Anecdota

Laughter is the Best Medicine

Robert Mankoff: “There is no Algorithm for Humor” | Talks at Google


UDI MANBER: Today
we have as a guest Bob Mankoff, who is the cartoon
editor of “The New Yorker,” has been the cartoon
editor of “The New Yorker” for over 20 years now. ROBERT MANKOFF: 1997. He can’t even do the math. This is the guy who
used to head Search. UDI MANBER: OK. I think I’d better let
him do the talking. ROBERT MANKOFF: All right,
I’m ready, thank you Udi. Thank you. Great to be here. I have a question,
is this noticeable? Actually, the reason
I’ve been wearing this– and I’ve been wearing it
around “The New Yorker” also– is that when I
actually wear Glass, it won’t be as obnoxious. [LAUGHTER] ROBERT MANKOFF:
Anyway, I’m going to talk a little
bit about my book, because I’m here
to sell my book. I’m running around
California doing that. “How About Never–Is Never Good
For You?: My Life In Cartoons”. But I’m also going to talk
about computers and humor, and the title of the
talk– well, no, oh, I’ve got to remember
what you guys are going to get if
you buy the book. Hey, GoPro, watch this. Is that exciting? Forget that wing gliding
stuff, this is dangerous. I got a paper cut doing this. OK, that’s what
you’re going to get. You buy the book,
whatever, you get that. And what I think I want to say
is if that little drawing I did makes the book more
valuable, I’d like it back. [LAUGHTER] OK, come on. Go, go, go. Oh yeah, this is the book. The thing I want
to say about this is there’s going to be jokes
and information in this lecture. The jokes will be
the information, and the information
will be the joke. Think about it. So that’s my cartoon. You see, that’s how old I am. It’s my birthday today, I’m 70. [APPLAUSE] ROBERT MANKOFF: That’s
great, just getting a hand for being 70, that’s wonderful. You know, they say 70 is
the new 50, and that’s good except that dead
isn’t the new alive. That’s the bad part of that. But anyway, I got in the
“Yale Book of Quotations”, businessman, Thursday’s out,
how about never–is never good for you? Next to that, other well-known
humorists, Mao Zedong. OK, so that is the title. There is no algorithm,
you need a brain. Now the brain doesn’t
have to be that big. It never ceases to amaze me
what little brains people have. But you need some
sort of brain, I’m going contend, to actually
do real human humor. Now I was up online,
and I watched Watson beat Ken Jennings, and
so I have this little clip. Let me see, make sure. [VIDEO PLAYBACK] -William Wilkinson’s “An
Account of the Principalities of Wallachia and
Moldavia” inspired this author’s most famous novel. 30 seconds, players, good luck. [JEOPARDY THEME MUSIC] -Over to Ken Jennings
now, $18,200 going in. Bram Stoker is what
we’re looking for. And we find– who is Stoker? I, for one, welcome our
new computer overlords. [END VIDEO PLAYBACK] ROBERT MANKOFF:
Well there, he did something Watson couldn’t
do, he made a joke. He was funny, right,
in that situation. Now, there are computer
programs that do create humor. This is called Standup, and
that looks pretty impressive. Wow, a lot of stuff
going on there, a lot of processing stuff. Really good. But the jokes aren’t that good. Why is an alone region different
from a sole habitation? One is a lonely air, the
other is an only lair. Thank you. [LAUGHTER] ROBERT MANKOFF: For real humor,
you need more than a brain. You need a body. Actually, I got this mixed
up, but you need a body like that, not like that. Because, see,
that’s a funny body. Humor must always deflate, and
when I say you need a body, you need a body because you
need to be a human being. Because humor is more
than just cognition. But even if you had robots–
this a great cartoon. If I can’t feel love,
what was the point of making me so
damn good looking. Well, of course, the
robots and the computers can’t feel anything. Now humor is an emotion. It’s a very, very
strong emotion. It’s an ecstatic promotion. [VIDEO PLAYBACK] [LAUGHTER] [END VIDEO PLAYBACK] ROBERT MANKOFF: We don’t
know what she’s laughing at, but it wasn’t a computer joke. That’s my first
wife, incidentally. You know, it’s funny
though, because this thing that we think is a good thing–
a little parenthetical point I want to make– was not always
thought that way at all. This is Lord Chesterfield
in a letter to his son, how to conduct himself. Frequent and loud laughter
is the characteristic of folly and ill manners. How low and unbecoming
a thing laughter is, not to mention the
disagreeable noise that it makes, and the
shocking distortion of the face that is occasions. I have never laughed in my life. It’s interesting,
our view of laughter and the importance
of it has changed. I can show you
that by, well, this is Plato and Aristotle
and Hobbes’ views of it. Mixture of pleasure
and pain that lies in the malice of amusement. Aristotle, comedy is
an imitation of people worse than themselves. Hobbes is all about
superiority theory. You’ll be happy to see this. It’s a Google Ngram. Nobody even had a sense
of humor until about 1870. Isn’t that amazing? It’s actually quite a new thing. It actually is a new thing. Before that, based
on what I showed you about Aristotle and
Plato, views of humor were very, very negative. Why? Because the thing that we
call the sense of humor, they said the thing was
called either a sense of the ridiculous or
a sense of ridicule. Because that’s one of the
primary function of humor. There’s wonderful functions
of humor, but when you think, well, why is it
in every culture, everywhere, and all the time? What does it do? What is its function? Why is it everywhere? You ask, what is it that keeps
everybody doing– conforming, basically. Not at Google, but
everywhere else. What is it? You don’t want to be
embarrassed, right? You don’t want to have that
toilet paper on your shoe, you don’t want to have
your fly open, right. We know what it’s like to
be in elementary school and high school. What’s the policeman
of embarrassment? Laughter. So one of the functions of
laughter– a harsh function, a conforming
function– is ridicule. People say laughter
is contagious. Well, that laughter
was contagious, but it’s not contagious
when it’s behind your back. So that’s one of the reasons
