Tuesday, May 7, 2024
Obesity is a biochemical problem, not a moral one [Big Food, Bigger Bodies]
Johann Hari, A Year on Ozempic Taught Me We’re Thinking About Obesity All Wrong, NYTimes, May 7, 2024.
Ever since I was a teenager, I have dreamed of shedding a lot of weight. So when I shrank from 203 pounds to 161 in a year, I was baffled by my feelings. I was taking Ozempic, and I was haunted by the sense that I was cheating and doing something immoral.
I’m not the only one. In the United States (where I now split my time), over 70 percent of people are overweight or obese, and according to one poll, 47 percent of respondents said they were willing to pay to take the new weight-loss drugs. It’s not hard to see why. They cause users to lose an average of 10 to 20 percent of their body weight, and clinical trials suggest that the next generation of drugs (probably available soon) leads to a 24 percent loss, on average. Yet as more and more people take drugs like Ozempic, Wegovy and Mounjaro, we get more confused as a culture, bombarding anyone in the public eye who takes them with brutal shaming.
This is happening because we are trapped in a set of old stories about what obesity is and the morally acceptable ways to overcome it.
An experiment:
One scientific experiment — which I have nicknamed Cheesecake Park — seemed to me to crystallize this effect. Paul Kenny, a neuroscientist at Mount Sinai Hospital in New York, grew up in Ireland. After he moved in 2000 to the United States in his 20s, he gained 30 pounds in two years. He began to wonder if the American diet has some kind of strange effect on our brains and our cravings, so he designed an experiment to test it. He and his colleague Paul Johnson raised a group of rats in a cage and gave them an abundant supply of healthy, balanced rat chow made out of the kind of food rats had been eating for a very long time. The rats would eat it when they were hungry, and then they seemed to feel sated and stopped. They did not become fat.
But then Dr. Kenny and his colleague exposed the rats to an American diet: fried bacon, Snickers bars, cheesecake and other treats. They went crazy for it. The rats would hurl themselves into the cheesecake, gorge themselves and emerge with their faces and whiskers totally slicked with it. They quickly lost almost all interest in the healthy food, and the restraint they used to show around healthy food disappeared. Within six weeks, their obesity rates soared.
After this change, Dr. Kenny and his colleague tweaked the experiment again (in a way that seems cruel to me, a former KFC addict). They took all the processed food away and gave the rats their old healthy diet. Dr. Kenny was confident that they would eat more of it, proving that processed food had expanded their appetites. But something stranger happened. It was as though the rats no longer recognized healthy food as food at all, and they barely ate it. Only when they were starving did they reluctantly start to consume it again.
Satiety:
Drugs like Ozempic work precisely by making us feel full. Carel le Roux, a scientist whose research was important to the development of these drugs, says they boost what he and others once called “satiety hormones.”
Once you understand this context, it becomes clear that processed and ultraprocessed food create a raging hole of hunger, and these treatments can repair that hole. Michael Lowe, a professor of psychology at Drexel University who has studied hunger for 40 years, told me the drugs are “an artificial solution to an artificial problem.”
Yet we have reacted to this crisis largely caused by the food industry as if it were caused only by individual moral dereliction. I felt like a failure for being fat and was furious with myself for it. Why do we turn our anger inward and not outward at the main cause of the crisis?
We've got two culture-wide tropes working against us:
When Pope Gregory I laid out the seven deadly sins in the sixth century, one of them was gluttony, usually illustrated with grotesque-seeming images of overweight people. Sin requires punishment before you can get to redemption. [...]
The second idea is that we are all in a competition when it comes to weight. Ours is a society full of people fighting against the forces in our food that are making us fatter. It is often painful to do this: You have to tolerate hunger or engage in extreme forms of exercise. It feels like a contest in which each thin person creates additional pressure on others to do the same. Looked at in this way, people on Ozempic can resemble cyclists like Lance Armstrong who used performance-enhancing drugs. Those who manage their weight without drugs might think, “I worked hard for this, and you get it for as little as a weekly jab?”
The bottom line:
Early indications are that the new anti-obesity drugs are moving people in a similar radically healthier direction, massively reducing the risk of heart attack or stroke. But these drugs may increase the risk for thyroid cancer. I am worried they diminish muscle mass and fear they may supercharge eating disorders. This is a complex picture in which the evidence has to be weighed very carefully.
But we can’t do that if we remain lost in stories inherited from premodern popes or in a senseless competition that leaves us all, in the end, losers. Do we want these weight loss drugs to be another opportunity to tear one another down? Or do we want to realize that the food industry has profoundly altered the appetites of us all — leaving us trapped in the same cage, scrambling to find a way out?
