"Every game needs a goal. In a Turing machine, goals are imposed from the outside. For true induction, the process itself should create its own goals. And there was a key constraint: Putnam realized that the dynamics he had in mind would only work mathematically if the system had just one goal governing all its behavior.
That’s when it hit him: The goal is to repeat. Repetition isn’t a goal that has to be programmed in from the outside; it’s baked into the very nature of things—to exist from one moment to the next is to repeat your existence. “This goal function,” Putnam wrote, “appears pre-encoded in the nature of being itself.”
So, here’s the game. The system starts out in a random mix of “on” and “off” states. Its goal is to repeat that state—to stay the same. But in each turn, a perturbation from the environment moves through the system, flipping states, and the system has to emit the right sequence of moves (by forming the right self-reinforcing loops) to alter the environment in such a way that it will perturb the system back to its original state."
I'm a big fan of this line of thinking. I've been arguing for years that RL should be based in homeostasis and this seems right along those lines. I wish I could have talked with him!
A captivating read, bringing life to a very interesting character. Thank you for posting this.
I do wonder about Putnam's research though. Has it been looked into by experts in the field more recently? The article doesn't really give an answer to this.
That was a fascinating article. I have no doubt that Putnam's work is far beyond me -- considering that it is beyond leading academics in the field -- but I appreciate the description of a man who tried his best to find the right path for himself, even if it was at odds with what the world expected of him.
If Putnam was really a genius I think he would have been able to communicate precisely some of this insight to mere mortals. It seems that he taught the same course during 10 years and that seems time enough to mature a good introduction to his ideas. Being ineffable don't entitle you to being a genius.
What seems clear is that he was in the right moment and with the right people to produce a great theory of the mind, he was thinking about defining the mind as a creator of heuristic for generating inductive rules, that is a very interesting idea, also that seems related to cellular automata and homeostasis, fixed point theory, error recovery and many other ideas that we, with our ability to look back in time, can see are located in their neighborhood. Recently we are using LLMs to produce heuristic or rules to generate python programs to explore proofs or new models, so the gems is still alive.
The field he was interested in it is full of little diamonds waiting for someone to find them, but perhaps he just put his foot in the door without entering (or creating) the field he was trying to envision. It could be that his paper could reveal some intuition that could guide us in a new search for meaning but from this post there is no any hint about that beyond such a desire.
If your fishing net is constructed just right--so as to pick out the many interesting gems in this article, you will be glad to know Gefter has also published a book that revolves around John Wheeler.
Beautifully written article about an interesting man. Hats off to Amanda (article author) for her determination!
But this story is also a great reminder on the importance of readability. I can't name Putnam a genius if his works were so incomprehensible, that even best of the best had trouble deciphering them. Just like you wouldn't call programmer a genius if he wrote his best work in brainfuck.
I guess he never intended his works to be read. So I won't.
> The neighborhood was quiet. There was a chill in the air. The scent of Spanish moss hung from the cypress trees. Plumes of white smoke rose from the burning cane fields and stretched across the skies of Terrebonne Parish. The man swung a long leg over a bicycle frame and pedaled off down the street.
I stopped halfway through this paragraph and just googled Peter Putnam.
If I want an article to take hours to get to the point I'll go read a recipe blog.
"Every game needs a goal. In a Turing machine, goals are imposed from the outside. For true induction, the process itself should create its own goals. And there was a key constraint: Putnam realized that the dynamics he had in mind would only work mathematically if the system had just one goal governing all its behavior.
That’s when it hit him: The goal is to repeat. Repetition isn’t a goal that has to be programmed in from the outside; it’s baked into the very nature of things—to exist from one moment to the next is to repeat your existence. “This goal function,” Putnam wrote, “appears pre-encoded in the nature of being itself.”
So, here’s the game. The system starts out in a random mix of “on” and “off” states. Its goal is to repeat that state—to stay the same. But in each turn, a perturbation from the environment moves through the system, flipping states, and the system has to emit the right sequence of moves (by forming the right self-reinforcing loops) to alter the environment in such a way that it will perturb the system back to its original state."
I'm a big fan of this line of thinking. I've been arguing for years that RL should be based in homeostasis and this seems right along those lines. I wish I could have talked with him!
A captivating read, bringing life to a very interesting character. Thank you for posting this.
I do wonder about Putnam's research though. Has it been looked into by experts in the field more recently? The article doesn't really give an answer to this.
After this article last week started to read his work (some available here https://www.peterputnam.org/) What a life, what a character.
That was a fascinating article. I have no doubt that Putnam's work is far beyond me -- considering that it is beyond leading academics in the field -- but I appreciate the description of a man who tried his best to find the right path for himself, even if it was at odds with what the world expected of him.
If Putnam was really a genius I think he would have been able to communicate precisely some of this insight to mere mortals. It seems that he taught the same course during 10 years and that seems time enough to mature a good introduction to his ideas. Being ineffable don't entitle you to being a genius.
What seems clear is that he was in the right moment and with the right people to produce a great theory of the mind, he was thinking about defining the mind as a creator of heuristic for generating inductive rules, that is a very interesting idea, also that seems related to cellular automata and homeostasis, fixed point theory, error recovery and many other ideas that we, with our ability to look back in time, can see are located in their neighborhood. Recently we are using LLMs to produce heuristic or rules to generate python programs to explore proofs or new models, so the gems is still alive.
The field he was interested in it is full of little diamonds waiting for someone to find them, but perhaps he just put his foot in the door without entering (or creating) the field he was trying to envision. It could be that his paper could reveal some intuition that could guide us in a new search for meaning but from this post there is no any hint about that beyond such a desire.
If your fishing net is constructed just right--so as to pick out the many interesting gems in this article, you will be glad to know Gefter has also published a book that revolves around John Wheeler.
https://www.peterputnam.org
Wow, I always thought Hilary was the only Putnam to come up with a computational theory of mind.
What a beautiful article about a fascinating man -- I'd be glad to read a full biography.
When first I saw the title I was thinking "how can he be forgotten when I've heard of the Putnam prize", but it's a totally different Putnam.
And according to my friendly neighbourhood LLM, they're not immediately related.
Beautifully written article about an interesting man. Hats off to Amanda (article author) for her determination!
But this story is also a great reminder on the importance of readability. I can't name Putnam a genius if his works were so incomprehensible, that even best of the best had trouble deciphering them. Just like you wouldn't call programmer a genius if he wrote his best work in brainfuck.
I guess he never intended his works to be read. So I won't.
I like how his theory* of the mind basically states that humans intrinsically are OK and everything is fine.
---
* if it's simplification in the article is correct and I understand that simplification correctly
> The neighborhood was quiet. There was a chill in the air. The scent of Spanish moss hung from the cypress trees. Plumes of white smoke rose from the burning cane fields and stretched across the skies of Terrebonne Parish. The man swung a long leg over a bicycle frame and pedaled off down the street.
I stopped halfway through this paragraph and just googled Peter Putnam.
If I want an article to take hours to get to the point I'll go read a recipe blog.
Sorry, this was no Vivian Maier.
Not even close.