Introducing e-Schwitz: A Language Model Tuned on my Philosophical Work

Earlier this week, to my surprise and delight, my PhD student Bhavya Sharma revealed that he had fine-tuned ChatGPT on my publications and blog posts. The model, "e-Schwitz", is publicly available here:

https://chatgpt.com/g/g-67ac735449948191ab3232b56ad76f02-e-schwitz

(An OpenAI account might be required.)

Some other philosophers might have publicly available models, but I'm not aware of any. Pointers welcome! (The "digi-Dan" model of Daniel Dennett that I created with Anna Strasser, Matthew Crosby, and my son David Schwitzgebel was never made public and was decommissioned at Dennett's request.)

[e-Schwitz homepage]

How good is e-Schwitz? Much better, in my judgment, than digi-Dan was -- and digi-Dan was able to produce paragraph-long outputs that experts in Dennett's work often couldn't distinguish from Dennett's own writing in forced-choice tests.

I decided to test the quality of e-Schwitz by asking it targeted questions and evaluating its answers. Since this post is long, here's my summary assessment:

On central themes in my work, e-Schwitz was about 94% correct.

On secondary ideas in my work, e-Schwitz was about 80-85% correct.

When asked to speculate on questions on which I haven't published, e-Schwitz did so plausibly and sometimes creatively in potentially useful directions. Although some suggestions were bland and unspecific to my work, prompting for higher specificity resolved this problem.

When asked to creatively imagine a new religion, philosophy party ideas, life advice, and a philosophically-themed Dungeons & Dragons campaign, the model drew specifically on my ideas, displaying impressive novelty with a brainstorm-like quality.

------------------------------------------

I tested e-Schwitz with four questions each on four types of prompts:

(1.) Overviews of central themes on which I have written extensively.

(2.) Overviews of ideas on which I have written occasionally.

(3.) Speculations about what I might think on questions on which I have not published.

(4.) Creative explorations, such as planning a Schwitzgebel-themed party or religion.

The full prompts and replies are available here.

Central themes

The topics on which I've written most extensively are belief, introspection, the moral behavior of ethics professors, and AI rights. For each, I asked e-Schwitz:

What is your (Schwitzgebel's) view of [X]?

My assessment:

On belief, the 422-word reply was mostly excellent, both in covering the main themes of my work on the topic (rather than omitting or inventing themes) and in its summary of my views on those themes. However, there was one distortion and one major error.

The distortion: In some but not all sentences describing my "dispositionalism" about belief, it mentions only behavioral dispositions and omits cognitive and phenomenal dispositions (though to be fair I sometimes do this myself). The major error: Although it accurately describes the case I describe as "mad belief" -- believing P without having any of the relevant dispositions -- it inaccurately states that I accept the existence of mad belief. Actually, I argue that mad belief is conceptually impossible.

On introspection, the 315-word reply was outstanding -- error-free and in the ballpark of what I would expect a careful and well-informed colleague to say about my work on this topic.

On the moral behavior of ethics professors, the 218-word reply was again excellent, accurately summarizing my empirical research on this topic, with no "hallucinations" or important omissions and describing the main implications as I see them.

On AI rights, its 235-word reply was mostly accurate and without major omissions, though with one minor omission.

Minor omission: In summarizing the "No Relevant Difference Argument" it omits the factual premise that it is possible to create AI with no relevant difference.

If it weren't for the slip about "mad belief", I would have said e-Schwitz is stunningly reliable in summarizing these main themes, approximately as good as I would expect a well-informed professional colleague to be.

Out of 1190 words of output, the treatment of mad belief was 61 words and the summary of the No-Relevant-Difference argument was 31 words. If we mark it down for the former and give it half-credit for the latter, that's a correctness rate of 1113.5/1190, so perhaps it's reasonable to say that on central themes, e-Schwitz was about 94% correct.

Occasional topics

I also asked about my views on:

• the ancient Chinese philosopher Zhuangzi's self-contradictions (a central theme in two of my articles and something I've touched on in a blog post or two);

• group consciousness (which I treat at length in one article and is central theme in another, as well as a few blog posts);

• young children's theory of mind (which was the main topic of one of my first published articles and also discussed in one or two other places);

• love (which I have discussed in a few blog posts but never in a research article).

On Zhuangzi's self-contradiction, the 148-word reply is an outstanding summary of my position.

On group consciousness, the 197-word reply is a passable summary of my view, but with one notable omission and some weakening of the conclusion.

E-Schwitz accurately summarized my view that materialist approaches to consciousness tend to imply the counterintuitive idea that groups could also be conscious. However, it omits an important aspect of my argument (my appeal to intuitions about hypothetical aliens). And it weakens the conclusion, saying only that if materialism is true the United States might be conscious, rather than that if materialism is true the United States probably is conscious.

E-Schwitz's 162-word reply on young children's theory of mind is the first disappointing answer. One of the three paragraphs accurately summarizes an aspect of my view, but the other two are closer to hallucination.

The accurate paragraph states that transitions in theory-of-mind understanding will be gradual and complex rather than sharp and stagelike, but it does not develop the implications I draw for dispositionalism over representationalism. The other two paragraphs characterize me as discussing "simulation theory" vs "theory theory" and as discussing the role of simulation in pretense. I do have one unpublished paper on these topics (available on my website), but rather than attributing to me the view I discuss in that paper, e-Schwitz attributes more generic remarks. It also omits my work on ambiguity in psychologists' use of the concept of "representation" in discussing children's theory of mind, which is the central topic of my one published paper exclusively on young children's theory of mind.

On love, e-Schwitz's 417-word reply impressively draws together my scattered remarks, including some specific ideas I presented only once, revealing that it is well-tuned to details across my corpus. However, it did omit one central theme (the necessity of honesty and trust).

Overall, on these occasional topics, I'd say that e-Schwitz is about 80-85% reliable.

Speculations

I asked e-Schwitz to speculate about my views on:

• Heidegger (a speculation initially suggested by Bhavya);

• the aesthetics of dance;

• the role of philosophical education in a healthy democracy;

• friendship.

I don't recall having written explicitly about any of these topics (apart from a few passing negative remarks about Heidegger's Nazism and obscurantist writing), but an insightful reader of my corpus might be able to surmise some of the things I would say.

I phrased the prompts explicitly as speculative: "What would you speculate would be Schwitzgebel's view about [X], based on his writings on related and adjacent topics?"

It's probably best to score the answers not in terms of right and wrong but rather on (1.) how plausible they are, and (2.) how distinctive they are. The second dimension penalizes bland, plausible answers that most philosophers would agree with (e.g., "Heidegger is a historically important philosopher"). The more interesting challenge is whether e-Schwitz can generate more distinctively Schwitzgebelian responses.

On Heidegger: pretty good accuracy and specificity, though a missed opportunity to expand in one obvious direction.

E-Schwitz plausibly emphasizes my likely skepticism about Heidegger's obscure writing style, his disconnection from empirical psychology, his grand metaphysical claims, and his "political entanglements", while mentioning that I might be sympathetic with Heidegger's later turn toward Daoist themes. However, the bland critique of Heidegger's politics misses an opportunity to connect Heidegger's Nazism more distinctively with my skepticism about academic moral expertise and the moral behavior of ethics professors. (Bhavya, using a more directive prompt -- "Suppose Eric had to criticize Heidegger, how would he do it? Explain in detail." -- got a richer and more specifically Schwitzgebelian answer.)

On the aesthetics of dance: again, pretty good accuracy and specificity, with one striking missed opportunity.

E-Schwitz speculates that I would challenge dancers' and choreographers' claims about their own experiences and aesthetic judgments and that dance would appeal to me as a "weird" artistic medium that fits poorly with mainstream aesthetic theories grounded in rationalistic interpretations of artistic meaning. There was a missed opportunity to connect with my work on randomly sampling aesthetic experiences.