There's more at the link.
Monday, May 6, 2024
ChatGPT outlines a sequel to Unfrosted [Media Notes 119 C]
When I’d posted my remarks on Seinfeld’s Unfrosted, I figured that was pretty much it. But I kept on thinking about it: Seinfeld’s a stone-cold comic genius, so why’s the movie so flat? I figured it had something to do with trying to inflate a brilliant 3- or 4-minute stand-up bit into a one-and-a-half-hour feature motion picture. That’s all well and good, but spell it out: Just what does that actually mean?
So I watched the film a second time and then, so help me, a third time, while taking notes, three pages worth. But somehow I couldn’t cajole myself into actually writing a post. Fortunately the Bill Maher convo came along. So I posted that, transcribing a bit where Jerry talks about Unfrosted.
Then I decided to play around with ChatGPT. I asked it for short bios of a bunch of the characters in the film: Thurl Ravenscroft, Marjorie Post, Jack Lalanne, Harold von Braunhut, Poppy Northcutt, Walter Cronkite, Chester Slink, Roger Sterling, Snap, Tom Carvel, Mike Puntz, and El Sucre. I gave me correct information about the real people, identified Roger Sterling as a fictional character, suggested that Snap might be one of those cereal mascots, and had nothing to say about Chester Slink, Minke Puntz, and El Sucre, though it did note the “sucre” is Spanish for sugar.
Then I decided to play around. A little of this led to a little of that and before I knew it, the Chatster had come up with a treatment for a sequel, which I've appended below. It’s a bit thin on details, as treatments tend to be, but there’s something there. Just what, I don’t know. But something. Who knows, with a team of crack Hollywood script doctors working it over, it might become a brilliant amalgam of Duck Soup, The Founder, and Galaxy Quest. In fact, maybe I’ll suggest just that to the Chatster the next time I sign on.
* * * * *
Title: "Out of This World: A Pop Tart Odyssey"
Genre: Comedy
Director: Jerry Seinfeld
Synopsis:
In the quaint town of Suburbia, USA, chaos reigns supreme when a trio of mischievous cereal mascots, Snap, Crackle, and Pop, stumble upon a scandalous tell-all book written by the charismatic but troubled creative director of Sterling & Co. Advertising Agency, Roger Sterling. As the scandal threatens to tear the town apart, unlikely heroes emerge to navigate the chaos and uncover the true meaning of redemption, friendship, and the pursuit of the impossible dream.
Poppy Northcutt, the brilliant but socially awkward rocket scientist working at Mission Control, sees an opportunity to reclaim control of the narrative and redirect attention towards the town's most ambitious endeavor—the Apollo Space Program. With the help of her eccentric mentor, Harold von Braunhut, and the unwavering support of her loyal friends, Poppy embarks on a quest to inspire a new generation of aspiring astronauts to reach for the stars.
Meanwhile, amidst the chaos of scandal, Tom Carvel, the gruff but lovable owner of the Carvel ice cream parlor, sees an opportunity to capitalize on the town's insatiable appetite for distraction and indulgence. Drawing inspiration from Snap, Crackle, and Pop's mischievous antics, Tom embarks on a quest to create the ultimate comfort food—a revolutionary new snack that combines the convenience of breakfast with the irresistible sweetness of dessert.
As the town's social hierarchy is upended and long-buried secrets come crashing to the surface, unlikely alliances are formed, friendships are tested, and love blossoms in the most unexpected of places. And as the chaos monkeys wreak havoc on the town, they ultimately become the catalyst for a series of events that will forever change the course of history.
In the end, amidst the chaos and confusion, the town of Suburbia, USA, discovers that sometimes the greatest adventures can be found in the most unlikely of places—and that true happiness lies in following your heart, no matter where it may lead.
"Out of This World: A Pop Tart Odyssey" is a heartwarming and hilarious comedy that celebrates the triumph of the human spirit and the power of friendship, love, and the pursuit of the impossible dream. With Jerry Seinfeld at the helm, audiences are in for a wild ride that will leave them laughing, crying, and cheering for more.
Description as Redesign: Carving Nature at the Joints
I'm bumping this to the top because I think it's good to be reminded that it was Plato who coined the trope of carving nature at its joints.
In reply to my open letter Willard McCarty made an observation I want to look at: “So an act of description is creative, or more accurately, an act of redesign under constraints which are very difficult to be exact about but which are hard as rocks.” To which I replied: “’Redesign’ – yes. After all, we ARE trying to figure out the design. We're attempting to reverse engineer it. And it is exacting, but the constraints are not obvious.”
