Wittgenstein, Watson, and Language Games

Here the term “language-game” is meant to bring into prominence the fact that the speaking of language is part of an activity, or of a form of life.

Review the multiplicity of language-games in the following examples, and in others:
Giving orders, and obeying them—
Describing the appearance of an object, or giving its measurements—
Constructing an object from a description (a drawing)—
Reporting an event—
Speculating about an event—

Forming and testing a hypothesis—
Presenting the results of an experiment in tables and diagrams—
Making up a story; and reading it—
Singing catches—
Guessing riddles—
Making a joke; telling it—
Solving a problem in practical arithmetic—
Translating from one language into another—
Asking, thanking, cursing, greeting, praying.

–Ludwig Wittgenstein, Philosophical Investigations

Jeopardy! and John Henry

Consider another kind of language game. A quizmaster reads a clue. For example, “William Wilkinson’s ‘An Account of the Principalities of Wallachia and Moldavia’ inspired this author’s most famous novel.” You then try to formulate a question that would evoke that clue as an answer. For example, “Who is Bram Stoker?” You are competing against two other players. Whoever formulates a correct question first wins a certain amount of money. The person with the most money after a set number of questions wins the game. The clues are drawn from general knowledge—history, geography, science, culture—and often involve wordplay: rhymes and puns and whatnot. To be good at this game it helps to have an encyclopedic knowledge of trivia, quick recall, and a corny sense of humor.

In 2011 a computer program built by IBM research called Watson1 appeared on the game show Jeopardy! and defeated its human opponents, among them Ken Jennings, the reigning champion. The contest was structured like a regular all-human competition: Watson played by the same rules as the people. The clue above about Bram Stoker is the final one with which Watson secured its victory.

The computer program Waston on the game show Jeopardy!

It is astounding that a computer defeated a person in this arena. Machines are good at certain things (storing and retrieving vast quantities of data, working without a pause for years at a time) and humans are good at other things (synthesizing, inferring, catching jokes). Jeopardy! would seem to favor the latter strengths. Furthermore, the standard assumption in the field of artificial intelligence is that the humans will always be smarter than the computers—our job as programmers is merely to make the computers smart enough to be useful. So Watson’s victory on Jeopardy! was a tremendous upset. It is not a case of the steam engine defeating John Henry. Rather it is if by some miracle a mass-produced dining room set turned out to be of higher quality than one hand-built by a carpenter.

In his book Philosophical Investigations from a few decades before, the philosopher Ludwig Wittgenstein claims that speech is a kind of game people play. He talked about games to emphasize the fact that speech exists not just for the purpose of transmitting information, but also as an end in itself. (What is the purpose of playing a game of chess, after all, other than to play a game of chess?) Later in the same book, Wittgenstein tries to come up with clearly articulated definition of the concept of a game, but after considering the varied activities that might fall under that heading (chess, athletics, jumprope, political contests, war) gives up. So in using the game metaphor he is also emphasizing how varied the act of speaking is. It is foolish to try and reduce language to some formula. Talking is simply one of the many things people do.

Wittgenstein knew nothing about computers. His notion of language games was a metaphor, but was prescient in this instance because Watson literally played a language game and won. So we may want to keep him in the back of minds in case there is other guidance he might offer.

A Russian Language Game

You can simulate what it’s like to be Watson right now. Type the phrase “William Wilkinson’s ‘An Account of the Principalities of Wallachia and Moldavia’ inspired this author’s most famous novel -IBM -Watson” into your favorite search engine.2 This brings up multiple links to articles about Bram Stoker and Dracula. It is easy for you to skim them and infer that the correct answer is “Bram Stoker”. (Sorry–“Who is Bram Stoker?”) So easy that it almost seems like cheating. But there’s a crucial difference between you and Watson: you understand English. Watson does not.

So in the interest of realism let’s play a different language game. The rules of this game are that I give you a clue in Russian and you have to return the correct answer in Russian. (I’ll permit you to skip the whole phrase-it-as-a-question business.) You are allowed to use a search engine, but you are not allowed to understand Russian. So now when you blindly perform the search “‘Рассказ о княжествах Валахии и Молдавии’ Уильяма Уилкинсона вдохновил самый известный роман этого автора” it brings back links to documents that are incomprehensible to you. What do you do next? How do you win?