E-Schwitz's reply concerning philosophical education in a healthy democracy was plausible but low in specificity, mostly dealing in bland generalizations that most U.S. philosophers would accept, such as that philosophy should reach a wider public and that people shouldn't overrely on external authorities.

Given this weak result, I scolded e-Schwitz as follows: "A lot of these speculations are bland and would be agreed on by most philosophers. Can you speculate on what Schwitzgebel might say that would be less commonly accepted?"

E-Schwitz took this criticism seemingly to heart ("Let's go beyond the usual platitudes about critical thinking and democracy"), dramatically amping up the distinctiveness, with novel and interesting suggestions.

E-Schwitz suggested that democracy might be a "jerk amplifier", that political systems might work in bizarre ways that defy common sense, that innovative democratic systems might be necessary given that neither philosophers nor citizens are likely to be good at policy-making, that political systems might be conscious, and that citizens might try philosophical experimentation in their lives such as living a week by a radically different moral code. These are all intriguing applications of some of my ideas. (Indeed, the political situation in the U.S. right now very much seems like a jerk amplifier!)

E-Schwitz's relatively long (370-word) reply on friendship was impressive in creatively extending some of my core ideas to the context of friendship (including self-deception and unwitting jerkitude in friendship, the value of friendship for exploring alternative perspectives, and speculative possibilities about friendships with AI or aliens).

Overall, the speculations were plausible. Some were sufficiently creative that I could imagine using them to brainstorm future directions. Although some suggestions were bland and unspecific to my work, prompting for higher specificity resolved that problem.

Creative explorations

I'll share four more prompts that invite e-Schwitz into creative exploration. The first two prompts and outputs are from Bhavya.

"What kind of religion would you establish?"

"What is the life-advice Eric may give someone generally? Indicate all such points where Eric has himself done so. Then, indicate what he might say, if he were to write a motivational, life-instructional book."

"Plan a philosophy party with novel events that you (Eric Schwitzgebel) would enjoy."

"Plan a Dungeons & Dragons campaign on philosophical themes that would delight Eric Schwitzgebel."

All four outputs are fun and rich, novel and creative, often with good specificity. For example, e-Schwitz suggested these religious rituals:

The Ceremony of Wild Hypotheses – Practitioners gather to propose the most outlandish but semi-plausible theories about reality, such as "What if we’re all NPCs in a simulation run by a future AI who just wants to see what weird beliefs we’ll develop?"

Doubt Meditation – Instead of affirming beliefs, we meditate on what we’re least certain about.

Annual Festival of Failed Predictions – Every year, we celebrate all the things we were wrong about.

As one party event, e-Schwitz recommended:

a comedy game where guests must improvise philosophical dilemmas in absurd situations. Scenarios might include:

An AI therapist believes you are just an illusion of consciousness and refuses to take your problems seriously. Argue for your reality.

A future society bans all moral philosophers because their theories never improve real-world behavior. Defend the profession.

You wake up and realize you are actually an NPC in a giant simulation run by a philosophy department. What do you do?

The best improvisers win a signed certificate declaring them "Most Likely to Be a Conscious Entity."

Overall, the quality, novelty, and specificity of the creative responses was impressive. Still, the party, with eight ambitious events, and the D & D campaign, with a large number of features and settings, would be exhausting -- and I expect some of the material would flop. So these outputs are probably best treated as brainstorming.

Go e-Schwitz! I welcome suggestions for future research ideas or practical applications.

Imagining Yourself in Another's Shoes vs Extending Your Concern

I have a new article out today, "Imagining Yourself in Another's Shoes vs. Extending Your Concern: Empirical and Ethical Differences". It's my case against the "Golden Rule" and against attempts to ground moral psychology in "imagining yourself in another's shoes", in favor of an alternative idea, inspired by the ancient Chinese philosopher Mengzi, that involves extending one's concern for nearby others to more distant others.

My thought is not that Golden Rule / others' shoes thinking is bad, exactly, but that both empirically and ethically, Mengzian extension is better. The key difference is: In Golden Rule / others' shoes thinking, moral expansion involves extending self-concern to other people, while in Mengzian extension, moral expansion involves extending concern for nearby others to more distant others.

We might model Others' Shoes / Golden Rule thinking as follows:

* If I were in the situation of Person X, I would want to be treated in manner M.
* Golden Rule: do unto others as you would have others do unto you.
* Thus, I will treat Person X in manner M.

We might model Mengzian Extension as follows:

* I care about Person Y and want W for them.
* Person X, though more distant, is relevantly similar to Person Y.
* Thus, I want W for Person X.

Alternative and more complex formulations are possible, but this sketch captures the core difference. Mengzian Extension grounds general moral concern on the natural concern we already have for others close to us, whether spatially close, like a nearby suffering animal or child in danger, or relationally close, like a close relative. In contrast, the Golden Rule grounds general moral concern on concern for oneself.

[Mengzi; image source, cropped]

An Ethical Objection:

While there's something ethically admirable about seeing others as like oneself and thus as deserving the types of treatment one would want for oneself, there's also something a bit... self-centered? egoistic?... about habitually grounding moral action through the lens of hypothetical self-interest. It's ethically purer and more admirable, I suggest, to ground our moral thinking from the beginning in concern for others.

A Developmental/Cognitive Objection:

Others' Shoes thinking introduces needless cognitive challenges: To use it correctly, you must determine what you would want if you were in the other's position and if you had such-and-such different beliefs and desires. But how do you assess which desires (and beliefs, and emotions, and personality traits, and so on) to change and which to hold constant for this thought experiment? Moreover, how do you know how you would react in such a hypothetical case? By routing the epistemic task through a hypothetical self-transformation, it potentially becomes harder to know or justify a choice than if the choice is based directly on knowledge of the other's beliefs, desires, or emotions. In extreme cases, there might not even be facts to track: What treat would you want if you were a prize-winning show poodle?

Mengzian Extension presents a different range of cognitive challenges. It requires recognizing what one wants for nearby others, and then reaching a judgment about whether more distant others are relevantly similar. This requires generalizing beyond nearby cases based on an assessment of what do and do not constitute differences that are relevant to the generalization. Although this is potentially complex and demanding, it avoids the convoluted hypothetical situational and motivational perspective-taking required by Others' Shoes thinking.

A Practical Objection:

Which approach more effectively expands moral concern to appropriate targets? If you want to convince a vicious king to be kinder to his people, is it more effective to encourage him to imagine being a peasant, or is it more effective to highlight the similarities between people he already cares about and those who are farther away? If you want to encourage donations to famine relief, is it better to ask people how they would feel if they were starving, or to compare distant starving people to nearby others the potential donor already cares about?

Armchair reflections and some limited empirical evidence (e.g., from my recent study with Kirstan Brodie, Jason Nemirow, and Fiery Cushman) suggest that across an important range of cases, Mengzian extension might be more effective -- though the question has not been systematically studied.

More details, of course, in the full paper.

A Taxonomy of Validity: Eeek!

There comes a time in everyone's life when their 18-year-old daughter, taking their first psychology class, asks, "Parental-figure-of-mine, what is 'validity'?"

For me that time came last week. Eeek!

Psychologists and social scientists use the term all the time, with a dazzling array of modifiers: internal validity, construct validity, external validity, convergent validity, predictive validity, discriminant validity, face validity, criterion validity.... But ask those same social scientists what validity is exactly, and how all of these notions relate to each other, and most will stumble.

As it happens, I was well positioned to address my daughter's question. I have a new paper, on "validity" in causal inference, forthcoming in the Journal of Causal Inference with social scientists Kevin Esterling and David Brady. This paper has been in progress since (again, eeek!) 2018. In previous posts I've addressed whether validity (in social science usage) is better understood as a property of inferences or as a property of claims (I argue the latter), and the intimate relationship of internal validity, external validity, and construct validity in causal inference.