it wasn’t looked on favorably. Now it’s very different,
because a sense of humor came to mean something else. What a sense of
humor came to mean– it’s easier to see by what you
say if someone doesn’t have it. Why do we say someone doesn’t
have a sense of humor? It’s more than they’re
not funny, right. They have rigid thinking. They have no perspective
on themselves. The transformation is
why it became important, and why it’s very difficult
for computers to have it. Because the sense of humor,
which originally was something that was directed out to
something deformed, something ugly that you didn’t
have empathy for, got directed back towards ourselves
as something to be perspective. That’s why it’s valued, and
that’s part of its function, now, also. What does evolution
tell us about humor? That’s a little cartoon of mine. [VIDEO PLAYBACK] -As we have listened
to animals playing– ROBERT MANKOFF:
It’s a type of play. -We have heard what appear
to be the sounds of laughter. And we studied these
for a couple of years without quite understanding
that this might be laughter. And then one day, we decided
to tickle some animals. We realized that we had to
look at the sounds at a very different register
than we can hear. So we obtained these
transducers that are called bat detectors that
can bring very high frequencies down to our auditory range. When we did this
and we listened in, we could tickle
animals and generate a lot of vocal activity
that appear to be laughter. These animals would begin
to enjoy our company, and they would start
to play with our hands, and wherever we would put our
hands, they would follow it. [CHIRPING NOISES] -And when we tested
these animals to ask whether they were
enjoying this kind of activity, the unambiguous answer was yes. [END VIDEO PLAYBACK] ROBERT MANKOFF: An
interesting result of that study is that
he has talking rats. I tried that in the
subway and it bit me, so not always the case. But one of the
things it shows you, it shows how deep
humor is an emotion. When you look at humor, overall,
you’ll see, first of all, it’s a social phenomenon. Whatever that woman was
doing, she was not alone. It’s a social
phenomenon, there’s something cognitive
that stimulates it. There’s this emotion,
there’s a feeling that happens before you laugh. So you could have people
judge comedy films and instruct them not
to laugh, and they still will be able to know that it’s
funny, and feel the funny. And then there’s this
expression in laughter, which is a type
of communication. OK. [VIDEO PLAYBACK] ROBERT MANKOFF:
So the same thing happens with chimps
as we go further up. -It’s coming. It’s going to get you. Here it comes, Alfie. Here it comes, Alfie. It’s going to get you. [END VIDEO PLAYBACK] ROBERT MANKOFF: OK. But this is a little
bit higher up, yet. This shows you a
lot about humor. This is a little
game of peekaboo. It’s not just tickling. It’s not tickling, and this is
the play face that’s happening, this is the play face
that’s happening. So we see, when
looking at this, you see that humor is
very, very deep, very, very, very deep in us. Very deep as a feeling, and
very deep as a function. So it’s a form of communication. What it communicates
is that we’re playing. Humor is a form of play. We’re always really
going between two modes. One is sort of the serious,
what’s called a telec mode, purposeful. OK, we want things to be done. And the other is the play mode. When we’re in a purposeful mode,
we actually like to be relaxed. We don’t like to be
anxious, we don’t like to be over stimulated. When we’re in a play mode,
exactly the opposite exists. We want to be excited. Look at this video. [VIDEO PLAYBACK] [SHOUTING] -Here we go. [LAUGHTER] [END VIDEO PLAYBACK] ROBERT MANKOFF: Well, that’s
interesting, he laughs, doesn’t he. So the other thing
that shows you is that humor is an emotional
regulation mechanism. Partly it’s based on
surprise, but it’s the surprise we expect. That’s what happened,
the surprise we expect. Surprise is always
bad for an organism. You should never be surprised
from an evolutionary point of view. I see all these looks, brow
looks furrowed because he said, he’s going to be funny. Don’t worry, just give me time. So surprise is always bad
because you should predict. So originally, when
you’re surprised, whether it’s by a
joke or whatever, there’s a slight
negative valence. You should know
what’s happening. Surprise can be met by
three different things. Fear, obviously,
awe, or laughter. All of them are
respiratory responses preparing the organism
for something. In the simplest
cases of laughter, you have surprise,
which is always bad, followed by something trivial. That’s just relief laughter,
you thought something bad. Jokes and everything
build on this. Here’s interesting–
this is my contract, because now I want to
get to a little bit of the cognitive part–
here’s my contract. They have this in the
book from “The New Yorker” when they gave me a
contract to be a cartoonist. I’ve blurred out all the
parts about the liquor, and the women, and
the money, because I didn’t think that
was your business. But look at the contract. This is the contract
for me as a cartoonist. It doesn’t have the
word cartoon anywhere. It says idea drawing,
for any idea drawing we accept from you
during this period, with respect to idea drawing. So that’s the other
part of humor, especially for “The New
Yorker,” that there is an idea. People ask me how I get my
ideas, and I glibly say, I think of them. [LAUGHTER] ROBERT MANKOFF: Here’s
some ideas I thought of. There is no justice
in the world, there is some justice in the
world, the world is just. It’s called what
lemmings believe. [LAUGHTER] ROBERT MANKOFF: Now, there
are just straight gags, and then there are
gags– and this is part of what I write
about in the book– that have some other value because
there’s some other idea. There’s an idea over
and above the idea that is happening in the joke. So the joke works just on
the levels of lemmings, but for me, this
is a broader joke. There’s another
meaning to it, which is about belief in general. For me, it’s a belief
about religion, and it’s my
expression of the idea that all the arguments–
horrible arguments– between religions are about who