Google search in Russian

Let’s make the rules of the game a little more forgiving. You are not allowed to understand Russian, but you are allowed to recognize Cyrillic characters, distinguish words, and tell when two phrases look similar or different. (Basically you are allowed to understand Russian as well as someone who doesn’t actually understand Russian.) If you carefully combed through the top hits returned by the search engine, you might notice the phrases Дракула and Брэма Стокера showing up repeatedly. Perhaps one of these is the answer.

It order to choose one term over the other it would be helpful to know the relationship between them. Knowing this offhand would be considered understanding Russian, however, if you did web searches on these two terms, you might discover that they often occur near each in the same documents. Furthermore you might notice that these documents often also contain the terms автора and роман, which appear in the question as well. You wouldn’t know what these words meant either, but their presence still might pique your interest. If you had enough examples of similar clues and answers about authors and novels3, you might be able to recognize typical word patterns that would lead you to guess the correct answer, “Брэма Стокера”.

Брэма Стокера Bram Stoker
Дракула Dracula
автора author
роман novel

In order to discern these patterns, you’d have to make careful tallies of Cyrillic word shapes across many thousands of documents and analyze them with sophisticated mathematical techniques capable of teasing out the subtle correlations between them. It would be a much too tedious job for a human being. You’d need a computer.

At Play in the World

Watson plays something very similar to the Russian language game. It analyzes the clue it is given, extracting relevant terms that are then used to perform a query of a general knowledge database. Candidate entity names returned by the query are ranked by a machine learning model trained on a long history of Jeopardy! questions and answers, and the highest scoring one is proposed as the answer.4 There is also knowledge about the grammar of English and basic ontology baked into the program, but for the most part Watson is just recognizing patterns of words.

This naturally raises the question, are we, the humans, playing the same game? Is our understanding of English, Russian or what-have-you ultimately just unconscious, statistically-driven pattern recognition? This is a contemporary rephrasing of the question of how to distinguish between appropriately conditioned behavior and true comprehension, which has a long philosophical history outside of software. Officially, engineers like myself are agnostic on this issue.5 We are only concerned with getting the appropriate responses and don’t care what underlies them. As a practical matter though, there remains a vast gulf between the variety of games a human and a computer program—even a world-class program like Watson—is able to play.

Consider another world-class computer program—the Google web search engine. To find the answer to the Bram Stoker clue, I typed it verbatim into Google, which instantly provided relevant results. It turns out that this is often the case: Google is an excellent way to cheat at Jeopardy! However, the Google search engine could not have gone on TV and beat Ken Jennings because it returns links to documents and relies on a human being to make sense of them. That is not sufficient for Jeopardy! There you must return the name of a specific entity couched in the form of a question, and as the Russian language example above demonstrates, going that final mile is harder than it looks. Google can’t win at Jeopardy! because it’s not playing by the rules of that game.

In fact the rules of this particular game impose all manner of non-obvious constraints. Both clues and answers must be concise. “What effect has the character Dracula had on film and literature?” is a valid question to ask, but no good for Jeopardy! because you could fill books answering it. Answers must not be a matter of opinion (“This gothic tale about a bloodsucking count is the greatest novel of the 19th century”) and clues must not contain incorrect presuppositions (“This male English author wrote the novel Frankenstein”). The convention of answering in the form “What is–?” “Who is–?” restricts answers to being well-defined entities, while the quiz show format disallows all manner of discourse. The designers of Watson knew their system would never have to produce a poem, or a joke, comfort a grieving widow, or maunder on about the weather. The set of utterances you don’t have to handle is as vast as the set of ones you do.