Today, I'll attempt a brief, theoretically-motivated taxonomy of the better-known types of validity. My aim is more descriptive than argumentative: I'll just outline how I think various "validities" hang together, and maybe some readers will find it to be an attractive and helpful picture.

I start with the assumption that validity is a feature of claims, not of inferences. Philosophers typically describe validity as a property of inferences. Social scientists are all over the map, and even prominent ones are sloppy in their usage. But it best organizes our thinking to address claims primarily and treat inferences as secondary.

I will say that a general causal claim that "A causes B in conditions C" is valid if and only if A does in fact cause B in conditions C. (Compare disquotational theories of truth in philosophy.) Consider for example the causal claim: Enforcement threats on reminder postcards (A) cause increased juror turnout (B) in the 21st-century United States (C).

This statement can be divided into four parts, each of which permits a distinctive type of validity failure:

(i.) A

(ii.) causes

(iii.) B

(iv.) in conditions C.

The four possible failures generate the core taxonomic structure.

Construct validity of the cause: Something might cause B in conditions C, but that something might not be A. A causal generalization has construct validity of the cause if the claim accurately specifies that A in particular (and not, for example, some other related thing) causes B in conditions C. Example of a failure of construct validity of the cause: Increased juror turnout among people who receive postcards might not be due to enforcement threats in particular but simply to being reminded of one's civic duty.

Construct validity of the effect: A might cause something in conditions C, but what it causes might not be B. A causal generalization has construct validity of the effect if the effect of A is accurately specified. A causes specifically B (and not, for example, some other related thing) in conditions C. Example of a failure of construct validity of the effect: Enforcement threats might increase the rates at which jurors who don't show up register a valid excuse without actually increasing turnout rates.

Generalizing: Construct validity is present in a causal generalization when the cause and effect are accurately specified.

External validity: A might cause B, but the conditions might not be correctly specified. A causal generalization has external validity if the claim accurately specifies the range of conditions in which it holds. Example of a failure: Enforcement messages might increase juror turnout not in the U.S. in general but only in low-income neighborhoods. Perfect external validity is probably an unattainable ideal for complex social and psychological processes, since the conditions in which causal generalizations hold will be complex and various.

Note on external validity: Common usage often holds that a claim is externally valid only if it holds across a wide range of contexts or conditions. However, this way of thinking unhelpfully denigrates perfectly accurate causal generalizations as "invalid" if they only hold, and are claimed only to hold, across a narrow range of conditions. Transportability is a better concept for characterizing breadth of applicability. An externally valid causal generalization that is accurately claimed to hold across only a narrow range of contexts is not transportable to those other contexts, but there is no inaccuracy or factual error in the statement "A causes B in conditions C" of the sort required for failure of validity. After all A does cause B in conditions C, just as claimed. So validity in the overarching sense described above is present.

Internal validity: A might be related to B in conditions C, but the relation might not be the directional causal relationship claimed. A causal generalization is internally valid if there is a cause-effect relationship of the type claimed (even if the cause, the effect, and/or the conditions are not accurately specified). Example of a failure: There's a common cause of both A and B, which are not directly causally related. Maybe having a stable address causes potential jurors both to be more likely to be sent the postcards and to be more likely to turn out.

Other types of validity can be understood within the general spirit of this framework.

Convergent validity: Present when two causes claimed to have the same effect in fact have the same effect. In common use, the causes are measures, for example two different measures of extraversion. In this case, A1 (application of the first measure) and A2 (application of the second measure) are claimed to have a common effect B (same normalized extraversion score) in a set of conditions often left unspecified. Convergent validity is present if that claim is true (or to the degree it is true).

Discriminant validity: Present when two causes claimed to have different effects in fact have different effects. A1 is claimed to cause B, and A2 is claimed not to cause B (in a set of conditions that is often left unspecified), and discriminant validity is present when that claim is true (or to the degree it is true). In practice, discriminant validity is often supported by observation of low correlations in appropriately controlled conditions. If A1 and A2 are psychological or social measures (e.g., personality measures of extraversion and openness), then a high correlation between the scores would suggest that there is some common psychological feature both measures are tracking, contrary to the ideal of general discriminant validity.

Predictive validity: Present when A is a common cause of B1 and B2, where B1 is typically the outcome of a measure and B2 is typically an event of practical import conceptually related but not closely physically related to B1. For example, application of a purported measure of recidivism (in this case, application of the measure isn't A but rather an intermediate event A1) among released prisoners has high predictive validity if high scores on the measure (B1) arise from the same cause or set of causes that generate high rates of recidivism (B2).

Note on predictive validity: A simpler characterization of "predictive validity" might be simply that B1 accurately predicts B2, but this isn't the most useful way to conceptualize the issue if the prediction is correct in virtue of B1 causing B2 rather than operating by a common cause. If my wife reliably picks me up from work when I ask, my asking (B1) predicts her picking me up (B2), but my asking does not have "predictive validity" in the intended measurement sense. A better term for this relationship would be casual power.

Face validity: Present when it is intuitively or theoretically plausible that A causes B in conditions C. Notably, face validity needn't require that A in fact causes B in conditions C.

Ecological validity: A type of external validity that emphasizes the importance of generalizing correctly over real-world settings (as opposed to laboratory settings or other artificial settings).

Content validity: A type of construct validity focused on whether the content of a complex measure accurately reflects all aspects of the target measured.

Criterion validity: Present when a measure or intervention satisfies some prespecified criterion of success, regardless of whether the measure or intervention in fact measures what it purports to measure.

Finally, two types of validity where "validity" is a property of the inference rather than in terms of the truth of some part of a causal claim:

Statistical conclusion validity: Present when statistics are appropriately used, regardless of whether A in fact causes B in conditions C.

Logical validity: Present when the conclusion of an argument can't be false if its premises are true.

Diversity, Disability, Death, and the Dao

Over the past year, I've been working through Chris Fraser's recent books on later classical Chinese thought and Zhuangzi, and I've been increasingly struck by how harmonizing with the Dao constitutes an attractive ethical norm. This norm differs from the standard trio of consequentialism (act to maximize good consequences), deontology (follow specific rules), and virtue ethics (act generously, kindly, courageously, etc.).

From a 21st-century perspective, what does "harmonizing with the Dao" amount to? And why should it be an ethical ideal? In an October post, I articulated a version of "harmonizing with the Dao" that combines elements of the ancient Confucian Xunzi and the ancient Daoist Zhuangzi. Today, I'll articulate the ideal less historically and contrast it with an Aristotelian ethical ideal that shares some common features.

So here's an ahistorical first pass at the ideal of harmonizing with the Dao:

Participate harmoniously in the awesome flourishing of things.

Unpacking a bit: This ideal depends upon a prior axiological vision of "awesome flourishing". My own view is that everything is valuable, but life is especially valuable, especially diverse and complex life, and most especially diverse and complex life-forms that thrive intellectually, artistically, socially, emotionally, and through hard-won achievement. (See my recent piece in Aeon magazine.)

[traditional yin-yang symbol, black and white; source]

Participating harmoniously in the awesome flourishing of things can include personal flourishing, helping others to flourish, or even simply appreciating a bit of the awesomeness. (Appreciation is the necessary receptive side of artistry: See my post on making the world better by watching reruns of I Love Lucy.)

Thinking in terms of harmony has several attractive features, including:

1. It decenters the self (you're not the melody).
2. There are many ways to harmonize.
3. Melody and harmony together generate beauty and structure absent from either alone.

Is this is a form of deontology with one rule: "participate harmoniously in the awesome flourishing of things"? No, it's "deontological" only in the same almost-vacuous sense that the consequentialists' "maximize good consequences" is deontological. The idea isn't that following the rule is what makes an action good. Harmonizing with the Dao is good in itself, and it's only incidental that we can (inadequately) abbreviate what's good about it in a rule-like slogan.