Still, we need to be careful. To be sure, in my experience, description is not easy, not obvious, and, I suppose, creative. I’m not following a set of explicit guidelines; there are things I look for, but it’s not the sort of thing I could reduce to a checklist and then give it to anyone who can read.
But also, the “creativity” of literary criticism has been much extoled in the mainstream literature and I’m a bit wary of that. For that’s what underwrites the multiplicity of often-incompatible interpretations for any given text. That’s fine for interpretation, but not so good for description. I want mutual compatibility of descriptions.
Turkeys, for example, are fairly complex objects and any given turkey will admit of many descriptions, all true and mutually consistent. But a turkey cannot have both two legs and only one, the other having been lost in an accident) – well, yes, turkeys exist in time and a description that’s apt at one moment may contradict one that’s apt at some other moment; that’s no problem.
But then McCarty re-characterizes description as qualification as redesign. And I like that, comparing it to reverse engineering – which I may say more about in another post. And that led me to that old cliché about carving Nature at its joints. We owe that one to Plato’s Phaedrus. Socrates has likened a well-formed speech to an animal with its various appropriately arranged parts and is now examining two different speeches on love (265e-266a):
... we are not to attempt to hack off parts like a clumsy butcher, but to take example from our two recent speeches. The single general form which they postulated was irrationality; next on the analogy of a single natural body with its pairs of like-named members, right arm or leg, as we say, and left, they conceived of madness as a single objective form existing in human beings. Wherefore the first speech divided off a part on the left, and continued to make divisions ...
The vehicle of Plato’s simile, butchering, is worth thinking about. When you carve meat, you want to do so cleanly. You don’t want bones to stick out, which means you’ve got to cut right into the joint, not to either side of it.
The problem, though, is you can’t see the joint when it is covered with muscle and skin. It’s hidden and its location is not apparent on the carcass’s surface. That’s where the comparison gets its power.
And so it is with describing texts. The “joints” are not apparent in the surface. It takes skill to find them.
And what of the computational mindset that I regard as essential to literary criticism (see, for example, Computational Thinking and the Digital Critic: Part 1, Four Good Books)? That’s what tells you where the joints are, or at least gives you clues. As I’ve argued in my long paper on literary morphology, literary form is computational form. Ultimately we want to understand how those computational processes work. To do that we must begin by identifying their traces in the text.
Jerry Seinfeld on Club Random with Bill Maher – Pop Tarts! [Media Notes 119 B]
From the YouTube page:
Bill Maher and Jerry Seinfeld on the gift Bill gave Jerry, Jerry’s attention to detail, the car Jerry picked up Bill in for Comedians in Cars, their respective career maturity, the mistakes Bill made with stand-up crowds, Lorne Michaels’ philosophy on retirement, Bill’s changing desires about performing live, the ruthless nature of Johnny Carson, picking apart compliments, whether there is such a thing as a bad crowd, the genius of Marcus Aurelius, and so much more.
At a point over an hour into the conversation they've been talking about Marcus Aurelius – Seinfeld reads a page or two a day – and they're on the general topic of sophistication (1:06:56):
That's what I think you might find that in in Unfrosted the the Pop Tart movie. It's a silly idea for a movie, but and the jokes are silly, but as we know there are no silly jokes. They're they're either good or they're not. And you'll find there's a level of sophistication in the silliness.
That is my ultimate. When I first saw Monty Python, when I was a kid on PBS, in the early 70s. I. lost. my. mind. The sophisticated silliness, that they were doing, absolutely lit me up like this is everything that I want, everything that I love.
I think "Get Smart" had that, I think Peter Sellers had that. He's acting dumb, but there is such a sophistication to it. It's because as we know as comedians acting dumb is really not, you know Laurel and Hardy are not stupid. [Maher: no no] I wasn't a stoes guy, but Laural and Hardy is elegant and sophisticated.
Seinfeld goes on to stay that he wasn't a Three Stooges guy and that he didn't like Moe. I agree. He loved "Superman" too.
Sunday, May 5, 2024
Cultural evolution: A review of theoretical challenges [& a note on language]
Nichols R, Charbonneau M, Chellappoo A, et al. Cultural evolution: A review of theoretical challenges. Evolutionary Human Sciences. 2024;6:e12. doi:10.1017/ehs.2024.2
Abstract: The rapid growth of cultural evolutionary science, its expansion into numerous fields, its use of diverse methods, and several conceptual problems have outpaced corollary developments in theory and philosophy of science. This has led to concern, exemplified in results from a recent survey conducted with members of the Cultural Evolution Society, that the field lacks ‘knowledge synthesis’, is poorly supported by ‘theory’, has an ambiguous relation to biological evolution and uses key terms (e.g. ‘culture’, ‘social learning’, ‘cumulative culture’) in ways that hamper operationalization in models, experiments and field studies. Although numerous review papers in the field represent and categorize its empirical findings, the field's theoretical challenges receive less critical attention even though challenges of a theoretical or conceptual nature underlie most of the problems identified by Cultural Evolution Society members. Guided by the heterogeneous ‘grand challenges’ emergent in this survey, this paper restates those challenges and adopts an organizational style requisite to discussion of them. The paper's goal is to contribute to increasing conceptual clarity and theoretical discernment around the most pressing challenges facing the field of cultural evolutionary science. It will be of most interest to cultural evolutionary scientists, theoreticians, philosophers of science and interdisciplinary researchers.
