The joys of analytic philosophy: A paradox

No comment really required, except that I found this amusing.

Posted in miscellany | Leave a comment

Friday Music: Thisbe’s Choice

In honor of my daughter’s first birthday,  I’m posting one of Thisbe‘s favorite performers, Aretha Franklin.  This is a nice performance of Rock Steady (from Soul Train, circa 1971):

The whole time Jennifer was pregnant, we had a running joke that the baby loved listening to Aretha Franklin.  We had observed that the foetus “danced” whenever we played Aretha, but this was probably due to some sort of observer bias.

Jennifer’s labor lasted about 20 hours; somewhere around hour 17 I thought it might help draw the baby out if we played my iPod mix of Aretha.  Pretty soon, everybody in the labor and delivery room (except Jennifer and me) was dancing to Aretha.  Sure enough, around hour 20, Thisbe arrived.  Aretha stayed on during the examination and clean up.

All of this would be nothing except that Aretha Franklin now has a weird hypnotic effect on Thisbe.  A particular favorite is Chain of Fools:

Chain of Fools has been known to stop even her worst meltdowns, the ones that threaten to go nuclear.  Jennifer would teach her seminar on media in Africa Wednesday evenings, and I was left on Thisbe watch.  Thisbe early on decided that the bottle was not for her, so I usually just have to wait out her inevitable Wednesday evening meltdown as stoically as possible.  Thank goodness for Chain of Fools!  The opening swamp guitar riff would immediately transfix her and Aretha singing “chain, chain, chain…” would send her.  I don’t know what I would have done without Chain of Fools, Rock Steady and Think:

I keep hoping that Ms. Franklin will record some of Thisbe’s favorite songs, like the ABC song or Six Little Ducks.

Posted in music | Tagged , , , | Leave a comment

Cheap Talk

Game theorists spend a lot of time thinking about signaling in situations where there is asymmetric information.  These are cases where one player knows more than another player; the first player may want to signal something (for strategic advantage) or the second player may want to find out what the first player knows. This is exactly the stuff of linguistic pragmatics, so there are good reasons for linguists to think about information asymmetries and signaling.  In particular, I want to think about cheap talk systems.

In a cheap talk system, it’s assumed that talk really is cheap: it costs little or nothing to send a signal.  The effects of the signal depend on the strategic properties of the game under consideration.  Consider a simple coordination game.  Suppose we’ve decided to meet up. You don’t remember whether we decided to meet at a pub or a coffeehouse. You do know that we both prefer the pub to the coffeehouse and we prefer either of those to not meeting up at all.  So here’s the game matrix:


A simple coordination game


Notice that both (pub,pub) and (coffeehouse,coffeehouse) are Nash equilibria; if we coordinate we have no reason to unilaterally defect.  Both of us choosing the pub is Pareto efficient: it gives us the best possible payoffs.

I dropped my cell phone and broke it and you know that I don’t check my email very often but you can send a message to me in the hopes of meeting.   You decide to send an email: “I’m going to the pub for our meeting.”  Suppose I get the email.  I should obviously go to the pub, since you have no reason to mislead me and every reason to tell me the truth.  Therefore, (pub,pub) is the cheap talk equilibrium of the game.  Notice that our interests coincide in this game.

Now suppose we’re playing a Prisoner’s Dilemma game:


A Prisoner's Dilemma Game


In this game, if I squeal and you stay silent, I go free and you go to jail for a long term (the situation is reversed if you squeal and I stay silent).  We both like going free so that gets the highest utility.  We both hate long jail terms so that gets the utility zero.  If we both stay silent, we both get a very short jail term based on another crime. We like that better than a long term but not as much as going free; I gave a short jail term a utility of 2.   Finally, if we both squeal, we get medium jail terms, which is better than a long term but not as good as a short jail term; it gets a utility of 1.

Notice that the action of staying silent is dominated by squeal.  If you’re silent, I can defect and squeal.  This gets me out of jail free.  If you squeal, then I had better squeal too; otherwise, I go to jail for a very long time, my least favorite outcome.

There’s an intuition, though, that we both ought to stay silent.  This gets us both a payoff of 2, which is better than the payoff of 1 we get if we squeal.  But because of the temptation payoff, (silent,silent) is not a Nash equilibrium.  We’re both drawn to the temptation payoff we get if we squeal and the other player is silent.  So we wind up with the mediocre payoff we get under (squeal, squeal).