has the most imaginary friends. OK, this is actually
a little diagram– I meant best imaginary
friend– here’s a diagram of how you
actually get ideas, which is a mash up of
different concepts. I was aked by Psychology Today
to show how you get an idea. So I said, OK, I’m going to
do an idea about the wheel. So originally what I did
was I just drew a wheel. I drew a wheel, I
drew a square wheel. And wheel is invention,
fire is invention. When I saw the
square wheel I said, oh, that’s like a bad thing. That could be a
clunker or a car. Trade in for a new
model, I wrote. Wheel, transportation,
car, bicycle, what. Then I came up with a
cartoon in which I said, the back part I call the wheel,
the front part, the brake. Then I didn’t know what I would
do, wheel on fire, invention. I didn’t know. That’s just a mash up,
that’s going to be an idea. That’s an incongruity
that will be an idea. Then, if there was a real
wheel, it could be a car. Dad, can I borrow
the wheel tonight. The interesting thing
about these ideas is that they don’t use up. They create other ideas. There are actually an infinite
number of ideas like this. Here’s one that we had
in “The New Yorker.” Yeah, yeah, and I
invented the ticket. And now we’ve taken it
completely out of that. So one of the things
that’s happening here is that you do have the basic
mechanism of surprise working. It works very, very fast. It works very fast because
you see the image which is incongruous, and then
you put this thing together. Built on the tiny
little surprise is cognitive thrill or pleasure. One of the things is,
when many things happen at the same time– my
belief anyway, what’s happening in humor– is
you can’t parse them out. So at the same moment
that the surprise– so if I tell a joke,
OK, so I’ll tell a joke. Here’s a joke. So a guy is walking
in a field and he sees a farmer walking a pig, and
the pig has an artificial leg. Now, in your minds, you’re
already a little worried. So this is good. I’m deconstructing the
joke a little bit for you. E.B. White said analyzing humor
is like dissecting a frog. Nobody’s much interested
and the frog dies. Well I’ll kill a frog. OK, so now the guy
is walking a pig, and the pig’s got
an artificial leg. Now you’re a little worried. Not really, because
you’re smart, you’re from Google, like
what’s going on here. I sure hope I get this. [LAUGHTER] ROBERT MANKOFF: Because
really, come on. OK, so he asks the
farmer, he says I’ve never seen a pig
with an artificial leg. He said, you see that house
up there, way up there? There was a fire in that house. This pig came up, he
saved me, then he went up and he saved my wife
and my two children. He said, so he burned his
leg in the fire, right? He said, no, no, no. You don’t eat a pig
like that all at once. [LAUGHTER] Oh, that’s horrible. OK, so all of a sudden, the
narrative completely changed, and then you put
the things together. Whether or not you
like that or don’t like that will depend
on many things. Your disposition, your
feelings about animal rights, a million other things. These are all enhancers or
inhibitors that don’t really have anything to
do with the joke. If they get enhanced
or inhibited enough they can inhibit it. If you have enough
enhancers, you could make anyone really
laugh at anything. But the point is, there’s
a narrative switch which is the surprise and
then the pleasure. And the pleasure,
then, could also be the pleasure in your own
sadistic and cruel impulses towards that pig. George Orwell said jokes
don’t degrade us, they just tell us we are degraded. OK. No, Thursday’s out. How about never–is
never good for you. That’s the cartoon,
that’s the cartoon that put my daughter through
college and everything. It’s great. Actually, it’s OK. One of the things they say ask
about it is, why is that part? Why can’t I just end it? Why can’t I just end
it with no, Thursday’s out, how about never. Hands? AUDIENCE: Because
that just sounds mean. ROBERT MANKOFF: What’s that? AUDIENCE: It just sounds mean. ROBERT MANKOFF: It
just sounds mean? Well, what do you mean? Why is it not mean, because is
never good for you makes it? [LAUGHTER] ROBERT MANKOFF: I
mean, is never good for you rubs it in, doesn’t it? AUDIENCE: But people
say that, right? How about 5:00, is
5:00 good for you? ROBERT MANKOFF: Exactly. Well, you’re onto something. You did more than
just eat that salad. [LAUGHTER] ROBERT MANKOFF:
What it is is this. Arthur Koestler, in a book,
“The Act of Creation”, talked about the similarity
between science, humor, and art, in that they
involved creating things from different matrices,
different associations. And humor certainly does that. What this is doing
is mashing up, but the mash up here is
between a message that is absolutely rude with the
syntax continuing to be polite. So that’s what you
were onto, really. That’s why is never good for
you continues the politeness. So what happens is
a cognitive synergy. At the both time we’re
experiencing rudeness and politeness, and
I think that’s what gives it its little frisson. Humor needs a target. How’s my GoPro, all right. So like I said, it
can be a little cruel. We’ve all felt like that. Now look at this cartoon. [LAUGHTER] ROBERT MANKOFF: So that’s
fairly benign, you would think. But is it benign– once
again, I want to point out the dispositional
nature of humor, which means your attitudes and
your opinions will determine the context of humor
and how you react to it. So that seems fairly benign,
but let’s look at the response. Email comment on this cartoon. Another joke on old white male. Ha ha, the wit. It’s nice, I’m sure
to be young and rude, but someday you’ll be old,
unless you drop dead as I wish. [LAUGHTER] ROBERT MANKOFF: So,
cordially, Robert Byron. Actually, he’s making a joke. You see, he’s mashing up, right. He himself is mashing up. He actually is
being funny himself, because he’s saying cordially,
Robert Byron, after something is completely non-cordial. OK. Discouraging data on
the antidepressant. [LAUGHTER] ROBERT MANKOFF: So I
like that little pause. That was a good laugh,
I want that on record. Because we had the regular
laugh, the real laugh, the emotional laugh, ha ha
ha, then we had the ooh. Which says I drive
a Prius, I’m good, I believe in global