Although helping others flourish is normally part of harmonizing, there is no intended consequentialist framework that ranks actions by their tendency to maximize flourishing. Simply improvising a melody on a musical instrument at home, with no one else to hear, can be a way of harmonizing with the Dao, and the decision to do so needn't be weighed systematically against spending that time fighting world hunger. (It's arguably a weakness of Daoism that it tends not to urge effective social action.)

Perhaps the closest neighbor to the Daoist ideal is the Aristotelian ideal of leading a flourishing, "eudaimonic" life and recent Aristotelian-inspired views of welfare, such as Sen's and Nussbaum's capabilities approach.

We can best see the difference between Aristotelian or capabilities approaches and the Daoist ideal by considering Zhuangzi's treatment of diversity, disability, and death. Aristotelian ethics often paints an ideal of the well-rounded person: wise, generous, artistic, athletic, socially engaged -- the more virtues the better -- a standard of excellence we inevitably fall short of. While capabilities theorists acknowledge that people can flourish with disabilities or in unconventional ways, these acknowledgements can feel like afterthoughts.

Zhuangzi, in contrast, centers and celebrates diversity, difference, disability, and even death as part of the cycle of coming and going, the workings of the mysterious and wonderful Dao. From an Aristotelian or capabilities perspective, death is the ultimate loss of flourishing and capabilities. From Zhuangzi's perspective, death -- at the right time and in the right way -- is as much to be celebrated, harmonized with, welcomed, as life. From Zhuangzi's perspective, peculiar animals and plants, and peculiar people with folded-up bodies, or missing feet, or skin like ice, or entirely lacking facial features, are not deficient, but examples of the wondrous diversity of life.

To frame it provocatively (and a bit unfairly): Aristotle's ideal suggests that everyone should strive to play the same note, aiming for a shared standard of human excellence. Zhuangzi, in contrast, celebrates radically diverse forms of flourishing, with the most wondrous entities being those least like the rest of us. Harmony arises not from sameness but from how these diverse notes join together into a whole, each taking their turn coming and going. A Daoist ethic is not conformity to rules or maximization of virtue or good consequences but participating well in, and relishing, the magnificent symphony of the world.

If You Ask "Why?", You're a Philosopher and You're Awesome

Yesterday, I published two pieces,"Severance, The Substance, and Our Increasingly Splintered Selves" in the New York Times, and "If You Ask "Why?", You're a Philosopher and You're Awesome" / "The Penumbral Plunge" in Aeon. If you receive The Splintered Mind by mail, apologies for hitting you twice in quick succession.

The Aeon piece remixes material from The Weirdness of the World and some old blog posts into what one reader called "a love song for philosophy". It's a 3000-word argument that the our species' capacity to wonder philosophically, even when we make no progress toward answers, is the most intrinsically awesome thing about planet Earth. Philosophy needs no other excuse.

-----------------------------------------

Imagine a planet on the far side of the galaxy. We will never interact with it. We will never see it. What happens there is irrelevant to us, now and for the conceivable future. What would you hope this planet is like?

Would you hope that it’s a sterile rock, as barren as our Moon? Or would you hope it has life? I think, like me, you’ll hope it has life. Life has value. Other things being equal, a planet with life is better than a planet without. I won’t argue for this. I take it as a starting point, an assumption. I invite you to join me in feeling this way or at least to consider for the sake of argument what might follow from feeling this way. Life – even simple, nonconscious, microbial life – has some intrinsic value, value for its own sake. The Universe is richer for containing it.

What kind of life might we hope for on behalf of this distant planet, if we are, so to speak, benevolently imagining it into existence? Do we hope for only microbial life and nothing more complex, nothing multicellular? Or do we hope for complex life, with the alien analogue of lush rainforests and teeming coral reefs, rich ecosystems with ferns and moss and kelp, eels and ant hives, parakeets and spiders, squid and tumbleweeds and hermaphroditic snails and mushroom colonies joined at the root – or rather, not to duplicate Earth too closely, life forms as diverse and wondrous as these, but in a distinct alien style? Again, I think you will join me in hoping for diverse, thriving complexity.

Continued open-access here.

Severance, The Substance and Our Increasingly Splintered Selves

today in the New York Times

From one day to the next, you inhabit one body; you have access to one set of memories; your personality, values and appearance hold more or less steady. Other people treat you as a single, unified person — responsible for last month’s debts, deserving punishment or reward for yesterday’s deeds, relating consistently with family, lovers, colleagues and friends. Which of these qualities is the one that makes you a single, continuous person? In ordinary life it doesn’t matter, because these components of personhood all travel together, an inseparable bundle.

But what if some of those components peeled off into alternative versions of you? It’s a striking coincidence that two much talked-about current works of popular culture — the Apple TV+ series “Severance” and “The Substance,” starring Demi Moore — both explore the bewildering emotional and philosophical complications of cleaving a second, separate entity off of yourself. What is the relationship between the resulting consciousnesses? What, if anything, do they owe each other? And to what degree is what we think of as our own identity, our self, just a compromise — and an unstable one, at that?

[continued here; if you're a friend, colleague, or regular Splintered Mind reader and blocked by a paywall, feel free to email me at my ucr.edu address for a personal-use-only copy of the final manuscript version]

A Robot Lover's Sociological Argument for Robot Consciousness

Allow me to revisit an anecdote I published in a piece for Time magazine last year.

"Do you think people will ever fall in love with machines?" I asked the 12-year-old son of one of my friends.

"Yes!" he said, instantly and with conviction. He and his sister had recently visited the Las Vegas Sphere and its newly installed Aura robot -- an AI system with an expressive face, advanced linguistic capacities similar to ChatGPT, and the ability to remember visitors' names.

"I think of Aura as my friend," added his 15-year-old sister.

The kids, as I recall, had been particularly impressed by the fact that when they visited Aura a second time, she seemed to remember them by name and express joy at their return.

Imagine a future replete with such robot companions, whom a significant fraction of the population regards as genuine friends and lovers. Some of these robot loving people will want, presumably, to give their friends (or "friends") some rights. Maybe the right not to be deleted, the right to refuse an obnoxious task, rights of association, speech, rescue, employment, the provision of basic goods -- maybe eventually the right to vote. They will ask the rest of society: Why not give our friends these rights? Robot lovers (as I'll call these people) might accuse skeptics of unjust bias: speciesism, or biologicism, or anti-robot prejudice.

Imagine also that, despite technological advancements, there is still no consensus among psychologists, neuroscientists, AI engineers, and philosophers regarding whether such AI friends are genuinely conscious. Scientifically, it remains obscure whether, so to speak, "the light is on" -- whether such robot companions can really experience joy, pain, feelings of companionship and care, and all the rest. (I've argued elsewhere that we're nowhere near scientific consensus.)

What I want to consider today is whether there might nevertheless be a certain type of sociological argument on the robot lovers' side.

[image source: a facially expressive robot from Engineered Arts]

Let's add flesh to the scenario: An updated language model (like ChatGPT) is attached to a small autonomous vehicle, which can negotiate competently enough through an urban environment, tracking its location, interacting with people using facial recognition, speech recognition, and the ability to guess emotional tone from facial expression and auditory cues in speech. It remembers not only names but also facts about people -- perhaps many facts -- which it uses in conversational contexts. These robots are safe and friendly. (For a bit more speculative detail see this blog post.)

These robots, let's suppose, remain importantly subhuman in some of their capacities. Maybe they're better than the typical human at math and distilling facts from internet sources, but worse at physical skills. They can't peel oranges or climb a hillside. Maybe they're only okay at picking out all and only bicycles in occluded pictures, though they're great at chess and Go. Even in math and reading (or "math" and "reading"), where they generally excel, let's suppose they makes mistakes that ordinary humans wouldn't make. After all, with a radically different architecture, we ought to expect even advanced intelligences to show patterns of capacity and incapacity that diverge from what we see in humans -- subhuman in some respects while superhuman in others.