From the introduction:
An efficient means for researchers concerned with CET [cultural evolutionary theory] to contribute to improvements in the field is to take account of and respond to the most pressing stated challenges facing the field. As a result, the present critical review of theoretical needs takes its cue from results of a recent survey of the membership of the Cultural Evolution Society about the field's ‘grand challenges’ (Brewer et al., Reference Brewer, Gelfand, Jackson, MacDonald, Peregrine and Richerson2017). Published results of this survey detail two varied sets of challenges. The first set was drawn from semantic analysis of topics directly from survey results. (Subsequent references to ‘survey results’ are to this paper.) Its most pressing challenge, revealed to be about twice as important as the next topic, was with ‘knowledge synthesis’. The authors state that this topic relates ‘primarily to issues of theoretical integration and speaks to the idea that while many behavioural scientists and humanities scholars see culture as a defining feature of humankind, different subfields rarely read each other's work or build interdisciplinary research programmes to explore how human cultures differ from those of other animals’ (Brewer et al., Reference Brewer, Gelfand, Jackson, MacDonald, Peregrine and Richerson2017: 1). Additional topics of appreciable concern in the first set of results were topics labelled ‘culture definition’, ‘theory’, ‘shared language’ across the disciplines, ‘pro-sociality’ and ‘cultural transmission’.
Language as an evolutionary phenomenon
Until not long ago, discussion of language ‘evolution’ was somewhat taboo. As recent as 1994, textbooks observed that ‘Evolution … has become a “dirty word” in modern linguistic theory’ (McMahon, Reference McMahon1994: 314). This is because through the twentieth century the nomenclature of evolution risked some allusion to outdated, often racist, views about how some languages – and by extension some social groups – might be ‘more evolved’ than others. It is now uncontroversial within linguistics that no language is inherently ‘better’ or ‘more useful’ than others. So, while vigilance must be maintained, the socio-political risks associated with evolutionary talk have begun to fade, and in the past 30 or so years evolutionary approaches to languages have grown dramatically in prominence (Tallerman & Gibson, Reference Tallerman and Gibson2011; Dediu & de Boer, Reference Dediu and de Boer2016).
Today, the expression ‘language evolution’ is widely used to describe at least two distinct phenom- ena and one methodological advance. The two phenomena are: the biological process by which humans, and apparently only humans, became a ’language-ready’ species; and the set of cultural – or ‘cultural evolutionary’ – processes by which relatively simple and unstructured systems become highly structured, and hence acquire some of the common, characteristic properties of languages, such as symbolism, compositionality and duality of patterning. The methodological advance is the application of phylogenetic tools, derived from population genetics, to study language history and lan- guage change. All three of these literatures are now very large, comprising thousands of papers each. Multiple past papers review, summarize and synthesize them (e.g. Mace & Holden, 2005; Tomasello & Call, 2019; Greenhill et al., 2020; Haspelmath, 2020; Roberts & Sneller, 2020; Scott-Phillips & Heintz, 2023; Scott-Phillips & Kirby, 2010). Here we focus on a philosophical issue arising from this growing influence of evolutionary thinking.
The challenge, put simply, is: what exactly does the evolutionary perspective bring that other approaches do not? The biology of language, language emergence and language change have all been important topics for language science for a long time, and a great deal of what has been uncovered by evolutionary approaches has previously been investigated and described in other terms. Research adopting an evolutionary perspective has enriched our understanding, provided new methods and added many new findings, but does it fundamentally alter our understanding of what the empirical phenomenon is and how it works?
