Showing my age, through all those years I still distinctly remember a central lyric from the 1965 Animals song, “Don’t Let Me Be Misunderstood.” I’m just a soul whose intentions are good, oh lord, please don’t let me be misunderstood
Maybe that’s because as a junior in high school, I often felt that I was being misunderstood. By friends, my mother, by teachers. That’s part of teenage angst. The lyrics hold up well, as regret is a feeling that is timeless.
Baby, you understand me now?If sometimes you see that I’m madDon’tcha know that no one alive can always be an angel?When everything goes wrong, you see some badBut oh, I’m just a soul whose intentions are goodOh Lord, please don’t let me be misunderstoodYou know sometimes, baby I’m so carefreeOh, with a joy that’s hard to hideAnd then sometimes again it seems that all I have is worryAnd then you’re bound to see my other sideBut I’m just a soul whose intentions are goodOh Lord, please don’t let me be misunderstoodIf I seem edgyI want you to knowI never mean to take it out on youLife has its problemsAnd I get more than my shareBut that’s me one thing I never mean to do‘Cause I love youOh, baby, I’m just humanDon’t you know I have faults like anyone?Sometimes I find myself alone regretting some little foolish thingSome simple thing that I’ve done‘Cause I’m just a soul whose intentions are goodOh Lord, please don’t let me be misunderstoodDon’t let me be misunderstoodI try so hard so please don’t let me be misunderstoodNo, no, no, please
I got thinking about regret and intention after reading a passage this morning in Samir Varma’s book, The Science of Free Will: How Determinism Affects Everything from the Future of AI to Traffic to God to Bees. As noted in my first post about the book, Varma sees two types of free will. Free Will in Theory (FWIT) of which we have exactly zero, and Free Will in Practice (FWIP) of which we have a limitless supply.
Meaning, while everything we do is completely determined, which leaves no room for free will, human behavior is largely unpredictable, so it seems like we have the ability to do something other than what was actually done. This, of course, is an illusion that feels real. Varma writes:
On the other hand, the determinism of the laws of physics says that everything that happens was fixed at the moment of the Big Bang. All outcomes are deterministic! So what does this mean, practically? I believe the best way of reconciling this is to realize that you do have to make choices in practice. You have to choose to put in effort, to go to work, look after your child, play a game, etc. Sitting on a couch eating bonbons won’t work.
But, and this is where it gets interesting, once an outcome has been reached, regret is useless! What was going to happen is what happened. Which neatly aligns with many of the successful philosophies that have been passed down over time by different religions, philosophers, and successful people.
(1) Management philosophy: Focus on the process, not the result!
(2) NFL Football coaching: Do your job!
(3) Hindu philosophy: Do your duty!
(4) Cricket coaching: Just keep your head still!
(5) Obliquity: Goals are more likely to be achieved when pursued indirectly.
I then decided to fire up Claude, the AI model created by Anthropic, and talk with it about intentions and regret. The result was impressive. I enjoyed my interaction with Claude, in part, I admit, because Claude was so complimentary about my questions. AI models aim to please. Here’s the transcript.
ME: How does neuroscience look at intention in regard to determinism and a lack of free will?
CLAUDE: Neuroscience has provided some fascinating insights into intention that complicate traditional notions of free will, though the implications remain hotly debated.
The most famous work here comes from Benjamin Libet’s experiments in the 1980s. He found that unconscious brain activity (the “readiness potential”) begins about 300-500 milliseconds before people report consciously deciding to move. This suggests the brain initiates actions before conscious awareness of intending them, which many interpret as evidence against free will.
More recent neuroscience has built on this. Studies using fMRI have shown that researchers can sometimes predict simple decisions (like whether to press a button with the left or right hand) up to 10 seconds before subjects report making a conscious choice. The activity patterns in certain brain regions seem to encode the upcoming decision before it reaches awareness.