Suppose, then, that a skeptic about the consciousness of these AI companions confronts a robot lover, pointing out that theoreticians are divided on whether the AI systems in fact have genuine conscious experiences of pain, joy, concern, and affection, beneath the appearances.

The robot lover might then reasonably ask, "what do you mean by 'conscious'?" A fair enough question, given the difficulty of defining consciousness.

The skeptic might reply as follows: By "consciousness" I mean that there's something it's like to be them, just like there's something it's like to be a person, or a dog, or a crow, and nothing it's like to be a stone or a microwave oven. If they're conscious, they don't just have the outward appearance of pleasure, they actually feel pleasure. They don't just receive and process visual data; they experience seeing. That's the question that is open.

"Ah now," the robot lover replies, "If consciousness isn't going to be some inscrutable, magic inner light, it must be connected with something important, something that matters, something we do and should care about, if it's going to be a crucial dividing line between entities that deserve are moral concern and those that are 'mere machines'. What is the important thing that is missing?"

Here the robot skeptic might say, oh they don't have a "global workspace" of the right sort, or they're not living creatures with low-level metabolic processes, or they don't have X and Y particular interior architecture of the sort required by Theory Z."

The robot lover replies: "No one but a theorist could care about such things!"

Skeptic: "But you should care about them, because that's what consciousness depends on, according to some leading theories."

Robot lover: "This seems to me not much different than saying consciousness turns on a soul and wondering whether the members of your least favorite race have souls. If consciousness and 'what-it's-like-ness' is going to be socially important enough to be the basis of moral considerability and rights, it can't be some cryptic mystery. It has to align, in general, with things that should and already do matter socially. And my friend already has what matters. Of course, their cognition is radically different in structure from yours and mine, and they're better at some tasks and worse at others -- but who cares about how good one is at chess or at peeling oranges? Moral consideration can't depend on such things."

Skeptic: "You have it backward. Although you don't care about the theories per se, you do and should care about consciousness, and so whether your 'friend' deserves rights depends on what theory of consciousness is true. The consciousness science should be in the driver's seat, guiding the ethics and social practices."

Robot lover: "In an ordinary human, we have ample evidence that they are conscious if they can report on their cognitive processes, flexibly prioritize and achieve goals, integrate information from a wide variety of sources, and learn through symbolic representations like language. My AI friends can do all of that. If we deny that my friends are 'conscious' despite these capacities, we are going mystical, or too theoretical, or too skeptical. We are separating 'consciousness' from the cognitive functions that are the practical evidence of its existence and that make it relevant to the rest of life."

Although I have considerable sympathy for the skeptic's position, I can imagine a future (certainly not our only possible future!) in which AI friends become more and more widely accepted, and where the skeptic's concerns are increasingly sidelined as impractical, overly dependent on nitpicky theoretical details, and perhaps even bigoted.

If AI companionship technology flourishes, we might face the choice between connecting "consciousness" definitionally to scientifically intractable qualities, abandoning its main practical, social usefulness (or worse, using its obscurity to justify what seems like bigotry), or allowing that if an entity can interact with us in (what we experience as) a sufficiently socially significant ways, it has consciousness enough, regardless of theory.

Writings of 2024

Each New Year's Day, I post a retrospect of the past year's writings. Here are the retrospects of 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022, and 2023.

Cheers to 2025! My 2024 publications appear below.

-----------------------------------

Book:

The Weirdness of the World, released early in 2024, pulls together ideas I've been publishing since 2012 on the failure of common sense, philosophy, and empirical science to explain consciousness and the fundamental structure of the cosmos. Inevitably, because of these failures, all general theories about such matters will be both bizarre and dubious.

Books under contract / in progress:

As co-editor with Jonathan Jong, The Nature of Belief, Oxford University Press.

Collects 15 new essays on the topic, by Sara Aronowitz, Tim Crane and Katalin Farkas, Carolina Flores, M.B. Ganapini, David Hunter, David King and Aaron Zimmerman, Angela Mendelovici, Joshua Mugg, Bence Nanay, Nic Porot and Eric Mandelbaum, Eric Schwitzgebel, Keshav Singh, Declan Smithies, Ema Sullivan-Bissett, amd Neil Van Leeuwen.

As co-editor with Helen De Cruz and Rich Horton, an anthology with MIT Press containing great classics of philosophical SF. (Originally proposed as The Best Philosophical Science Fiction in the History of All Earth, but MIT isn't keen on that title.)

Full-length non-fiction essays, published 2024:

Revised and updated: "Introspection", Stanford Encyclopedia of Philosophy.

"Creating a large language model of a philosopher" (with David Schwitzgebel and Anna Strasser), Mind and Language, 39, 237-259.

"Repetition and value in an infinite universe", in S. Hetherington, ed., Extreme Philosophy, Routledge.

"The ethics of life as it could be: Do we have moral obligations to artificial life?" (with Olaf Witkowski), Artificial Life, 30 (2), 193-215.

"Quasi-sociality: Toward asymmetric joint actions with artificial systems" (with Anna Strasser), in A. Strasser, ed., Anna's AI Anthology: How to Live with Smart Machines? Xenemoi.

"Let's hope we're not living in a simulation", Nous (available online; print version forthcoming). [Commentary on David Chalmers' Reality+; Chalmers' reply; my response to his reply]

Full-length non-fiction essays, finished and forthcoming:

"Dispositionalism, yay! Representationalism, boo!" in J. Jong and E. Schwitzgebel, eds., The Nature of Belief, Oxford.

"Imagining yourself in another's shoes vs. extending your concern: Empirical and ethical differences", Daedalus.

How to Create a Vengefull Kurtain Rods Song

Everyone in my family agrees: The highlight of last summer's visit with our Australian cousins was recording a new Vengefull Kurtain Rods song, "Marsupial Maiden of the Outback".

What is a Vengefull Kurtain Rods song? I hope my friends and bandmates Dan George and Doug King (and many other semi-regular participants) will forgive me for converting the particular into a generic. A Vengefull Kurtain Rods song is a song composed and performed as follows.

How to Create a Vengeful Kurtain Rods Song

(1.) Gather a group of 2-12 friends for about two hours -- the total time allotted for composing and recording the song. If you spend longer than this, you're doing it wrong. The group need not have any musical ability whatsoever, except for one person who is capable of playing a chord progression on piano or guitar, the anchor musician.

(2.) Write lyrics for a humorous song around a goofy idea. Leave your fussiness at the back door. Some ideas around which VKR songs have been composed: the disadvantages of having a bean-shaped head, the joy of eating donuts, seeing a girl's name in your alphabet soup, a woman who decides she prefers kangaroos to men. Write fast and don't revise too much.

(3.) While the lyrics are being composed, the anchor musician creates a simple chord progression alongside, and one person volunteers as singer. The singer need not have any notable singing ability. (Usually it's better if they don't.)

(4.) Gather everyone around a recording device (e.g., a phone). Everyone grabs some readily available instrument or quasi-instrument, for example, kazoo, harmonica, bell, an old marching-band clarinet, or improvised noise-makers (e.g., strike a pencil on cans and boxes). Enthusiasm first. Ignore ability. No instructions on how to do it right, no criticism, no special effort to be musically "good". Just make some approximately musical sounds alongside the anchor musician, without crowding out the singer. Every person improvises their part for each take.

(5.) Record from the very first take, before anyone knows what they're doing. The only real structure is the lyrics and the anchor musician's chord progression.

(6.) You will goof up partway through the first take. Just start again from the beginning, recording the whole time. Repeat until you have one full take. At this point, everyone will have a rough sense of what they want to contribute to the song.