warming, you wouldn’t believe the organic
foods, everything. Honestly, that first
part was a mistake. Something I regret. [LAUGHTER] ROBERT MANKOFF: This is
a survey I did online to see– this shows
you, actually, the real diversity of opinion. So people could rate it to
10, and 109 people liked it. So really, the majority of
people liked, just about. Or a lot, anyway. But look at this person
who did not like it. They voted only three
and said I like animals. Look how much they liked them. One, two. You can’t like animals more
than five exclamation points. If you do, it’s unhealthy. Look at this person. I don’t like to see animals
suffer, even in cartoons. So I am solicitous
to these people. I wrote back and I
said they don’t suffer. We use anesthetic ink. [LAUGHTER] ROBERT MANKOFF: Seems just
like the results of my– so, this is actually,
when humor is not affected by contagion and
the contagion of laughter, the diversity of
opinion is very great. But it depends on what
you’re– actually, things that you feel
really strongly about, you usually will
not think is funny. When you have really,
really strong beliefs, then humor, to you, seems
like some sort of blasphemy. OK. Humor is a type
of entertainment. This is a type of
entertainment, too. The tiger in a cage. What’s exciting about it
is, ooh, he could kill us. What’s good about
it is no, he can’t. It’s just like the guy
in the roller coaster. You’re safe, but in danger,
safe, but in danger, safe, but in danger. Wow, that’s fun. That’s sort of fun, I guess. That’s not that much fun. That’s sort of a bad zoo. But there are worse zoos. [LAUGHTER] ROBERT MANKOFF: So
that’s the line. When you look at humor
within context, and this would be the difficulty for
computers, because not just in the creation of it, but
in the evaluation of it. To know– for my job– what’s
right and wrong for “The New Yorker.” Look at this cartoon. Oh, I like that. We got the oh before. That’s definitely– if you got
the oh before for that one, how about this. Oh, that is sad. Who did you think was
going up the ramp? But this is the point
of the titration of humor, how you
move it around. One of the ways to look at
humor, or a lot of humor, as the enhancement and not
the essential part of it, is the right amount of wrong. Like the Las Vegas ad,
the right amount of wrong. So just enough where you
like the wrongness of it, because actually,
we want freedom from the rightness of our lives. But with this thing, by
putting the guy up there, or if I had him hanging,
it’s all terrible. One of things, of course, we
have is our caption contest. And one of the reasons
I’m here besides, I’m a good friend of Udi, he’s a
great guy, big fan of cartoons, is that I have data on
the caption contest. I have millions of entries. You are data
people, help me out. So on the caption contest,
every week there’s a winner, three finalists,
I won’t go through it, and one that you can enter. So here aliens are coming
right behind the car, right. I gave you data to
computer linguists at the University of
Michigan– I really hope you guys can
do better than this. Actually a very nice guy. Dragomir Radev and Ben King. And what they did
was, they took– I get about 5,000 to
10,000 captions– they took all the words that came in,
compared them to basically all the words on the
internet, standard words, and picked out the words of that
were unusual, were not common. And from that– I changed
the formula to what the fuck, because I didn’t understand
the first formula. [LAUGHTER] ROBERT MANKOFF: Oh, they come
up with this mapping, which showed the clusters
of words, and really the clusters of captions. And from that– and this
is like a blow up of it– and from that, you
got basically– when those 5,000 or
10,000 comes in, it’s not 5,000 or 10,000
different captions. It’s called
collective discourse, so it’s very interesting that no
one knows what other people are doing, but they doing this. But the truth is,
the computer did not create headlight, beam,
alien, UFO, tailgate. Someone had to look at these
clusters to say, do that. So it actually was not
tremendously useful to me because truthfully,
as it comes in, I can see these clusters
very, very fast. I don’t care what
planet they’re from, they can pass on the
left like everyone else. So these were sort of the three
finalists, rather ordinary. One of the things about humor
is it can be meta humor. This is a site called
What a Misunderstanding, and every single contest
it has the same caption. What a misunderstanding. Then, I’m just showing
how complicated humor is. Then there’s the
monkeys you ordered, which just simply says
what’s in the caption. There’s a UFO right behind us. [LAUGHTER] ROBERT MANKOFF: Then there’s
the anti-caption contest, which I’ll read. Back off, you mother-fucking
alien piece of shit, and go back to
your own planet you uninsured little
green cock-sucker. Which, of course,
is much funnier because of the transgression. That is a complete
enhancer joke. OK, now look at this. This is one of them. And one of the things you see
in the generational of humor and crowd sourcing, OK, this
was the winning caption. Objection, your honor,
alleged killer whale. [LAUGHTER] ROBERT MANKOFF: And in
the caption contest, I get people complaining. I can’t believe it,
someone stole my caption, I can’t believe it. I was on a plane, I wrote it on
the back of “The New Yorker,” I left it in the thing and
then someone clearly stole it. OK, no they didn’t. Hundreds of basically
the same caption come in for every
single contest. OK. Sometimes the best caption
is the very simplest. It sort of makes you stop
and think, doesn’t it. And that’s what I
hope I’ve done today, and now we can talk about humor. Thank you. [APPLAUSE] ROBERT MANKOFF: Man, this hurts. We’ll sit and talk. I’ll answer questions
and Udi will moderate. UDI MANBER: OK. Let me ask the first question
and we’ll go from there. So you mentioned
that surprise is something against our nature. Are you still surprise
by seeing cartoons, can they surprise you? ROBERT MANKOFF: I am
surprised, in other words, I am in the position of not
enjoying humor, but evaluating it. So whenever you mediate a
response– in other words, I’ll never be as surprised
as you are by the joke, I’ll always sort