Navigating open but still constrained domains is where the field of artificial intelligence stands at the moment. We know how to play particular games. Given a task—find a webpage, recommend a movie, transcribe a spoken utterance, win at Jeopardy!–and a large set of exemplars of how human beings have successfully performed that task before, we can find a way to train a machine to imitate them. Usually not surpass6, but at least emulate to some reasonable degree. It’s not easy—machine learning is still as much an art as a science—but for the foreseeable future there is a clear way forward that lies in making incremental progress by winning incrementally different games.

Computer programmers have a instinct towards generalization, and so naturally wonder if there is an approach that could subsume the current piecemeal state of the art. You’d want there to be a single human game—call it reason, rationality, intelligence—that we could learn to play just once, and have particular tasks fall out as special cases. This was the dream of both an earlier generation of artificial intelligence researchers and an earlier incarnation of Wittgenstein, who in the Tractatus Logico-Philosophicus proposed a sort of gnomic form of predicate calculus as the definitive solution to all outstanding philosophical problems. Wittgenstein later renounced this position, and his description of the multiplicities of language games in the quote above reflects his later view that a description of human experience will never be able to fully abstract beyond the particulars. Currently, engineers are okay with attacking sets of particular problems and undecided as to whether these may someday be unified. Wittgenstein, however, warns that any attempt at this unification is a fool’s errand. There is no master template from which all reasoning derives. Instead it’s just games, games, games all the way down.

1 Usually I find these disclaimers superfluous, but since my job title is “Watson NLP Developer” I suppose I should state for the record that the opinions expressed on this blog are entirely my own and do not reflect those of my employer.

2 The “-IBM” and “-Watson” are necessary for the sake of fairness, because all the web pages that contain this phrase verbatim discuss Watson’s victory on Jeopardy!

3 Of course you wouldn’t know they were about authors and novels. You’d just know that certain Russian words tended to pattern with certain other Russian words in certain ways.

4 The system is described in detail in the May-June 2012 edition of the IBM Journal of Research and Development.

5 Though personally I have to say that it sure doesn’t feel that way to me.

6 It is interesting that the other great publicity coup IBM has scored in the past twenty years was Deep Blue’s defeat of the chess grandmaster Garry Kasparov. I wonder if Wittgenstein had lived later in the 20th century he would have defined a game as “something a computer could eventually defeat a human at”.

Posted in Fabulous ones, Mermaids, Those that have just broken the flower vase | Leave a comment

Religious Holiday Themed Starbucks Coffee Cups

–Excuse me, this cup is empty.
–Well, it’s not dark yet.

–Yes it’s small, but you get eight of them.

–Ugh! This coffee tastes like paint.

–I’ll have a double-tall soy iced-Ouch!

The Feast of Corpus Christi
[Spits out a mouthful of blood.]
–Jesus Christ!

Posted in Etcetera | Leave a comment

I May Not Know Much…

–If you leave a penny in a glass of Coke overnight it will dissolve.
–Nikola Tesla invented the electric car.
–It’s called a Brazilian because many famous Brazilian men are bald.
–George Washington made his wooden teeth out of a cherry tree he chopped down himself.
–In New Zealand they don’t use the letter “X”.
–Unicorns are extinct.

Posted in Those that at a distance resemble flies | Leave a comment

America 2050: The End of Racism

By 2050 or so white people will be a minority in the United States, just one ethnic group among many. They will not possess a disproportionate amount of the nation’s wealth, nor will their particular folkways be the default national culture. The whole notion of skin color based ethnicity will lose its power, though ethnicity will remain as important as ever. There will still be Russians, Hmong, Jews, Tamils, Lebanese, and some variety of pan-European mutt called white. Second-generation Somali immigrants and the descendants of 19th century west African slaves will feel varying degrees of affinity for and distance from one another, but no one will say they belong to the same group merely because an earlier generation of Europeans would have thought they looked alike. (“They don’t look alike,” someone born in 2030 would say. “No more than Swedes look like Sicilians.”) The descendants of west African slaves will remain a culturally distinct group, and many will develop a prickly nativist pride in having been the “original” Americans. (Which, when you think about it, they were.) A shared sense of national seniority and tradition of military service will unite “black” and “white” Americans into a consistently jingoistic voting block in the mid 21st century. Motorcycle and prison gangs will divide themselves up along lines that would seem utterly surreal today. There will still be pockets of intractable poverty that trace their origin back to slavery’s long echo, but they won’t define an entire ethnic group’s relation to society any more than the entrenched poverty of Appalachia represents something essential about what it means to be white.