Now, suppose I can send you the message “I’m going to stay silent.”  Should you believe me?  Well, talk is cheap so you’re well advised to assume that I’m trying to trick you.  If you believe me and stay silent, then I have every incentive to squeal; you’d better squeal too, if you want to avoid a long jail term.  If you think I’m sincere, then you ought to squeal, since you’ll get your preferred outcome.  In any event you should just squeal.  So my message has no real effect on your behavior; there’s no cheap talk equilibrium.  Rather there’s a “babbling equilibrium.”  I might just have well said “ooga booga” for all the good it did!

Now you might think that I should keep my word; but this assumes that there are consequences for my duplicity: That’s a different game.  In this instance, we can assume that if I squeal and go free, I can just fade into the woodwork (or the witness protection program) and go on with my life, no consequences for my betrayal.

Linguists find the cheap talk framework a bit peculiar; of course we should take the costs of speech production into account!  Suppose, for example, that I have to choose between using a pronoun like “it” and a full description like “the pen on the table over there.”  Of course, producing the pronoun is relatively inexpensive so if I think you’ll interpret it correctly, I’m likely to use it.  It costs less effort for me to produce it than the relatively long description.  If, however, I think you won’t interpret the pronoun correctly, then I’ll go to the effort of producing the long description; this has higher production costs—I have to decide what description to use and then actually produce it, which is work—but it’s at least likely to have the desired effect.

The fact that we use pronouns at all shows that considerations of production costs do impact our linguistic behavior.  So talk isn’t always cheap.

Notice that there’s an effect of scale here.  At the micro-level of speech production, considerations of production costs have an impact on our behavior.  At the macro-level of influencing your strategic decision-making, then talk is cheap:  I’ll certainly go ahead and produce the signal if I think it’s to my strategic advantage to do so.

Sometimes, of course, talk isn’t cheap even at the macro-level.  Suppose that I develop a reputation as deceitful in a population.  Other members of the population assume that I’m a liar and are less likely to cooperate with me.  They might even refuse to play.  In that case, talk has damaged my social standing, my “face” in the terms of Erving Goffman, and I become less capable of getting things done socially.  Then talk certainly isn’t cheap.

I’ll almost certainly be returning to face in other posts; I think it’s an important principle organizing a lot of our linguistic behavior.

Posted in game theory, linguistic meaning | Tagged , , , , , | 1 Comment

Entrenchment: “Befriend” vs. “friend”

In “Old words, new meanings,” I talked a little about how words can come to occupy semantic niches; really, this is just the question of how words can come to express the meanings that they express.  The idea is that this is a microeconomics phenomenon; populations give value to words by using them to mean something, so meaning is a kind of emergent property of language use in a population.

Of course there’s more—a lot more—that needs to be said here, but I want to focus on the problem of “entrenchment” that I raised in the earlier post.  Entrenchment is a complex phenomenon, one that helps fix the language as a social medium, and I think you can look at it from a number of different perspectives.  Right now, I want to think about the choice that speakers had when trying to express the new social relationship, establish a Facebook friendship.

We can assume that, at some point, speakers were faced with a choice between at least two forms to express the meaning establish a Facebook friendship: They could verb a noun—that is, they could use the noun friend as a verb—-or they could extend the old form befriend into a new semantic domain, one that is presumably adjacent to its fixed meaning.

Notice that friend as a noun doesn’t express the right relationship; a Facebook friend is not necessarily a friend in real life.  But the meaning that friend as a verb would express (had it existed) is already expressed by the old form, befriend.

At this point, strategic reasoning can take over.  The speaker can reason that if she uses befriend, then the odds are that the hearer will take her as meaning the old concept of become friends which is not what she intends to signal.  In other words, befriend is entrenched in that meaning.

If the speaker uses friend, however, the hearer can infer that she must not have meant become friends; had she meant that, then she would have been more likely to use befriend (although there is always the possibility of error—perhaps she couldn’t “find” befriend in her mental lexicon).  Friend as a verb does not have an entrenched meaning, but its noun use is in a nearby semantic niche, so (presumably) friend as a verb can now occupy the new semantic niche of establish a Facebook friendship.

As I mentioned, there’s another perspective to take here.  Befriend had to become entrenched in the sense of make friends.  In particular, it had to occupy the make friends niche with such tenacity that friend as a verb could not invade that niche and drive it out.  It has to be the case that, originally, befriend was the grammatically regular way creating a form that meant become friends; verbing the noun friend was blocked.  As the use of the prefix be- declined, it would have to be that the frequency of befriend remained high enough  to drive out competitors.

This is one of the other senses of “entrenchment.” It’s very similar to the sense of entrenchment that Nelson Goodman was talking about in his book Fact, Fiction and Forecast.  Once a term is entrenched in a semantic niche, it becomes difficult for a new form to drive it out of the niche.

Notice, though, that strategic reasoning lies at the heart of the whole process.