of be ahead of it. I am usually not surprised, but
yet, I can evaluate the humor, so I know what’s
going to surprise you. That’s really what a
professional does in any art. That I’m aware of the audience,
so when I constructed– right now, this thing here,
what I did for you guys– when I constructed it, some
jokes were new, some old. And I had to be able
to sort of predict. And in that case, I’m
not surprised at all, because I know exactly
what I’m going to do. I know exactly what
I’m going to do. I’m going to put
this on, I know it’s going to create tension because
it’s weird, looking at me. I know that actually is the
fuel which humor can work with. But I’m not going to
be surprised by it, so I’m not really. The things I’m surprised
at are the things that you’re not
going to find funny. UDI MANBER: Interesting. So the cartoons
that make you funny, nobody here will laugh at. ROBERT MANKOFF: What’s that? Well, for the most
part, because you tend to laugh at just weird
things or a type of absurdity. So I made myself laugh for no
reason at all in the airport, crazy. I don’t know what
the guy was saying, he said please watch
your unattended luggage, it may be confiscated,
damaged, destroyed. And then I just said, or eaten. [LAUGHTER] ROBERT MANKOFF: I don’t know. It just made me laugh,
I just said, or eaten. I would die to have
him say that over that. But it’s interesting. So one of the things we talk
about is the play frame, so we’re in the play
frame and you heard that. If you had heard that
at the airport, imagine. You wouldn’t laugh,
you’d be frightened. What is loose in the airport
that is eating luggage? UDI MANBER: So you do
laugh, that’s good. Yeah, any questions? Go ahead, get a mic in back. AUDIENCE: Do you often
come across a cartoon you don’t get at all? ROBERT MANKOFF: Well, there
are different varieties of don’t get. There’s the don’t get in which
you’re not intended to get. In some way, you’re not
intended to the get Eden. In other words, it doesn’t
have closure in the same way. It doesn’t, like, oh,
oh, now I understand it. If I tell you about a cartoon
and the title is French Army Knife, and it has all
corkscrews, you understand it. You will put together, and
there’s complete closure. I understand the joke. The other thing, all of a
sudden you’ve been surprised, and it’s your enjoyment
of the absurd. And so a lot of that happens
because we definitely have cartoons like that
in “The New Yorker.” It might have a title,
Paninis of the Old West. Now, in and of itself, the
phrase Paninis of the Old West is funny. It’s just, if you were to
evaluate funny phrases for one, you’d say that’s funny. Do you say, what does it mean? Now, I can make
it mean something, because I can have a villain
on the tracks with a panini, so now it’s going to be smushed. So this is going back
to the computer thing, this is what makes humor
the difficult problem. So many different ways that
it can interact, so many ways that it interacts
because we’re human beings, because
of how surprised, how emotion, all of
these come together. So I definitely will sometimes
have a cartoon I don’t get, because I’m not seeing
the association, it’s some reference I don’t get. That happens. But most of cartoons that people
complain about not getting are because they have this
very, very loose structure. UDI MANBER: Let me ask you
another, actually, an advice. So Dave Berry was here, he
gave Search very good advice. He said what you need to do
is, every few minutes, send out messages saying, now
get back to work. Do you have some advice for us? ROBERT MANKOFF: I
think whenever you need to click Search
it should just say search your conscience. UDI MANBER: Search
your conscience. ROBERT MANKOFF: First. You know what I mean? Look at all the wrong
things you did today. UDI MANBER: OK. ROBERT MANKOFF: Then, you know. UDI MANBER: But what
would I click on? ROBERT MANKOFF: What? I’m no interface designer. UDI MANBER: OK. Good point. ROBERT MANKOFF: All of
a sudden, [INAUDIBLE]. UDI MANBER: Next question. AUDIENCE: Do
cartoonists actually do any research on what
lemmings actually look like? Or do they– ROBERT MANKOFF: Excuse me? AUDIENCE: The lemmings in your
cartoon, do you actually know– ROBERT MANKOFF: Those are
completely anatomically accurate. Not only is the
outside accurate, the inside of the lemmings. I just don’t draw the
outside of the lemming. First I construct the complete
skeleton on the lemming, then I clothe it. So actually, there is fact
checking at “The New Yorker,” so here’s a funny
story about that. I did a cartoon when
Prozac first came out, maybe the early ’90s, so it’s
based on the bluebird cliche, bluebird of happiness. So there’s a
bluebird, but he’s got a vial of pills in his mouth,
and he’s saying hi, hi, you guy, I’m the blue
bird of Prozac. So this is when
it first came out. So the fact checkers came to
me and said the bird’s too big. I said, well, it’s
hard, because you’ve got to see the pills, and stuff. Back and forth, back and forth. They fact check everything. So I called her over,
her name was Dusty. I said Dusty, I don’t want
anyone else to hear this. They can’t talk. The birds can’t talk. [LAUGHTER] ROBERT MANKOFF: So we do. Some of the things,
for instance, if you draw the
Supreme Court, you’re going to get the number
of pillars right, columns, if there’s an American flag. So there is checking like that. UDI MANBER: Christina. AUDIENCE: So I’m
wondering if you can talk a little bit
about the cartoons that David Remnick did
not let you publish. So you said humor is safe,
but in danger, and getting the right amount of wrong. So I’m wondering if you
could square that with– ROBERT MANKOFF: OK. So here, once again, The
New Yorker, like Google, is a commercial institution that
is supported by advertising. So the only– I forget the
guy’s name, great comedian, Stanhope is his last name– the
only real freedom is stand up in a club where you
can say just what you want, anything you want. I can’t say that
here because why? Because there are real
punishments for it. You have to decide
whether you want those commercial
punishments on you. But here are cartoons that “The
New Yorker” did not publish. One of them, which
is funny, I did this when “The Sopranos”