Elderly people in 2050 will be confused by the younger generation’s attitudes. We will have spent our whole lives learning how to scold someone for their unconscious racism, but when some kid makes a disparaging remark about Nigerians and we call him on it, he’ll respond, “What do you mean? I don’t like Nigerians. You’re the one who brought all black people into it. You’re the racist.” The notion that everyone who can trace their family tree back to central or south America belongs to the same (race? ethnic group?) called “Latino” will simply make no sense. “Because they speak Spanish? Does that mean Malians and Québecois are the same race too?” (But whatever you do, do not wonder aloud “What does ‘race’ even mean anyway?” because that’s just begging for some old geezer to yammer on endlessly.) Elderly whites’ reflexive shows of cultural sensitivity will be taken as the condescension of a bygone era. A black septuagenarian will attempt to engage in a bit of good-natured solidarity with a 25 year-old Eritrean trust fund baby who will literally not know how to respond. There will, perversely, come to be a generational identity built around the notion of racial difference. Kids today don’t understand that some degree of inter-group friction is necessary in order for multiculturalism to flourish. In its dotage, our generation will believe that we were the ones who solved humanity’s ethnic conundrum, striking the ideal balance between diversity and uniformity, and it’s been all downhill ever since.

Posted in Belonging to the emperor | Leave a comment

Basic Training

Whence basic? It began as black slang meaning “trashy”, “low-class” or “contemptible”, and–in its full form, “basic bitch”–a woman who has these characteristics.

There is a hint of delusion about the basic bitch—she thinks she’s better than she is—but this is drowned out by general disdain. For my money, basic bitch isn’t much of a dis. It gets by on alliteration and a dash of misogyny without saying much more than “I don’t like you”. As insults go, it’s pretty basic.

As the term moved beyond its community of coinage, the notion of self-overestimation was amplified. This comes across in Kreayshawn’s indelible 2011 earworm, “Gucci Gucci”. The song is mostly about bad bitches—women who effortlessly occupy the top of the Great Chain of Sexily Insouciant Being—but they exist in contrast to basic bitches, who yearn to breathe that rarefied air but just don’t have what it takes. These bitches try to hide their basicness beneath name brands—Gucci, Gucci, Louie, Louie, Fendi, Fendi, Prada—but Kreayshawn sees right through it. Her contempt is not for luxury goods per se, but rather the idea that style can be purchased. She is a full-blooded aristocrat, peering down her nose at the arrivistes.

“Basic” mutated again when it moved from slang and pop music into the Zeitgeist at large. It became explicitly gendered but lost its connotation of trashiness. A basic bitch was now a middle class woman with cloyingly average tastes: Ugg boots, Sex and the City, and Pumpkin Spice lattes. Her delusion is that these tastes make her interesting, when in fact they are painfully trite.

It’s the Pumpkin Spice lattes that do it. Why, among all the flavors on offer at Starbucks, is Pumpkin Spice the one that brings a stereotype to life? A generation ago the punchline would have been simply “latte”, and it would have been directed against pretentious strivers in general instead of this one particular kind of of female striver who, hilariously, doesn’t even realize she’s striving. This detail is at once preposterously specific and immediately recognizable. It’s poetry.

Imagine explaining to the proverbial anthropologist from Mars—or just your grandmother—why “Pumpkin Spice” evokes a whole risible world. You can’t. Or actually, you can, if you’re willing to trundle out the postmodern heavy artillery, but the difficulty is an indication that we’re operating at a high symbolic altitude, playing a game in which the delight comes from picking up on specific cultural references. Kreayshawn’s aristocratic disdain is still in effect here. Basic joins other terms of contempt—inauthentic, uncool, consumerist—whose essence lies in their inscrutability. The trappings of basic bitchness cannot be inherently contemptible. They have to be pleasant enough to win aficionados. Otherwise no woman would want them, and there’d be no basic bitches to mock.