Posted in linguistic meaning | Tagged , , , , , , , , , , | 1 Comment

Stephen Fry on language

A nice oration by Stephen Fry, illuminated with kinetic typography.

Posted in miscellany | 1 Comment

Old words, new meanings

In 2009, the Oxford English Dictionary made “unfriend” the word of the year.    The word is so new, in fact, that my electronic dictionary doesn’t have it.  The entry it gives for “unfriended” is:

without friends : murder left innocent people bereft and unfriended.

It further advises that this use is “poetic.”  The spell checker I’m using now doesn’t even have it; “unfriend” is underlined as misspelled.

Of course, the above sense is not the one intended by the Oxford English Dictionary.  As everyone knows, “unfriend” comes from Facebook, the name of a website that is now its own verb:  Ads on the subway implore us to Facebook us.  Facebook makes available to its users the possibility of linking to another user; the linking relationship was, of course, one that involved mutual consent.  The link established a “Facebook friend.”

A Facebook friend is not the same as a real friend; we might be Facebook friends but not friends in real life.  Notice we still maintain a distinction between the real and the virtual.  Hence, a new social category was born: the Facebook friend.

With the new social category came the need for a new word to express that category, and, in particular, the need for a way to express the establishment of the Facebook friend relation.  A new “semantic niche” had been created and, just as in ecology, something was going to come and consume the new resource.  The language already has “befriend” to denote the establishment of a real (not virtual) friendship.

So people started using “friend” as a verb with the special meaning “establish a reciprocal Facebook link with another person.”  A glance at Google’s timeline for the word “friending” shows the explosion in its use:

A timeline for the new verb "friending"

The growth of "friending"

The timeline is, of course, a histogram; but it suffices to show how “friending” has exploded.

There’s an interesting microeconomic note here.  The growth of “friend” is possible, first, because there is a new semantic niche that was not occupied but, second, because speakers started coordinating around this new meaning.  If someone said, “I friended Joe,” I knew what they meant and I knew, in particular, that they hadn’t befriended Joe.  The whole population coordinated around the new usage; this sort of mutual tacit coordination is the very stuff of Thomas Schelling‘s work:  See is book Micromotives and Macrobehavior for lots of examples.

Going back to “befriend,” notice that this word is entrenched in the meaning “become friends,” a semantic niche that it has occupied for years.  The new “friend” has no hope of displacing “befriend” in this niche.  Interestingly, “befriend” didn’t move in to occupy the “Facebook friend” niche; instead, we did what we so often do: We verbed an existing noun.

New technologies are opening up a way of studying this process.  First, it’s making new things available that we need words for: iPods and other mp3 players, smartphones, text messaging (“texting” which figures in Meaningful Games) and so on.  But it’s also giving us new ways of studying the process.  We can actually watch words spread into semantic niches and, sometimes, compete with each other for priority in the niche.  A new field—social ecology—is possible.

Back to “unfriend.”  It happens that “unfriended” already has a kind of weak existence, as my electronic dictionary shows.  The new social relationship, Facebook friend, is often transitory.  Deleting a Facebook friendship is not nearly as weighty an action as breaking a friendship!  We needed a new word for dissolving a Facebook friendship and, voila!, “unfriend” immediately suggests itself due to regular English morphology; its weak, unentrenched meaning is now gone (if it was ever there) and the new use caused the frequency of “unfriend” to explode:

timeline for unfriend

A timeline for the frequency of "unfriend"

Notice that “unfriend” has a long history—it has uses back to 1650—but it’s a sparse one until recently.

Another note:  There is another morphological form possible, “defriend.”  But my intuition is that this would be wrong.  If I defriend someone, I would have to eliminate all his Facebook friends.  Thus, “I unfriended Joe” means I removed him from my Facebook friends but “I defriended Joe” means that I somehow got access to his Facebook account and removed all his Facebook friends.  Compare “defriend” with “deforest.”

The grammar has a place here: it makes forms available. But the workhorse here is the social decision-making that enlivens the forms and endows them with real meaning.

As a footnote, Creeper the Hamburger Pimp has his own Facebook page.  I’d friend him, but I’m not on Facebook.

Posted in linguistic meaning | Tagged , , , , , , , , | 3 Comments

The Hamburger Pimp

From Rudy Ray Moore‘s masterpiece, Dolemite.  What more can I say? (Except I kick my own ass twice a day as well!)  Dave Embick pointed me to the Wikipedia entry for Dolemite, which says that the Hamburger Pimp was played by Vainus Rackstraw, aka “Creeper.”  The Internet Movie Database gives Dolemite as is only credit, alas.  (Is that a microphone in the bottom right of the shot at the beginning?)

Posted in blaxploitation | Tagged , | Leave a comment