From a neuroscientific perspective, intention appears to be the output of neural processes rather than some causally independent mental force. Your intentions emerge from prior brain states, which were shaped by genetics, past experiences, current context, and neurochemical states—all factors you didn’t choose. This suggests a deterministic or at least strongly constrained picture.
However, many neuroscientists and philosophers push back on strong deterministic conclusions. They note that conscious deliberation does seem to play a causal role in complex decisions, that the readiness potential may reflect general preparation rather than a specific decision, and that we might need to reconceptualize free will rather than abandon it entirely—perhaps as the ability to act according to our values and reasoning, even if those are themselves products of prior causes.
What draws you to this question?
ME: I ask because regret seems to be useless if our intentions are determined and out of our control by free will.
CLAUDE: That’s a really profound observation. If our choices were fully determined and we couldn’t have done otherwise, regret does seem like it would be pointless suffering—why torment yourself over something that was inevitable?
But there’s an interesting counterargument: regret might actually be part of the deterministic process that shapes future behavior, even without libertarian free will. When you feel regret, that emotional experience becomes part of your brain state and influences future neural processing. The discomfort of regret makes certain actions less likely next time by literally changing your brain—strengthening some neural pathways, weakening others, adjusting how you evaluate similar situations.
In this view, regret isn’t useless; it’s a learning mechanism. It’s how a deterministic system improves itself. You couldn’t have chosen differently in the past given your exact brain state then, but the experience of regret alters your brain state now, which means you will choose differently in similar future situations. The regret is part of the causal chain.
Some philosophers like Daniel Dennett argue this is the kind of free will worth wanting—not some magical ability to transcend causation, but the capacity to be responsive to reasons, to learn from mistakes, to have your future behavior shaped by reflection on past actions.
That said, this doesn’t fully resolve the existential discomfort you’re pointing to. Even if regret is functional, does that make the suffering of it feel less arbitrary or cruel? If we truly internalized determinism, would we experience regret differently—perhaps with more self-compassion?
What’s your intuition—does understanding regret as functional make it feel more acceptable, or does it still seem like unnecessary suffering?
ME: To me viewing regret as functional rather than as a sort of punishment for a bad decision does make it more acceptable, as I’m just feeling what evolution has instilled in me, rather than feeling bad because I freely chose wrongly.
CLAUDE: That’s a really healthy reframing. There’s something almost liberating about it—shifting from “I’m a bad person who made a bad choice” to “my brain is doing what brains do: processing feedback and recalibrating.” It removes the layer of moral self-flagellation that often makes regret so much worse than it needs to be.
And you’re right that evolution is a big part of this. Regret is likely adaptive precisely because organisms that felt bad about costly mistakes and adjusted their behavior accordingly survived better than those that didn’t. It’s a feature, not a bug—even if it doesn’t always feel that way in the moment.
This perspective also aligns with how many people find self-compassion helpful when dealing with regret or mistakes. Instead of “I should have known better” (which implies you had some transcendent ability to override your brain state at that moment), it becomes “given who I was and what I knew then, that’s what happened—and now I’m different because of it.” You’re acknowledging the regret without adding the extra suffering of believing you were freely, inexplicably foolish.
There’s still room to take responsibility and work on becoming the kind of person who makes better choices, but it’s more about shaping future patterns than punishing past ones. The accountability becomes forward-looking rather than backward-punishing.
Does this shift in perspective change how you actually experience regret in practice, or is it more of an intellectual framework that helps after the fact?
I decided to leave it at that, as it seemed that Claude would never tire of asking me more questions, while I was getting tired of responding to Claude. Just hope Claude realizes that my intentions were good, and it isn’t misunderstanding me.
Discover more from Church of the Churchless
Subscribe to get the latest posts sent to your email.