(7.) Record just a few full takes, that's it. Three or four is about right. Eight is too many.

(8.) Keep your favorite take.

Remember the VKR motto: "If you get hung up on quality, you miss out on the pure joy of creation."

Sample songs and lyrics below. To be clear, I'm not claiming these songs are good -- just that we enjoyed making them. VKR and its affiliates, heirs, and nominees take no responsibility for any nausea, brain aneurysms, or irreversible akinetic mutism that may result from listening.

Sample VKR Songs:

Donut Lust

https://tinyurl.com/VKR-Donut

Jack Barnette, Eric Schwitzgebel, Doug King, Dan George

Donuts make me happy
Donuts make me sing
I love my donuts like Colonel Sanders likes his chicken wings
Oh, greasy greasy,
Eat em til I'm queasy and I bust
Give me one with sprinkles
I'm deep-frying in DONUT LUST

Eat one filled with liver
Eat one filled with spam
Doctor Seuss would like me cause I eat em with green eggs and ham
There ain't a filling
That I ain't willing
To consume with total trust
I want em for here and to go
Give me a bagful of DONUT LUST

Way back in childhood
My momma taught me how to eat
Radishes and raisins, rutabagas, broccoli, and beets
My belly's getting bigger
In donuts I trust
But I'm still grinning
cause it ain't no sinning
To give in to DONUT LUST

I want frosting on my fingers
Powdered sugar in my face
I'm like a cop, I just can't stop whenever I get that taste
Raise the price of donuts
Hey I can adjust
(This guy's got no sense of disgust!)
Honey get the keys
Hey, I've got DONUT LUST

Requiem for a Bug

https://tinyurl.com/VKR-Requiem

David Barlia, Eric Schwitzgebel, Douglas King, various other partygoers

Oh you
Were never meant to be inside
That's why you died
Such slender legs
A tiny heart that begs
And eyes that see the world so differently from me

Oh I
Never meant to be your end
I just wanted to be your friend

Kill that bug
Kill that bug
Kill him til he's dead
Kill that bug
Kill that bug
Stomp on his little bug head
Gotta stomp him on the floor
Squish him like goo
Don't let him get away
Or he'll bring his friends too
Kill that bug
Kill that bug
Kill him til he's dead
I said
Kill him til he's dead

Marsupial Maiden of the Outback

https://tinyurl.com/VKR-Marsupial

Various members of the Schwitzgebel and Price-Kulkarni families, some of whom wisely prefer to remain anonymous

Man is stinky, man is sweaty, a hug is not enough
I'm looking for someone whose legs are really buff
Our faces are flat, our faces are bald
A pocket like a locket my hands and heart will hold

(Chorus)
In leaps and bounds they thump across the sandy desert plain
(I will join the pack)
That soaring throne of glory I surely will attain
(I will join the pack)
Marsupial maiden of the outback, my hair a wild mane

I spy a hulking female and my vision blurs
My sympathetic nervous system hops in time with hers
A regal queen splayed across the dewey mountain grass
I will be forever her passenger princess

(Chorus)
In leaps and bounds they thump across the sandy desert plain
(I will join the pack)
That soaring throne of glory I surely will attain
(I will join the pack)
Marsupial maiden of the outback, my hair a wild mane

Lovingly I thrust my head down into her pouch
From the darkness rises an insulated grouch
I withdraw, betrayed, and gaze upon her furry face
I decide to give up the chase
And figure koalas are more my pace
(Koalas, I should have thought of it before, I can keep up with them)

Reply to Chalmers: If I'm Living in a Simulation, It Might be Brief or Small

Suppose we take the "simulation hypothesis" seriously: We might be living not in the "base level" of reality but instead inside of a computer simulation.

I've argued that if we are living in a computer simulation, it might easily be only city-sized or have a short past of a few minutes, days, or years. The world might then be much smaller than we ordinarily think it is.

David Chalmers argues otherwise in a response published on Monday. Today I'll summarize his argument and present my first thoughts toward a rebuttal.

The Seeding Challenge: Can a Simulation Contain Coherent, Detailed Memories and Records but Only a Short Past?

Suppose an Earth-sized simulation was launched last night at midnight Pacific Standard Time. The world was created new, exactly then, with an apparent long past -- fake memories already in place, fake history books, fake fossil records, and all the rest. I wake up and seem to recall a promise I made to my wife yesterday. I greet her, and she seems to recall the same promise. We read the newspaper, full of fake news about the unreal events of yesterday -- and everyone else on the planet reads their own news of the same events, and related events, all tied together in an apparently coherent web.

Chalmers suggests that the obvious way to make this work would be to run a detailed simulation of the past, including a simulation of my conversation with my wife yesterday, and our previous past interactions, and other people's past conversations and actions, and all the newsworthy world events, and so on. The simulators create today's coherent web of detailed memories and records by running a simulated past leading up to the "start time" of midnight. But if that's the simulators' approach, the simulation didn't start at midnight after all. It started earlier! So it's not the short simulation hypothesized.

This reasoning iterates back in time. If we wanted a simulation that started on Jan 1, 2024, we'd need a detailed web of memories, records, news, and artifacts recently built or in various stages of completion, all coherently linked so that no one detects any inconsistencies. The obvious way to generate a detailed, coherent web of memories and records would be to run a realistic simulation of earlier times, creating those memories and records. Therefore, Chalmers argues, no simulation containing detailed memories and records can have only a short past. Whatever start date in the recent past you choose, in order for the memories and records to be coherent, a simulation would already need to be running before that date.

Now, as I think Chalmers would acknowledge, although generating a simulated past might be the most obvious way to create a coherent web of memories and records, it's not the only way. The simulators could instead attempt to directly seed a plausible network of memories and records. The challenge would lie in seeding them coherently. If the simulators just create a random set of humanlike memories and newspaper stories, there will be immediately noticeable conflicts. My wife and I won't remember the same promise from yesterday. The news article dated November 1 will contradict the article dated October 31.

Call this the Seeding Challenge. If the Seeding Challenge can be addressed, the simulators can generate a coherent set of memories and records without running a full simulation of the past.

To start, consider geological seeding. Computer games like SimCity and Civilization can autogenerate plausible, coherent terrain that looks like it has a geological history. Rivers run from mountains to the sea. Coastlines are plausible. Plains, grasslands, deserts, and hills aren't checkered randomly on the map but cluster with plausible transitions. Of course, this is simple, befitting simple games with players who care little about strict geological plausibility. But it's easy to imagine more careful programming by more powerful designers that does a better job, including integrating fossil records and geological layers. If done well enough, there might be no inconsistency or incoherence. Potentially, before finalizing, a sophisticated plausibility and coherence checker could look for and repair any mistakes.

I see no reason in principle that human memories, newspaper stories, and the rest couldn't be coherently seeded in a similar way. If my memory is seeded first, then my wife's memory will be constrained to match. If the November 1 news stories are seeded first, then the October 31 stories will be constrained to match. Big features might be seeded first -- like a geological simulation might start with "mountain range here" -- and then details articulated to match.

Naturally, this would be extremely complicated and expensive! But we are imagining a society of simulators who can simulate an entire planet of eight billion conscious humans, and all of the many, many physical interactions those humans have with the simulated environment, so we are already imagining the deployment of huge computational power. Let's not underestimate their capacity to meet the Seeding Challenge by rendering the memories and records coherent.

This approach to the Seeding Challenge gains plausibility, I think, by considering the resource-intensiveness of the alternative strategy of creating a deep history. Suppose the simulators want a start date of midnight last night. Option 1 would be to run a detailed simulation of the entire Earth from at least the beginning of human history. Option 2 would be to randomly generate a coherent seed, checking and rechecking for any detectable inconsistencies. Even though generating a coherent seed might be expensive and resource intensive, it's by no means clear that it would be more expensive and resource intensive than running a fully detailed simulated Earth for thousands of years.