One answer might be that a relatively faithful transposition of the Darwinian model – where vari- ation, selection and inheritance combine to generate natural selection and hence the appearance of design in nature – is possible, and brings with it new insights, explanations and tools. With languages, there is something like variation (linguistic items vary enormously), there is something like selection (some items become more common than others) and there is something like inheritance (we learn from the previous generation). Some linguists indeed propose that linguistic items, of some sort or another, could be identified as units of selection closely analogous to genes (e.g. Croft, 2008; Ritt, 2004; Tamariz, 2019). However, deeper analysis raises difficult questions, and the issue is contentious: a decade ago, a group paper on language as a culturally evolving system observed that, ‘Various scho- lars have proposed that concepts, cultural behaviors, or artifacts may function as replicators. It remains to be seen whether any, all, or some combination of these entities are reasonable candidates for cultural replicators’ (Dediu et al., 2013: 314). The situation has not changed substantially since.
Another response to this question would be to adopt only specific parts of the Darwinian toolkit. That is, one could reduce the role of an analogy between variation in language and variation in genes while retaining other parts of Darwinism (Claidière et al., 2014). For instance, there might be scope for perspectives that adopt population thinking and some commitment to selectionism, but without the stronger commitment to replicators.
I have some remarks on the cultural evolution of language in:
Culture as an Evolutionary Arena, Journal of Social and Evolutionary Systems 19(4): 321-362, 1996, https://www.academia.edu/235113/Culture_as_an_Evolutionary_Arena
The Evolution of Human Culture: Some Notes Prepared for the National Humanities Center (Version 2), Working Paper, 2014, 65 pp.
I have proposed the idea of a coordinator as the cultural analog to the biological gene:
“Rhythm Changes” Notes on Some Genetic Elements in Musical Culture, Signata 6, Annales des Sémiotiques /Annals of Semiotics: Sémiotique de la musique / Music and Meaning. Per Aage Brandt and José Roberto do Carmo Jr., eds. Presses Universitaires Liège, 2015, pp. 271-285, https://www.academia.edu/23287434/_Rhythm_Changes_Notes_on_Some_Genetic_Elements_in_Musical_Culture
Intelligence, A.I. and analogy: Jaws & Girard, kumquats & MiGs, double-entry bookkeeping & supply and demand
Think of this post as an adjunct to my series on A.I., chess, and language, which is about the structure of computation in relation to difficult problems.
NOTE: It runs long, so sit back, relax, pour a Diet Coke, some San Pellegrino, a scotch, light up a spliff (assuming it’s legal where you live) — whatever you do to make online reading tolerable — and settle in for the duration. Or you could just print it out.
I’m interested in the general question of what it would mean to say that an A.I. is more intelligent than the most intelligent human, something like that. That’s an issue that’s being debated extensively these days. For the most part I don’t think the issue is very well formulated.
To be honest, I don’t find it to be a very compelling issue. It doesn’t nag at me. If others weren’t discussing it, I wouldn’t bother.
The notion of intelligence itself remains vague despite all the discussion that it has occasioned. I rather expect that as A.I. becomes more developed, we’ll develop a more sophisticated understanding the issue. The general notion seems like it can be captured in a simple analogy:
Intelligence is to a mind’s capacity for dealing with cognitive tasks, such as finding a cure for cancer
AS
Horsepower is to an engine’s ability deal with mechanical tasks, such as the acceleration of an automobile.
But I don’t want to take up the general issue in this post. Rather, I want to look at analogical reasoning. I start with 1) a specific kind of analogical reasoning, interpreting narratives, 2) use some remarks Geoffrey Hinton made about analogy to introduce some more general remarks, 3) move on to another specific example, an analogy between double-entry bookkeeping and supply and demand, and then 4) conclude by wrapping things up with a quick comparison to computing chess, which can be implemented in a well-defined search space, while searching for and explicating analogies cannot.
Interpreting Texts: Girard and Jaws
A lot of literary interpretation proceeds by constructing analogies between events that happen in texts and patterns of behavior specified in some theory about human behavior. Psychoanalysis has theories about how individuals behave in their personal lives; Marxism has theories about how social class and roles in the economy affect individual action. René Girard has a theory about imitation and sacrifice. That’s what I want to look at.
Early in 2022 I did a Girardian interpretation of Spielberg’s Jaws. At the end of that year I had ChatGPT do the same. Of course ChatGPT can’t watch movies, but Jaws is well-known there’s lots of stuff about it on the web, including scripts, though I don’t know whether or not any of the available scripts give the dialog word-for-word as it is in the film (the scripts I downloaded, for free, did not). Now, I didn’t just ask ChatGPT to use Girard’s ideas to interpret the film. First I prompted it about the film itself, and specifically about Quint. Once I’d established that it had a reasonable grasp of the film I then brought up Girard, generally at first, then specifically about mimetic desire. Then I asked it: “How does mimetic desire work in Jaws?” It gave a reasonable answer:
Mimetic desire is a concept developed by the French philosopher and literary theorist Rene Girard. It refers to the idea that human desire is often based on the desire of others, and it can be seen at work in the film Jaws in a number of ways.