was a hot show. There’s a Tony
Soprano guy and he’s in a room with his
guys, a closed room, and he’s saying, if
these walls could talk and they knew what was good for
them, they’d shut the fuck up. [LAUGHTER] ROBERT MANKOFF: OK, so
we’re not going to use fuck. Because we don’t do it. Because a child
might see the word fuck, and you know
what would happen then. They would die. We know all the children who
die every day because they see the word fuck,
and that happens. We can’t have on our head. Another cartoon, a
great cartoon, actually, was done by Sam Gross. It was right after the
Danish cartoon controversy, and everybody was
running scared, everybody was scared,
physically scared. I gave talks on stage with
imams and stuff like that, and I was worried,
physically worried because of what I
was going to say. I was worried. So during that period, I
said, I remember what I did. I knew I had to
avoid Muhammad, I couldn’t say anything
about Muhammad because I was afraid they
were going to kill me. Honestly. So I said OK, I’m not
going to say anything, I’m going right to the top. You can make fun of God, you
can’t make fun of Muhammad. But anyway, so after
the controversy Sam Gross, great cartoonist,
submitted this cartoon. It’s in heaven and
there’s a suicide bomber who’s all blown to pieces. Different parts, he’s got
his head, stuff like that. And there’s a, let’s
say, an imam there, and the imam is
saying to him, you’ll get the virgins when
we find your penis. [LAUGHTER] ROBERT MANKOFF: So
we didn’t print that. So there are lots of
cartoons that we don’t print, and that’s partly because
“The New Yorker,” basically, its avenue is benign humor. I think there are definitely
a place for these cartoons. I don’t run “The New Yorker,”
and David Remnick doesn’t even run “The New Yorker,” and part
of what “The New Yorker” is is taste. But in terms of cartoons,
and this question I always get asked is, where do you
draw the line between doing this and offending? And my real answer to that,
my real feeling about that is, offending is an
overrated offense. It’s just, OK,
someone was offended. What happened? Did anybody die? OK, a person was offended. Then what happened
right after that? Did they have lunch? Did they go about the
rest of their day? Could they go to sleep at night? In other words, the
only thing that happened is they were offended. So I don’t think it’s a big
deal, but in a perfect world, we could offend more people. You have a question. AUDIENCE: I was going to
ask in general terms of how much the cartoons have
changed since 1925, but I think I’ll
be more specific. Do you think that one
would have worked in 1925? ROBERT MANKOFF: That’s actually
sort of a more meta– first of all, in 1925 the cartoons in
“The New Yorker” were terrible. They were just like the
cartoons everywhere, they were these long,
drawn out pieces. They didn’t have the
American vernacular, I could give you a whole
history of cartooning going back to Punch in 1843. So they were very tedious. Even the cartoons in “The
New Yorker” in ’28 and ’29, sort of ground-breaking
cartoons, would seem very mild today. In 19– I think it’s ’28–
by Carl Rose, captioned by E.B. White was, it’s
mother at a dinner table, and she’s saying to a
little Lord Fauntelroy son, eat your broccoli. And he says, I say it’s
spinach and to hell with it. That was actually very
sort of a breakthrough, just because of the vulgarity
of the language for that time. Basically, it would be like
he was saying, well, fuck it. But it still has a two
he said, she said thing. The way cartoons actually
changed over time is, by the ’30s and ’40s, I
could show that you basically have this incongruity thing. A strange picture
and the punch line. In the ’40s and ’50s,
most of the humor is within the cartoon frame
when it’s observational humor. In other words, the
person within the cartoon is not making a joke. So you might have a woman
looking at, in 1947, at a watch at a
baseball game saying, annoyed, why didn’t
they tell us there was going to be an extra inning. OK, she’s not making a joke. An Alex Gregory
joke from the ’90s in which one woman
is saying to another, I started my vegetarianism
for health concerns, then moral reasons, and now,
just to annoy people. That’s a modern
cartoon, and the reason is that the people who came of
age, their comedy background is sitcom where people
make their own jokes. So it keeps changing, it
keeps evolving over time. One of the things that I’ve
written about in this book is how a whole new generation
of people are doing it, and they’re bringing
their own take to it. UDI MANBER: So let
me ask you another, to give you an opportunity. So you don’t think the
computers can generate jokes, people have tried
it for a while– ROBERT MANKOFF:
It can generate– UDI MANBER: Good jokes. ROBERT MANKOFF: It can
generate very bad jokes. It’s like a lot of people who
look at a cartoon in “The New Yorker” and it’s a bad cartoon. Maybe it is bad, it’s a lousy
cartoon, terrible cartoon. I’m admitting that,
world’s most horrible, I can’t believe we
published that cartoon. It’s like someone who goes to
go a baseball game at Yankee Stadium and watches the
player strike out and says, I can do that. Let me play. So here you have the very
worst types of jokes. I know, computers can’t do it. UDI MANBER: So let me
give you a premise. The reason I’m saying that,
actually, I found out, there’s a field in
computer science called computational humor. ROBERT MANKOFF: I showed you it. UDI MANBER: Right. And it’s not a very funny name. If they want to succeed, they
have to find a better name. But the point is it reminds
me of chess at the beginning. And somebody who
was very famous, David Levy was his
name, and he bet that no computer will
beat him in 10 years. And he actually got the
10 years exactly right, because the computer beat
him in ten and a half years. But in 10 years did
not, so he won the bet. They did it in 1968. ROBERT MANKOFF: Well, it’s
simply because there are rules and there are algorithms,
you can play chess. UDI MANBER: Let me just. The question is, are you
willing to make this bet? ROBERT MANKOFF: Oh, of course. Well, of course. Even although
these programs that are generating those