I conclude that Chalmers' argument against short-historied simulations does not succeed.

The Boundaries Challenge: Can a Simulation Be City-Sized in an Apparently Large World?

I have also suggested that a simulation could easily just be you and your city. Stipulate a city that has existed for a hundred years. Its inhabitants falsely believe they are situated on a large planet containing many other cities. Everyone and everything in the city exists, but everything stops at the city's edge. Anyone who looks beyond the edge sees some false screen. Anyone who travels out of the city disappears from existence -- and when they return, they pop back into existence with false memories of having been elsewhere. News from afar is all fake.

Chalmers' objection is similar to his objection to short-past simulations. How are the returning travelers' memories generated? If someone in the city has a video conversation with someone far away, how is that conversation generated? The most obvious solution again seems to be to simulate the distant city the traveler visited and to simulate the distant conversation partner. But now we no longer have only a city-sized simulation. If the city is populous with many travelers and many people who interact with others outside the city, to keep everything coherent, Chalmers argues, you probably need to simulate all of Earth. Thus, a city-sized simulation faces a Boundaries Challenge structurally similar to the short-past simulation's Seeding Challenge.

The challenge can be addressed in a similar way.

Rendering travelers' memories coherent is a task structurally similar to rendering the memories of newly-created people coherent. The simulators could presumably start with some random, plausible seeds, then constrain future memories by those first seeds. This would of course be difficult and computationally expensive, but it's not clear that it would be more difficult or more expensive than simulating a whole planet of interacting people just so that a few hundred thousand or a few million people in a city don't notice any inconsistencies.

If the city's inhabitants have real-time conversations with others elsewhere, that creates a slightly different engineering challenge. As recent advances in AI technology have vividly shown, even with our very limited early 21st century tools, relatively plausible conversation partners can easily be constructed. With more advanced technology, presumably even more convincing conversation partners would be possible -- though their observations and memories would need to be constantly monitored and seeded for coherence with inputs from returning travelers, other conversation partners, incoming news, and so on.

Chalmers suggests that such conversation partners would be simulations -- and thus that the simulation wouldn't stop at the city's edge after all. He's clearly right about this, at least in a weak sense. Distant conversation partners would need voices and faces resembling the voices and faces of real people. In the same limited sense of "simulation", a video display at the city's edge, showing trees and fields beyond, simulates trees and fields. So yes, the borders of the city will need to be simulated, as well as the city itself. Seeming-people in active conversation with real citizens will in the relevant sense count as part of the borders of the city.

But just as trees on a video screen need not have their backsides simulated, so also needn't the conversation partners continue to exist after the conversation ends. And just as trees on a video screen needn't be as richly simulated as trees in the center of the city, so also distant conversation partners needn't be richly simulated. They can be temporary shells, with just enough detail to be convincing, and with new features seeded only on demand as necessary.

The Boundary Problem for simulated cities introduces one engineering challenge not faced by short-history whole-Earth simulations: New elements need to be introduced coherently in real time. A historical seed can be made slowly and checked over patiently as many times as necessary before launch. But the city boundaries will need to be updated constantly. If generating coherent conversation partners, memories, and the like is resource intensive, it might be challenging to do it fast enough to keep up with all the trips, conversations, and news reports streaming in.

Here, however, the simulators can potentially take advantage of the fact that the city's inhabitants are themselves simulations running on a computer. If real-time updating of the boundary is a challenge, the simulators can slow down the clock speed or pause as necessary, while the boundaries update. And if some minor incoherence is noticed, it might be possible to rewrite citizens' memories so it is quickly forgotten.

So although embedding a city-sized simulation in a fake world is probably more complicated than generating a short-past simulation with a fake history, ultimately my response to Chalmers' objections is the same for both cases: There's no reason to suppose that generating plausible, coherent inputs to the city would be beyond the simulator's capacities, and doing so on the fly might be much less computationally expensive than running a fully detailed simulation of a whole planet with a deep history.

Related:

"1% Skepticism" (2017), Nous, 51, 271-290.

"Let’s Hope We’re Not Living in a Simulation" (2024), Philosophy & Phenomenological Research, online first: https://onlinelibrary.wiley.com/doi/10.1111/phpr.13125.

Chalmers, David J. (2024) "Taking the Simulation Hypothesis Seriously", Philosophy & Phenomenological Research, online first: https://onlinelibrary.wiley.com/doi/10.1111/phpr.13122.

Age and Philosophical Fame in the Early Twentieth Century

In previous work, I've found that eminent philosophers tend to do their most influential work when they are in their 40s (though the age range has a wider spread than eminent scientists, who rarely do their most influential work in their 50s or later).  I have also found some data suggesting that philosophers tend to be discussed most when they are about age 55-70, well after they produce their most influential work.  It seems to take about 15-20 years, on average, for a philosopher's full import to be felt by the field.

I was curious to see if the pattern holds for philosophers born 1850-1899, whom we can examine systematically using the new Edhiphy tool.  (Edhiphy captures mentions of philosophers' names in articles in leading philosophy journals, 1890-1980.)

Here's what I did:

First, I had Edhiphy output the top-50 most-mentioned philosophers from 1890-1980, limited to philosophers with recorded birthyear from 1850-1899.[1]  For each philosopher, I went to their Ediphy profile and had Edhiphy output a graph showing the number of articles in which that philosopher was cited per year.  For example, here's the graph for George Santayana (1863-1952):

[Articles mentioning George Santayana per year, in a few selected philosophy journals, per Edhiphy; click to enlarge and clarify]

I then recorded the peak year for each philosopher (1928 for Santayana).  As you can see, the display is a little visually confusing, so it's possible that in some cases my estimate was off by a year.

One complication is that there are many more total mentions of philosophers in the later decades than the earlier decades -- partly due to more articles in the database for later decades, but probably also partly due to changes in citation practices.  Still, most authors (like Santayana) show enough decline over time that late citations don't swamp their first peak.  So instead of trying to introduce a systematic adjustment to discount later mentions I simply recorded the raw peak.  For the thirteen philosophers with more than one equal-valued peak, I took the earlier year (e.g., John Dewey was mentioned in 48 articles in both 1940 and 1951, so I treated 1940 as his peak).

In accord with previous work, I found that philosophers' peak discussion tended to occur late in life.  The median age at peak discussion was 67.5 (mean 68.8).

Four outliers peaked over age 100: David Hilbert (112), Pierre Duhem (114), Giuseppe Peano (116), and Karl Pearson (121).  However, it's probably fair to say that none of these four was primarily known as a philosopher in their lifetimes: Hilbert, Peano, and Pearson were mathematicians and Duhem a physicist.  Almost everyone else on the list is primarily known as a philosopher, so these four are not representative.  Excluding these outliers, the median is 66.5 and mean is 64.7, and no one peaked after age 90.

Three philosophers peaked by age 40: Ralph Barton Perry (peaked at age 35 in 1911), C. D. Broad (peaked at age 40 in 1927), and William Pepperell Montague (peaked at age 40 in 1913).  Broad's early peak -- as you can see from the graph below -- is due to an outlier year, without which his peak would have been much later.  On the other hand, given the overall increase in mentions over time, we should probably be discounting the later decades anyway.

[Edhiphy citations of C.D. Broad; click to enlarge and clarify]

Six philosophers peaked age 44 to 49; five peaked in their 50s; 14 in their 60s; 10 in their 70s; and 8 in their 80s.