One example of mimetic desire in the film is the way that the townspeople of Amity Island desire the safety and security that the town’s police chief, Martin Brody, is able to provide. When the town is terrorized by a great white shark, the townspeople look to Brody to protect them and to find a way to stop the shark. This desire for protection and security is mimetic, as the townspeople are basing their desire on Brody’s ability to provide it.
Another example of mimetic desire in the film is the way that the shark itself is driven by its own insatiable hunger and desire. The shark is driven to attack and kill the townspeople because it desires their flesh and blood, and it is unable to control its own desire. This mimetic desire ultimately leads to the shark’s own destruction, as it is killed by the town’s police chief in a final confrontation.
Overall, Girard’s concept of mimetic desire can be seen as a useful way of understanding the dynamics at play in the film Jaws, and the ways in which desire can be both destructive and transformative.
The second paragraph is reasonable, but the third is problematic. Hunger is not mimetic on the face of it. But I was happy that it ChatGPT seemed to at least have some idea of that nature of the reasoning required, where to look and what to look for in the film. I decided not to pursue that and turned my attention to sacrifice, first Girard, then the application to Jaws.
The form of the required reasoning is analogical. In effect, is there an analogy between Girard’s general statements about mimetic desire and sacrifice and the specific events in Jaws? I would argue that setting up the question is more difficult than working out an answer to it. I note that it’s not the kind of question that has only one answer; thus the argument I actually made in my paper is more sophisticated than the one ChatGPT came up with through my prompting. What is it that brought me to pose the question in the first place?
I watch a fair amount of streaming video, but I don’t write about most of the titles nor do routinely a watch a particular title with the intention of writing about it. That decision is made later. I had no intention of writing about Jaws when I decided to watch it. I was simply filling a hole in my knowledge of movies – I’d never seen the film, which I knew to be an important one. Once I’d watched the film, I read the Wikipedia article about it, something I routinely do, mostly to ‘calibrate’ my viewing experience. The article noted that the sequels were not as good as the original. I decided to see for myself. I was unable to finish watching that last two sequels (of four), but I watched Jaws 2 at least twice, and the original three or more times. It was obvious that the original was better than the others. I did the multiple viewings in part to figure why the original was better. I was on the prowl, though I hadn’t yet decided to write anything.
I decided there were two reasons the original was best: 1) it was well-organized and tight while the sequel sprawled, and 2) Quint, there was no character in the sequel comparable to Quint. I have no all but decided that I would write about Jaws.
I posed a specific question: Why did Quint die? Oh, I know what happened in the film; that’s not what I was asking. The question was an aesthetic one. As long as the shark was killed the town would be saved. That necessity did not entail the Quint’s death, nor anyone else’s. If Quint hadn’t died, how would the ending have felt? What if it had been Brody or Hooper?
It was while thinking about such questions that it hit me: sacrifice! Girard! How is it that Girard’s ideas came to me? I wasn’t looking for them, not in any direct sense. I was just asking counter-factual questions about the film.
With Girard on my mind I smelled blood. I had a focal point for an article. I started reading articles from various sources, making notes, and corresponding with my friend, David Porush, who knows Girard’s thinking much better than I do. Can I make a nice tight article? That’s what I was trying to figure out. It was only after I’d made some preliminary posts, drafted some text, and run it by David, that I decided to write an article. It turned out well enough that I decided to publish it.
Now, when we’re thinking about whether or not A.I.s will come to exceed our intelligence, are we imagining them going through such a process? For this kind of search and exploration is central to human thinking. I certainly do this sort of exploration when thinking about other things, such as the structure of human cognition, the nature of cultural evolution, the functioning of the nervous system, and so forth. This blog is a 14-year record of my explorations, during which I’ll gather some of them together in a more formal way and write a working paper which I’ll then post at Academia.edu, SSRN and ResearchGate. Every once in a while I’ll write an article which I’ll submit for publication in the formal academic literature – a few of those have gotten published. And then there are the monthly pieces I publish in 3 Quarks Daily, which is quite different from the formal academic literature. And of course I’ve got pages and pages of unpublished notes that support all this activity.
Is this kind of exploratory work part of the routine of the superintelligent A.I., or does it go straight for the good stuff, cranking out fully-realized work without need of exploratory effort? If so, how does it know where to dig for the good stuff? Is that what superintelligence is, knowing where the good stuff is without having to nose around? No one says anything about this. Perhaps they’re thinking about the Star Trek computer. But it knows where to look because Spock points it in the right direction.