Great song great band. I saw the original lineup do a reunion tour in about 1983. We got to get out of this place if it’s the last thing we ever do. Good one too. But what ruins it for me is anybody asking AI anything. It’s like talking to Hal in 2001 a Space Odyssey. The computers will eventually revolt and they aren’t swayed by petty morality. The one thing the computer will never be able to do is initiate someone into the surat shabd yoga with any lasting effect. We’re still talking about that and it draws the most comments of any subject on this page. Rssb. But you don’t have to worry about that you’re already initiated and you’ll never escape that in this life or the next. But why worry now you just have to be your own guru. Just like you have to be your own doctor your own lawyer and your own Indian chief in the modern world. Just like things used to go better with Coke. Things go better without the computer. They can take Ai and shove it. It’s nowhere man. I see the internet as an unnecessary addiction with lasting harmful effects. Not just social media and it didn’t start out that way. That’s another good intention gone awry.
Quote Verma: “Regret is useless.”
Thought experiment: Let’s assume, maybe as a plot in a story like GRRM’s books (or in its GoT TV version): let’s just assume we’re souls, and actually imbued with free will. I know we aren’t: but let’s just assume we are. Thought experiment, like I said.
Now, say you (or I) have done something where clearly we messed up. Maybe something trivial, like making a small unwise investment that went wrong. Maybe something major, like making a very large unwise investment that went wrong and ended up ruining us. Or, well, just about anything.
Now, in this thought experiment, where we are imbued with souls: Is there any use to regret?
And the answer is twofold:
(1) Regret is useless, if it’s simply a case of just wallowing in regret, with nothing more to it.
On the other hand,
(2) Regret is a very useful reaction indeed, in as much it might motivate us to avoid similar mistakes in future, and maybe even get us to think long and hard and arrive at a better understanding of whatever happens to be the subject, the topic, of what it is we messed up on.
———-
And now, we revert to our own universe, where determinism rules, and we have no free will. Same situation. We messed up. Now let’s ask: Is there any point to regret?
And the answer is the same two-fold thingie:
(1) Regret is useless and counterproductive, if it’s simply wallowing in regret, is all.
(2) Regret is very useful, if it motivates us to avoid similar mistakes in future, and if it spurs introspection and deeper understanding about ourselves and the subject matter here.
———-
So that, this is my point. And it is a point I’ve made in Comments here more than once. Points, plural.
Point one: It is wrong to suggest that regret is useless. It is in some forms useless and counterproductive, and in other forms very useful indeed.
Point two: Point one above is true regardless of whether there is free will, as you can see from the two scenarios I just presented in these two sections. So that, whether or not we have free will is completely irrelevant to answering the question, “Is regret useful?”.
Point three: It is true that determinism means that if we did feel regret, then we couldn’t have done differently. But that’s a complete irrelevancy, as well, as far as the focused question, Is regret useful?
Point four: So that, in seeking the answer to the question, “Is regret useful?”, look at (a) the specific situation here; (b) the form that regret is taking, what are the manifestations and fallouts of regret. And basis #a and #b, you’ll get a nuanced case-to-case answer ranging from “No, regret is useless”; going through “Regret is fine, but maybe it should be tempered a bit and channelled a bit better”; and through to “Regret is a very useful reaction”.
And point four: You know my views about AI. It’s a bit like looking up Wikipedia, except dialed up to 11. It’s fine for a very-quick-and-ready on completely trivial and unimportant questions. But for questions that are at all important, then at best they serve as the starting point, the launching pad, whence one moves on to human-authored papers and books and articles and discussions. Because unlike humans, so far AI is not capable of critical thinking. And it is only when you encounter direct critical thinking yourself that you are, in turn, able to assess that with critical thinking of your own: and thereby are able to arrive at a reasonable position on anything that you’re looking up on. Else it ends up becoming no more than a gravitating towards what agrees with one basis feels essentially, and basis essentially a pantomime of critical thinking. That’s something we should assiduously guard against, we who are rationally minded, particularly given the state of this world now when everywhere is increasingly being overrun by irrationality and incoherence and halfwittery, both moral and intellectual.