bad jokes, they don’t even know they’re bad. Those are the best that
people have to pick out. So there’s no way
for it to evaluate. What I’m trying
to say– Udi and I talked a little about
[INAUDIBLE] is that, it really comes back
to the hard problem. And the problem is consciousness
and feeling itself. Which is the hard problem,
that’s the hard problem. That’s how we react to humor,
that’s how we generate humor. We generate humor
because we have bodies. And so there’s a
small element of it which is this cognitive
development, which is very, very difficult
in and of itself. So all the rules I
gave you, they’re just sort of narrative rules. When I say benign violation,
the right amount of wrong, well, the right amount. Now, there’s no question
that you can pair computers with crowd sourcing
of people to say, is this the right amount of
wrong, and things like that. Just computers in
and of themselves, because all they have
by themselves is syntax, there’s no grounding in reality. Yeah, I don’t– UDI MANBER: Point is,
we can set up a contest and we can make it
real, for example. We can pick up a topic,
and you have half an hour, and you generate five jokes,
and the computer generates five jokes, and you give
it to random people, and see which one
they think is funnier. So the point is if you
put a name to this bet, he got very famous
because he made this bet. You should make your own bet. ROBERT MANKOFF:
I’ll make the bet. The question is, I don’t think
the computer can generate– it can’t simply draw on jokes. I mean, it can’t simply peg
the huge database of jokes. You actually have to
have a program that has no human knowledge, no
pointers to say, this is funny, this is funny, this is funny. I pointed out that they have as
much chance of generating jokes as they do of having sex. OK, can computers have sex? Can they enjoy it? Why would they do it? Why not, why can’t? Because there’s nothing there. There’s no experience. The idea that experience itself,
the actual core of our lives, is somehow irrelevant for one
of the most important things in our lives, which is
humor, to me is– anyway, I’m willing to be proven wrong. And it would be
interesting, too too. But you have to look at those
things that are computational, driving, or algorithmic
things, or certainly chess, and those things that are
actually not computational. I don’t think what I’m
doing now is computational. I don’t think I’m
computing anything. I don’t think I’m drawing from
a vast database of sentences that I’m stringing
together accidentally. I think most of the things
I’m saying right now I have never said
before in this way. And it’s not because there’s
a random generator generating sentences, I’m actually
making sense to you. Although as I
pointed out to Udi, one of the great advantages
of being funny and reasonably intelligent– you see, I’m
reasonably intelligent, but I’m really not that smart. But I’m very funny,
so it seems that I’m more intelligent than I am. And because I’m
pretty intelligent, it also seems that
I’m funnier than I am. That’s a joke. [LAUGHTER] UDI MANBER: OK, next question. AUDIENCE: Sure. So at Google, it’s become pretty
popular to share information and humor using memes. I was wondering if you
enjoy writing memes or if there’s any
memes you’ve written that you think are funny. ROBERT MANKOFF: What
I enjoy, pretty much, is I do stuff on Twitter. I like Twitter for
the one liners, so I think I’ll put up,
maybe today, because I’m 70– as I was saying to
Udi, I like aphorisms– I said about being 70. The good news is could be worse,
the bad news is, it will be. Just think about that. I don’t want to tout that
as the world’s greatest aphorism or anything. It demands a very deep type
of understanding of existence that is very elliptical. So much is left out
that you then put in there to understand, as a
human being, what it means. Basically, you have
to understand, oh, he looks like he’s
in pretty good shape. He’s 70, you say, hey, 70. OK. 20 years. He’s not going to
stay in good shape, the train’s only going
in one direction. It could be worse,
and it will be. So all of that. And the thing is, you
immediately process that. In fact, to the degree,
the people who laughed processed it probably a little
faster than less than a half a second, and the other people
maybe didn’t, or whatever. So the fact is, for it to
work as a joke– there’s some experiments
that show that– it’s got to take about less than
half a second to put together all the things. And you’re putting together
a huge amount in that time. How are you doing it? I don’t think it’s by searching
very quickly, Google-like, parallel processing
through vast things. I think that the fact that we
just exist, feel, have quality, have sensation, that it’s
immediately happening. It’s like if I had a
stick that was 186,000— how fast is the speed of light? UDI MANBER: That’s about right. ROBERT MANKOFF: It’s
186,286 miles long, OK? And you can’t go faster than
the speed of light, right. But my stick is 186,286, about. And you’re on the end of it, uh. That’s it. One faster than light. I pushed the stick, right? I transmitted the information. I know there’s a flaw
in it, but I don’t care. See, one of the great
things about funny is that when you’re
funny, you distract people from the actual logic
of thinking about it. So people say, maybe
he’s right, maybe Google should invent this long stick. Maybe this is a
whole new– let’s give up that whole
self-driving car. The stick, the big stick. That’s going to be the thing. So I think something
like that happens, and of course I don’t know what
happens, and no one– something with that happens in
the mind where somehow the physical essences
and substances are being transmitted almost immediately
for you to do that. The idea that billions
of computations are happening through
huge databases and things just doesn’t make sense to me. I think it’s a
limitation of our minds that the only way we
can think about it is through computation. That’s the only way we can
think it could possibly be done, because
that’s what you do. UDI MANBER: Since you
mentioned your age– ROBERT MANKOFF:
So the bet is on. UDI MANBER: The bet is on. Well, you have to give a name,
and you have to have rules– ROBERT MANKOFF: My