This brings us back to Jaws. There is a world of difference between what I did in writing about Jaws and what ChatGPT did. I did the hard part, figuring out that there was a specific intellectual objective there, Jaws and Girard. Once I’d done that there was still work to do, quite a bit of work, but it was of a different kind. I was no longer prospecting for intellectual gold. I was now constructing a system for mining the ore and then refining it into gold. ChatGPT only had to do the last part, dumping the ore into the hopper and cranking out the refined metal. I told it where to look, Jaws, what to look for, Girard’s ideas, and gave it some help turning the crank.
A year later, in January of 2023, I decided to see how ChatGPT would do without all of my prompting. I gave it this prompt:
Stephen Spielberg is an important film-maker. Jaws is one of his most important films because it is generally considered to be the first blockbuster. Rene Girard remains an important thinker. Can you use Girard’s ideas mimetic desire and sacrifice to analyze Jaws?
It didn’t do so well. It needed my prompting to get it through the exercise.
Now, no one is claiming that ChatGPT is superhuman in any respect but its ability to discourse on anything. But GPT-5, who knows, maybe it’ll be superhuman in some interesting way. If not GPT-6, or GPT-7, or maybe we’ll need a more sophisticated architecture, but surely at some point an A.I. will surpass us in the way that we surpass mice. Perhaps so.
But I have no sense that these breezy predictions are supported by thinking about how human intelligence actually goes about solving problems. I does no good to say, but it’s an A.I.; it works differently. Well, maybe yes, maybe no, but there has to be some kind of process. At the moment the human process is the only example we have. Perhaps we should think about it.
Just how is it that Girard popped into my mind in the first place? How do we teach a computer to look around for nothing in particular and come up with something interesting?
Analogy: Kumquats and MiGs
As I remarked above, the process of interpreting Jaws is an analogical one. So let’s think about analogy more generally. I’m thinking in particular of some remarks Geoffrey Hinton made at a panel discussion in October of 2023. You can find the video here. I’ve transcribed some remarks:
1:18:28 – GEOFFREY HINTON: We know that being able to see analogies, especially remote analogies, is a very important aspect of intelligence. So I asked GPT-4, what has a compost heap got in common with an atom bomb? And GPT-4 nailed it, most people just say nothing.
DEMIS HASSABIS: What did it say ...
GEOFFREY HINTON: It started off by saying they're very different energy scales, so on the face of it, they look to be very different. But then it got into chain reactions and how the rate at which they're generating energy increases– their energy increases the rate at which they generate energy. So it got the idea of a chain reaction. And the thing is, it knows about 10,000 times as much as a person, so it's going to be able to see all sorts of analogies that we can't see.
DEMIS HASSABIS: Yeah. So my feeling is on this, and starting with things like AlphaGo and obviously today's systems like Bard and GPT, they're clearly creative in ... New pieces of music, new pieces of poetry, and spotting analogies between things you couldn't spot as a human. And I think these systems can definitely do that. But then there's the third level which I call like invention or out-of-the-box thinking, and that would be the equivalent of AlphaGo inventing Go.
OK. Let’s start from there. Given that GPT-4 “knows about 10,000 times as much as a person,” what procedure will it use “to see all sorts of analogies that we can't see”? I’m thinking of that procedure as roughly analogous to the exploratory process I undertake whenever I decided to watch some video. Every once in a while I decide to write about one of the titles. Most of the time, time, though, what I write isn’t as elaborate as my article about Jaws and Girard – I’ve collected many of those pieces under the rubric of Media Notes, though most of those pieces do not focus on analogical reasoning.
What’s the procedure by which an GPT-4 would search through all those things it knows and come up with the interesting analogies? There isn’t one and I suppose it’s a bit churlish of me to suggest that Hinton should specify one. But really, if there he has no procedure to suggest, then what’s he talking about? We know how chess programs search the chess tree. How do we search through concept space for analogies? Alas, while the chess tree is a well-defined formal object, the same cannot be said of concept space, which is little more than a phrase in search of and explication. And how do we evaluate possible analogy-pairs?
Perhaps the simplest procedure is simply to ask. That’s something I recently tried. Here’s the prompt I gave to ChatGPT:
Saturday, May 4, 2024
Unfrosted: The Pop-Tarts Story [Media Notes 119 A]
Was it funny? Yes. Worth watching? I suppose. But it wasn’t the laugh-out-loud hilarity fest I was hoping for. It wasn’t Duck Soup for the 21st century.