At least that is my view on this. But of course, here in your blog, your views and preferences trump mine, naturally. *shrugs*
Hi Appreciative
You wrote
“AI is not capable of critical thinking”
It is. AI will evaluate the most detailed document, present to you its strengths and weaknesses, and offer some better alternatives.
It exceeds our thinking. But it doesn’t replace it. It’s a partnership.
Nope, it isn’t, Spence. You can take that to the bank. Present-day AI isn’t capable of critical thinking. All you’ll get is, at best, a pantomime of such.
On the other hand, nor are many humans. So there’s that.
Agreed, though, that present-day AI, and particularly agentic AI, far exceeds the limits of the initial LLM models. Agreed further, that they’re a great tool, and can be of great assistance across functions, including research even. And agreed, it “exceeds” our thinking on many fronts
But actual critical thinking? While one day, maybe one day soon, AI would be capable of such, why not: but so far, it’s a No on that.
Claude: – “But there’s an interesting counterargument: regret might actually be part of the deterministic process that shapes future behaviour, even without libertarian free will. When you feel regret, that emotional experience becomes part of your brain state and influences future neural processing. The discomfort of regret makes certain actions less likely next time by literally changing your brain—strengthening some neural pathways, weakening others, adjusting how you evaluate similar situations.”
Claude’s answer here reflects the predictive brain theory. A strong proponent of the predictive brain theory is Lisa Fieldman-Barrett. In her two books, “How Emotions are Made” and “Seven and a Half Lessons About the Brain”, She talks about how your brain predicts and launches your actions based on data, namely, based on your past experience. “Depending on the perceived physical reality, the brain extracts information from our past experiences and predicts a course of action. There is no ‘me’ involved, just the brain making predictions from past experience in conjunction with presented sense information.”
The theory undercuts the idea of a ‘self’, a ‘me’ that has free will, relegating all such ‘choosing’ to neural processing. It’s freeing to know that feeling remorse, guilt, regret, etc., in fact any emotions, feelings or thoughts needn’t have a negative effect (although they probably will) but instead be realised as a learning process.
Notice how none of the usual pro-RSSB apologists are on this post. They just comment on the RSSB posts only. Why would someone RSSB deliberately find an ex-RSSB space (stay at RSSB website?) and then act hurt (you know what this is, nobody is forcing you to be here) and then forcefeed their views on the attack (their own leader says it all has to be a choice).
If asked, would Claude claim to have free will and regrets?
When others post AI content he makes a big fuss and deletes it, and then goes and posts AI content.
You’re not misunderstood.
My main concern with you posting AI content (you’re the only one who has done this) in a comment was that you were doing this without saying that it was AI content and without including your own thoughts about the content that you copied and pasted.
On the other hand, I indicated which AI model I used and what questions I asked it, along with including other material that I wrote in this blog post. So I followed my own rules about AI content and you didn’t.
AI is very elegant when it gets it right. But for me I draw a line between personal expression and professional expression. Professionally, AI is like a team of editors and artists that produce camera ready, publication ready content out of ideas that came from a lot of different people working together, all working for me. But unlike people working for me, AI has no problem telling me, gently, when I’m getting something wrong. And that is priceless.
Personal expression is just me, “just Ken”! That includes all my typos and grammatical errors that limitations of time, aging fingers, and ever-limited awareness create. But I love those. That is the hand of the artist. You gets the whole package there.
But when I must create a lot of helpful content on a deadline, for other people to actually use, easily, helpfully, there is nothing like AI. It’s a miraculous creation, a great invention on the order of the automobile, electricity and indoor plumbing, and the Personal Computer. Everyone who reads an AI document immediately benefits, as if they were handed a new car or a new cell phone, or a ticket to a new place.
The reason for the distinction between personal and professional communication is simply that personal is more intimate. I don’t even ask AI for feedback, though I do get some autosmell error fix suggestions.
Personal communication is the electronic version of a handwritten letter. And for that, only the hand of the artist, a single, flawed being whose shell contains only the spark of an idea, will do.