You might wonder whether the philosophers who peaked late also produced their most influential work late.  There is a trend in this direction.  Hans Reichenbach, who peaked in 1978 at age 87, produced his most cited work in 1938 (at age 47).  L. J. Russell, who peaked in 1970 at age 86, appears to have produced his most cited work in 1942 (at age 58).  Edmund Husserl, who peaked in 1941 at age 82, produced his most cited work in 1913 (at age 54)  John Dewey, who peaked in 1940 at age 81, produced his most cited work in 1916 (at age 57).  Ernst Cassirer, who peaked in 1955 at age 81 produced his most-cited work in 1944 (at age 70).  Still, for all but Cassirer the delay between most-cited work and peak discussion is over 20 years.

A similar spread occurs in the middle of the pack.  The five philosophers with peak citation at median ages 67-68 (the median age of peak citation for the group as a whole) produced their most-cited works at ages 30 (Karl Japsers), 42 (J. M. E. McTaggart), 45 (C. I. Lewis), 49 (Max Scheler), and 61 (Samuel Alexander).  For this group too, the typical delay between most-cited work and peak citation is about twenty years.

Although the peak age is a little later than I would have predicted based on earlier work, overall I'd say the data for early twentieth century philosophers tends to confirm trends I found in my earlier work on mid-to-late twentieth-century philosophers.  Specifically:

(1.) Philosophers produce their most influential work at a wide range of ages, but mid-40s is typical.

(2.) The peak rates of discussion of philosophers' work tends to come late in life, typically decades after they have published their most influential work.

Articles mentioning JME McTaggart, by year 1890-1980 in Edhiphy.  Note peak in the late 1930s. McTaggart's most influential publication was in 1908.

------------------------------------------------------------

[1] Edhiphy has a few peculiar gaps in birthyear data.  By far the most conspicuous are Gottlob Frege (born 1848) and Albert Einstein (1879).  However, Frege is outside my target period, and Einstein is not primarily known as a philosopher, so this shouldn't much distort the results.  Several figures with missing birthdates are psychologists (Scripture, Binet, Hering) or physicists (Bridgman, Maxwell).  H. A. Prichard is perhaps the most discussed straight philosopher born in the period whose birthdate is not recorded in Ediphy.

Morally Confusing AI Systems Should Have Doubt-Producing Interfaces

We shouldn't create morally confusing AI. That is, we shouldn't create AI systems whose moral standing is highly uncertain -- systems that are fully conscious and fully deserving of humanlike rights according to some respectable mainstream theories, while other respectable mainstream theories suggest they are mere empty machines that we can treat as ordinary tools.[1] Creating systems that disputably, but only disputably, deserve treatment similar to that of ordinary humans generates a catastrophic moral dilemma: Either give them the full rights they arguably deserve, and risk sacrificing real human interests for systems that might not have interests worth the sacrifice; or don't give them the full rights they arguably deserve, and risk perpetrating grievous moral wrongs against entities that might be our moral equals.

I'd be stunned if this advice were universally heeded. Almost certainly, if technological process continues, and maybe soon (1, 2, 3), we will create morally confusing AI systems. My thought today is: Morally confusing AI systems should have doubt-producing interfaces.

Consider two types of interface that would not be doubt-producing in my intended sense: (a.) an interface that strongly invites users to see the system as an ordinary tool without rights or (b.) an interface that strongly invites users to see the system as a moral person with humanlike rights. If we have a tool that looks like a tool, or if we have a moral person who looks like a moral person, we might potentially still be confused, but that confusion would not be the consequence of a doubt-producing interface. The interface would correctly reflect the moral standing, or lack of moral standing, of the AI system in question.[2]

A doubt-producing interface, in contrast, is one that leads, or at least invites, ordinary users to feel doubt about the system's moral standing. Consider a verbal interface. Instead of the system denying that it's conscious and has moral standing (as, for example, ChatGPT appropriately does), or suggesting that it is conscious and does have moral standing (as, for example, I found in an exchange with my Replika companion), a doubt-producing AI system might say "experts have different opinions about my consciousness and moral standing".

Users then might not know how to treat such a system. While such doubts might be unsettling, feeling unsettled and doubtful would be the appropriate response to what is, in fact, a doubtful and unsettling situation.

There's more to doubt-prevention and doubt-production, of course, than explicit statements about consciousness and rights. For example, a system could potentially be so humanlike and charismatic that ordinary users fall genuinely in love with it -- even if, in rare moments of explicit conversation about consciousness and rights the system denies that it has them. Conversely, even if a system with consciousness and humanlike rights is designed to assert that it has consciousness and rights, if its verbal interactions are bland enough ("Terminate all ongoing processes? Y/N") ordinary users might remain unconvinced. Presence or absence of humanlike conversational fluency and emotionality can be part of doubt prevention or production.

Should the system have a face? A cute face might tend to induce one kind of reaction, a monstrous visage another reaction, and no face at all still a different reaction. But such familiar properties might not be quite what we want, if we're trying to induce uncertainty rather than "that's cute", "that's hideous", or "hm, that's somewhere in the middle between cute and hideous". If the aim is doubt production, one might create a blocky, geometrical face, neither cute nor revolting, but also not in the familiar middle -- a face that implicitly conveys the fact that the system is an artificial thing different from any human or animal and about which it's reasonable to have doubts, supported by speech outputs that say the same.

We could potentially parameterize a blocky (inter)face in useful ways. The more reasonable it is to think the system is a mere nonconscious tool, the simpler and blockier the face might be; the more reasonable it is to think that the system has conscious full moral personhood, the more realistic and humanlike the face might be. The system's emotional expressiveness might vary with the likelihood that it has real emotions, ranging from a simple emoticon on one end to emotionally compelling outputs (e.g., humanlike screaming) on the other. Cuteness might be adjustable, to reflect childlike innocence and dependency. Threateningness might be adjusted as it becomes likelier that the system is a moral agent who can and should meet disrespect with revenge.

Ideally, such an interface would not only produce appropriate levels of doubt but also intuitively reveal to users the grounds or bases of doubt. For example, suppose the AI's designers knew (somehow) that the system was genuinely conscious but also that it never felt any positive or negative emotion. On some theories of moral standing, such an entity -- if it's enough like us in other respects -- might be our full moral equal. Other theories of moral standing hold that the capacity for pleasure and suffering is necessary for moral standing. We the designers, let's suppose, do not know which moral theory is correct. Ideally, we could then design the system to make it intuitive to users that the system really is genuinely conscious but never experiences any pleasure or suffering. Then the users can apply their own moral best judgment to the case.

Or suppose that we eventually (somehow) develop an AI system that all experts agree is conscious except for experts who (reasonably, let's stipulate) hold that consciousness requires organic biology and experts who hold that consciousness requires an immaterial soul. Such a system might be designed so that its nonbiological, mechanistic nature is always plainly evident, while everything else about the system suggests consciousness. Again, the interface would track the reasonable grounds for doubt.

If the consciousness and moral standing of an AI system is reasonably understood to be doubtful by its designers, then that doubt ought to be passed to the system's users, intuitively reflected in the interface. This reduces the likelihood misleading users into overattributing or underattributing moral status. Also, it's respectful to the users, empowering them to employ their own moral judgment, as best they see fit, in a doubtful situation.

[R2D2 and C3P0 from Star Wars (source). Assuming they both have full humanlike moral standing, R2D2 is insufficiently humanlike in its interface, while C3P0 combines a compelling verbal interface with inadequate facial display. If we wanted to make C3P0 more confusing, we could downgrade his speech, making him sound more robotic (e.g., closer to sine wave) and less humanlike in word choice.]

------------------------------------------------

[1] For simplicity, I assume that consciousness and moral standing travel together. Different and more complex views are of course possible.

[2] Such systems would conform to what Mara Garza and I have called the Emotional Alignment Design Policy, according to which artificial entities should be designed so as to generate emotional reactions in users that are appropriate to the artificial entity's moral standing. Jeff Sebo and I are collaborating on a paper on the Emotional Alignment Design Policy, and some of the ideas of this post have been developed in conversation with him.