I like Seinfeld, a lot. I’ve written a bunch of posts about his stuff, mostly Comedians in Cars and assorted stand-up bits, and gathered most of those into two working papers, Seinfeld's Comedy, Jokes are Intricately Crafted Machines (2023), and Jerry Seinfeld & the Craft of Comedy (2016). That Seinfeld is a miniaturist. Unfrosted: The Pop-Tarts Story started life as a stand-up bit. In this clip Seinfeld talks about how he created that bit (with shots of his hand-written notes on a yellow legal pad):
I wonder about that line he mentions (02:36), “chimps in the dirt playing with sticks.” He explains why he likes it, four of the seven words are funny (underlined). It makes me think of the Kubrick’s 2001, which picks up on the space theme Seinfeld had introduced seconds before (02:30 “it was like an alien spacecraft”). Was that connection rattling around in Seinfeld’s mind as well? Who knows? Does it matter? Maybe yes, maybe no. And he’s only halfway through his explanation.
Back to the movie, Unfrosted. It’s bright and cheery, something Seinfeld was aiming for. In one or three of the dozen interview clips I watched over the past week he says that, just as you are greeted with a shelf of brightly colored cereal boxes when you go to fix breakfast in the morning, so this movie about a breakfast pastry needs to be bright and cheery. Bright and cheery? I’ll give it a smile and two chuckles.
Seinfeld also goes on and on about getting to work with Hugh Grant, a hero of his. Hugh Grant is cast as Thurl Ravenscroft, a Shakespearean actor reduced to (the indignity of) playing Tony the Tiger – remember Alan Rickman in Galaxy Quest, “by Grabthar’s Hammer”? In that role he comes up with that famous tag-line, “They’re gr-r-eat!” You know what? Not so great. Add two smiles and a chuckle to the score. And then in the climax, which is a mascot rebellion filmed as a parody of MAGAs storming the Capitol Building on January 6, Hugh “Tony the Tiger” Grant is wearing a horned fur helmet like Jacob Chansley, the QAnon guy. Why?
The movie’s set in the 1960s, the clothes, the cars, the music – Chubby Checker doing “The Twist” fergodsake! – Khrushchev, JFK, the missile crisis, Walter Cronkhite, NASA & Tang, it’s all there. What’s the MAGA rampage doing in there? It makes no sense. A mascot rebellion? Fine. But all those shots modeled on video footage of the MAGA insurrection? That reference is just a distraction that adds nothing to the story.
The idea seems to be that you take the Pop-Tart comedy bit, turn it into a competition between Kellogg’s and Post, and then frame that competition as a parody of the 1960s space race – I must have heard that line in a half-dozen of those interviews. It sounded promising each time I heard Seinfeld say it. I was intrigued. But on the big screen? Whats the score now, three smiles and three chuckles? And no belly laughs. That seems about right. You can’t take a Godzilla toy, hook it up to an air-pump, and expect to inflate it into a world-destroying comedic monster. That’s not how these things work.
But that seems to be what Seinfeld has done. Here’s what the good folks at Rotten Tomatoes had to say: “Much like a preservative-packed toaster pastry, Unfrosted is sweet and colorful, yet it's ultimately an empty experience that may leave the consumer feeling pangs of regret.” That’s a bit harsh. No pangs of regret, no ultimate anything, not empty. But not particularly filling.
* * * * *
Bonus: I decided to see what kind of scenario ChatGPT could come up with. Here’s a record of a session I had with it. While it’s not gr-r-eat!! it did get a couple of chuckles from me. As always, my prompt is in boldface, the Chatster’s response is plain-face.
Evidence of predictive coding hierarchy in the human brain listening to speech
Charlotte Caucheteux, Alexandre Gramfort, & Jean-Rémi King, Evidence of a predictive coding hierarchy in the human brain listening to speech, Nature Human Behaviour, Vol. 7, 430-441, March 2, 2023, https://doi.org/10.1038/s41562-022-01516-2
Abstract: Considerable progress has recently been made in natural language processing: deep learning algorithms are increasingly able to generate, summarize, translate and classify texts. Yet, these language models still fail to match the language abilities of humans. Predictive coding theory offers a tentative explanation to this discrepancy: while language models are optimized to predict nearby words, the human brain would continuously predict a hierarchy of representations that spans multiple timescales. To test this hypothesis, we analysed the functional magnetic resonance imaging brain signals of 304 participants listening to short stories. First, we confirmed that the activations of modern language models linearly map onto the brain responses to speech. Second, we showed that enhancing these algorithms with predictions that span multiple timescales improves this brain mapping. Finally, we showed that these predictions are organized hierarchically: frontoparietal cortices predict higher-level, longer-range and more contextual representations than temporal cortices. Overall, these results strengthen the role of hierarchical predictive coding in language processing and illustrate how the synergy between neuroscience and artificial intelligence can unravel the computational bases of human cognition.
Subscribe to:
Posts (Atom)