computer overlords. UDI MANBER: And you have
to promise some money to whoever– OK. People have looked
at humor in medicine and showed that, in fact, you
can get better if you laugh and if you involve humor. And yet, when you go
to the doctor’s office, you see news magazines,
which depressing, rather than cartoon books. Why don’t you guys
go and try to put cartoon books all
over hospitals? ROBERT MANKOFF:
Well first of all, let me say about
humor and health. Don’t give up on the chemo. Humor is great as long
as you don’t really have something wrong with you. For all aches and
pains and other things, if you really have something
bad wrong with you. I am not a humor evangelist. I am fascinated with the
cognitive, emotional, the functional sides of humor. But I don’t think humor
is this great health cure. It’s OK, it can act
as a mild analgesic. You do these things, you
stick the person’s hand in cold water, and
then you show them comedy, and that lasts a little
longer than watching something depressing, all of that. I think in terms of
medicine, though, I do think it is emotional
regulation mechanism. So there’s a number of
different ways it can be used. There are different
types of humor. There’s affiliative humor,
which is, basically, the humor that, in difficult
situations, we use to cope. We just kid around
and do things. I try to use that a lot
in my– as you get older, you get things wrong with you. I’ll be at the doctor and
I’ll just say to myself hey, why not be happy for all
the diseases you don’t have. So you make a little
joke for yourself. And then there’s
self-deprecating humor. And then there’s humor
that works in a social way, that just bonds, you
don’t feel the doctor is this impersonal automaton. But the problem is
you, humor is a skill. And if you’re very
unskilled at it, it’s not a good thing to
use, and most doctors are. Why don’t we sign
some books, right? Then these people can
buy the book, oh no, no. OK, go ahead, some
more questions. UDI MANBER: Why don’t
we have a last question. AUDIENCE: There’s some debate
whether we can do a question or not, I guess we can. The captions that readers
submit, many of them are brilliant, but
some of them are not. ROBERT MANKOFF: Is this
the caption contest? AUDIENCE: I was wondering– ROBERT MANKOFF: If it’s a
criticism, I can take it. AUDIENCE: No, what
sometimes I think is that I’d like to see
the original caption. ROBERT MANKOFF: Oh, see
the original caption? Yeah, yeah. I’m going to do that
at a point because I’ve got all the original captions. Almost always, whatever
was in the caption contest as the original caption is
also one of the entries. But one of things to
remember is the cartoonists do 10,15,20 ideas every week. Many of them aren’t very good. The difference between a
professional and an amateur is a professional thinks most
of their own work is crap and an amateur is
really in love with it. So they have a failed idea, and
they just go do another idea. And people are like that. They become very precious
about their ideas. Like I get all the time, I
have an idea for a cartoon, but the person doesn’t
want to tell me it because they’re afraid
I’m going to steal it. So I say, OK, I’m not. Let me finish with
one story about ideas. And I think this is for
Google and for cartoons, and in general. So Conde– Conde’s this big,
fancy, wonderful, interesting, annoying place, like
every other corporation. And there are a lot
of guys in great suits and stuff like this,
this isn’t one of them, this is like a bad Conde suit. So I’m in the elevator with
this guy, and we’re going down, I don’t know the guy. This starts on 20. He said, they stole my idea. They stole my fucking idea. Bitching all the way, I can’t
believe it, they stole it, they stole my fucking idea. Finally get down to the bottom
floor, I don’t know the guy. I tap him on the shoulder
and I said, get another idea. And that’s my suggestion
to all of you. Thank you. UDI MANBER: Good. Thank you very much. ROBERT MANKOFF: Thank you, Udi. [MUSIC PLAYING]

12 thoughts on “Robert Mankoff: “There is no Algorithm for Humor” | Talks at Google

  1. This guy is like the Joker. He laughs and stops it in a matter of a blink! There is no way of knowing when it is a sincere one. Plus, he was the only one laughing after his jokes….. 

  2. when he says to have humor you need a body as well as a mind is he referring to having experience rather than mere knowledge?
     like your knowledge must come from firsthand sources and not secondhand like books and the like.

    if so could it mean bottom-up emergent neural nets would be the key to machine humor?  
    like a robot that learns from trial and error could theoretically log the near misses and corrected mishaps as like anecdotes to its report or something and as it recognizes common elements in each mishap it can then comment on them possibly making observational humor,if i make sense.

    like how in he proverbial dissection of the frog it is better to raise your own personal frog than to pick apart someone else's.
      have the humor grow up organically as you notice the odd happenings in your life rather than figure out what will qualify as "odd"

  3. so what is the message, of couse "computers" can't make jokes because they don't have a body. What if we would have robots walking and living with us…could they do jokes. Or is he saying that its basically impossible for technology (robots, computer, whatever) to make jokes?

  4. Wouldn't expect the people of Google to be some of the most stone-faced, cold people out there, but everyone seemed to be unresponsive to any of his jokes, and looked at him almost as a crazy old man, rather than the brilliant mind he really is.

  5. The New Yorker cartoons still look like they were drawn and written in the 1950s by a group of ageing cartoonists. Not sure why Mankoff thinks he's an authority on humour as the best of his work just breaks a smile (as per audience's response). 800 cartoons over 30 years, sooo that is exactly 1 cartoon every 2 weeks??? Hmmm. As 'the' editor and cartoonist he chose to put in one of his pieces every 2 weeks, I guess one every week would have been too much.

Leave a Reply

Your email address will not be published. Required fields are marked *