling that this particular aspect of Greek needed to be spelt out very painstakingly for English learners in particular. He wrote back "All languages have their peculiarities and Greek of course is not an exception. For a student of Greek it is difficult to understand how a verb looking and conjugated as a passive voice verb has an active meaning. Most of these verbs come from ancient classical Greek." It might be easier in other language pairings, but deponency is just cruel and unusual for us anglophones. Just because something looks a certain way, doesn’t mean that it behaves that way.

At one stage, in a bold manoeuvre by which she sought to outflank me, Kelly came in and told me that she had confirmed such and such a point about deponency or passivity by checking with her husband, who is also a Greek speaker (and in IT as far as I remember), therefore QED and shut up. She had unwittingly plunged headlong into the abyss of anti-pattern no 5...

5. Argumentum ad Naturalis Omnisciens

...which is the mistake of thinking that native speakers have some innate handle on the grammar points of their mother tongue. If you suffer from this delusion, turn to the person beside you at work and tell them you’ve forgotten when to use the subjunctive again, would they mind running you through the rules? And the forms the words take too. If they wouldn’t mind, like? Seeing all too clearly where I’m going with this you might say, alright, they may not know all the stinkin’ rules, but they’d surely know whether something sounded right. Ah, would that this was true, or..hang on... is that ‘were true’? I’m not sure.

And that’s English, with the world’s tiniest subjunctive and verbs that hardly move at all. Greek verbs flex like a prepubescent Russian gymnast, so the idea that Greek speakers can rattle off the rules of irregular verb formation, or the subjunctive, or the deponent stuff, I have found to be, well, laughable. My wife, with respect, rabbits away to her Mum in Greek. She proved to be of little or no use to me, God love her, when I was hitchhiking through the badlands of Greek grammar and got waylaid by delinquent local constructions.

6 Meaning, meaning, everywhere

An overarching misconception, common to several of these anti-patterns, is born of a failure to grasp just how arbitrary, random, and plain ol’ in flux much of language is. This anti-pattern manifests itself in the form of the old Platonic chestnut wherein, say, the building blocks of Chinese characters are seen to have such a rich, meaning-laden storytelling role in the language that when they are used as part of a character they bring that semantic basket of goodies they have preserved intact through the centuries with them. They might tell a little cautionary tale here, or proverb there, like the one my brother-in-law told me about the compound character where two of the ‘building block’ characters which ‘stand for’ a woman, all under a roof character, means strife or conflict. Because, you know, the wily old Chinese realised, chuckle chuckle, that having two women in one house would be like having a bag of cats. This I grandly call The Raconteur Theory of Chinese Characters. Alas, there is no deep insight into Chinese or the human condition to be found here: words and characters are just tools, not lessons.

Subscribing to 'Raconteur’ you are like someone who thinks we’re referring to horses when we talk about someone’s chivalrous behaviour simply because cheval means horse in French. Many a word’s meaning has moved on over the years in a process of analogical extension, blurring and Chinese whispers. You are like the sailor who expects to find his bundle bobbing in the same spot if you believe otherwise. You, sir or madam, have St. Tibald’s fever.

As anyone who’s studied the origin of Chinese characters knows, “there has been a progression from pictographic, writing the picture; to ideographic, writing the idea; and then logographic, writing the word” to quote James Gleick in The Information. Nowadays we’re just writing words, not necessarily ideas. Which means that in many cases, much as you might invite an acquaintance along to a party simply because he has a guitar and can sing, part of a Chinese character is there just for its sound: it’s as semantically void as a waffler’s promise.

There is a very human tendency to overexplain or retrofit explanations to try and justify something that’s actually just the way it is for no better reason than euphony (it sounds better) or historical contingency. You don’t want to confuse a mem, a way of remembering something, often involving a story, with the ipso facto thing itself. We like stories, but beware the Narrative Fallacy which says that stories are often the wrong causal explanation for something that may be better ‘explained’ in statistical terms. In fact, as far as learning words is concerned, the best explanation is often “that’s just the way it is.” Let it go at that without burdening a language with the naive expectation of logic everywhere, and be happy that they’re much more idiosyncratic and poetic than that.