Posted in Mermaids | Leave a comment

Puerto Ricans Have Relatively Few Words for Snow

Irish has no standalone words for “yes” and “no”. To give an affirmative response to a question you repeat the verb, and to give a negative response you repeat the verb with a negation particle. Much the same situation obtains in Chinese. This may be why the Irish and Chinese are both so squirrely about answering direct questions.

French grammar divides all nouns into “masculine” and “feminine” genders, as does Arabic, which explains why the French and the Saudis have such famously similar attitudes towards sex.

Mandarin, Cantonese, Taiwanese etc. are tonal languages–variations in vocal pitch are used to distinguish between vocabulary items. This is also true of Bantu languages. This may be a reason why the music of China and sub-Saharan Africa sounds so much alike.

Portuguese has a subjunctive mood, so the notion of hypothetical action is represented as a verb ending, woven into the very fabric of Portuguese words. Contrast this with English, which to impart the same quality of irrealis to a clause must prepend a whole other helping verb like “may” or “might”. Is it any wonder that Brazilians are so much better than Australians at reasoning about hypothetical situations?

German has a lot of harsh consonants. It employs regimented speech elements that sound like machine parts grinding away with well-engineered if somewhat forbidding precision. Italian, on the other hand, has a lot of fluid, expressive, romantic vowel sounds. This may be part of the reason why Germans are so rigid and Italians are so emotional.

Where a white American English speaker would say “He works at the post office” many black American English speakers would say “He be working at the post office”, the extraneous, unconjugated helping verb “be” adding nothing to the essential meaning. Their insistence on using more words than necessary is an indication that blacks are less intelligent than whites.

Where a white American English speaker would say “She is tall” many black American English speakers would say, “She tall”, omitting “is”, the sentence’s grammatical keystone. Their insistence on using fewer words than necessary is an indication that blacks are less intelligent than whites.

The Chinese writing system is not an alphabet in which a small set of letters is mapped to discrete speech sounds which must then be strung together like beads in order to have meaning. Chinese employs a rich vocabulary of many thousands of written characters, each one representing a fully-formed concept. A single character may express an extremely abstract idea, and its meaning may play off that of other characters in endlessly subtle ways. This writing system gives rise to a world view which is holistic rather than atomizing, which is why traditional Chinese medicine treats the human body as an organic whole instead of adopting western allopathic medicine’s approach of breaking it down into a set of specific disease symptoms that must be treated by specific drugs, and also why even second-generation Chinese-Americans are notoriously bad at math.

English has no single word for a short legless garment a woman wears about her waist while playing tennis, which is why if you say the phrase “tennis skirt” to an English speaker, they will have no idea what you are talking about.

Posted in Mermaids | Leave a comment

A Monday Kind of Love

The girl walked down St. Mark’s Place, yoga mat under her arm, sweating from the summer heat. She shouldered her way through crowds and darted across the street, dodging taxis. She took a sip from her Nalgene water bottle and contemplated which gluten-free pizza chain she’d stop at.

She turned the corner onto a quiet side street where a man sat in a plastic lawn chair. Late 50s, early 60s, lined face, and rock and roll T-shirt. As she walked past, she could feel him glaring at her. He started talking, in a normal tone of voice, but clearly meant for her to hear.

“This place used to mean something,” he growled. “Punk rock. Graffiti. Trash in the streets. It was real. Real. CBGB. Debbie Fucking Harry.” His voice trailed off into almost a sob, then she felt the laser of his contempt refocus on the back of her neck. “You weren’t there. You could never understand. You just weren’t there.”

He couldn’t see her smile, and she didn’t realize she was smiling. God bless this city, where even hostility from random old guys put a spring in your step. A blast of hot air from a subway entrance caught her square in the face, bringing out fresh beads of sweat just below her hairline. She was nineteen years old and living in New York City and the world was hers.

Posted in Those that at a distance resemble flies | Leave a comment