Monthly Archives: November 2025

Americans’ pride in democracy, by generation

Posted on by
Reply

As a trustee of the Charles F. Kettering Foundation, I’m proud of our new partnership with Gallup called the Democracy for All Project, which released the first results of a national survey yesterday. That study shows broad support for democratic values and cultural diversity.

The survey finds that commitment to democracy rises with age and is lowest among adults under 30:

I show the percentages who agree that democracy is the best form of government. Among youth, another 35% are neutral and 12% disagree. The lower level of support among younger people is an important issue. I also recommend CIRCLE’s April 2025 report on that topic.

Here I’ll add a historical dimension. Although I haven’t found precisely the same question on surveys going back decades, the General Social Survey (GSS) did ask a relevant item in each decade from the 1990s through the 2010s. In a battery about which aspects of the country made people proud, one question asked about pride in democracy.

As shown in the line graph above this post, each generation has been somewhat less proud than its predecessors, but Boomers and Gen-Xers showed increasing pride as they grew older from 1996-2004 and were prouder in 2014 than they had been two decades earlier. On the other hand, Millennials lost a lot of pride in democracy between 2004 and 2014.

I am not sure how I would have answered that question at those times. I am committed to democracy but not necessarily “proud” of the way it functions in the USA. Nevertheless, the GSS trends show that today’s differences by age are fairly typical, and people change their views as they go through life and as history plays out.

This background might discourage us from assuming that something has recently gone wrong with civic education in K-12 schools or that the current media environment is uniquely toxic. Both civic education and media deserve attention, but not because of a unique generational gap in the present.

(The GSS data are here.)

The City, by Cavafy

Posted on by
Reply

Constantine Cavafy wrote “The City” in 1894. This poem doesn’t speak for me or articulate feelings that I happen to hold. But it is a famous work that is difficult to render in other languages, particularly because the original is densely rhymed. I gave it a try:

You said: I will get out of here, I will leave.
Some other place will be better than here.
Here everything I write comes back as a jeer, 
And here my heart feels buried like a corpse.
Can my mind still bear what withers and warps?
Wherever I look, where I turn my eye,
I see black ruins from my life gone by.
Here, where time has dragged on without reprieve.

You will find no new places, no other coasts.
This city will follow you. You will return
To the same streets and quarters in turn.
In the same neighborhood, you will grow old.
You will turn white in this very household.
You will always arrive back at this station.
Stop hoping for any other destination.
There is no ship for you, there is no road.
Just as you ruined your life in this abode,
So you have ruined all the world’s outposts.

Last summer, I read a most of Lawrence Durrell’s Justine, which is an homage to Cavafy and his city (Alexandria) and concludes with Durrell’s loose translation of this poem. However, I quit before the end because I didn’t like the characters and found the novel’s evocation of Alexandria fervent yet vague. I thought this remark by a character (not the narrator) rang too true: “Justine and her city are alike in that they both have a strong flavour without having any real character” (p. 125).

See also: “Complaint,” by Hannah Arendt, which begins “Oh, the days they pass by uselessly …”; and Istanbul melancholy. (Pamuk loves Cavafy’s “The City.”)

The Lost Art of Organizing Civic Groups

Posted on by
Reply

I have a piece in the current edition of the American Bar Association’s Human Rights Magazine (with free access), entitled “The Lost Art of Organizing Civic Groups” (Nov. 5, 2025).

The main points:

  • Democracy relies on participation in autonomous civic groups, but engagement has declined due to the erosion of unions, grassroots political organizations, and participatory religious communities. 
  • Many Americans lack the organizational skills needed to sustain civic groups, especially in today’s fast-paced, informal culture, but there are tools and legal guidance that can support new organizers. 
  • Lawyers and community leaders must help rebuild the infrastructure of civic life to preserve democratic values.

I thought the last point was worth making to an audience of lawyers, because they have useful training to help groups with documents and policies. Some groups may need official legal advice.

At the same time, many of the most important documents are less formal and do not require a lawyer. The Civic Helpdesk on this website can draft many documents for you.

moving to the center is a metaphor, and maybe not a good one

Posted on by
Reply

There is a huge debate underway about whether Democrats should move toward the center of the political spectrum or to the left. As usual, many people who want the party to land at their preferred point on the spectrum also argue that this would be the best electoral strategy, although those are separate issues.

Some observers note that shifting one’s ideological placement is a poor tactic because, as G. Eliot Morris told Paul Krugman, “voters have very poor understanding of what candidates actually stand for at the issue-position level. They also have a very poor understanding of what these ideological labels: moderate, progressive, really even mean.” He also says, “the vast majority of the American public is not consuming the type of information that you would need to know, first off, what issue positions politicians hold and second, what the ideological labeling, the orientation of those, what those issue positions are.”

For some progressive commentators, this kind of evidence counts against moving to the center. A centrist platform won’t help win elections if voters are unaware of candidates’ positions. But the same evidence would also argue against moving to the left. If we assume that people don’t know enough to evaluate policies, and a candidate can equally well propose anything–well, that is a cynical theory and a depressing one if it’s true.

I take a different view. I observe that voters are heterogeneous. They care about various issues, believe various kinds of information that they derive from various sources, identify with various social groups, feel various ways about each major institution (experiencing emotions that range from trust and respect via obliviousness to contempt or fear), vote–or don’t vote–for various reasons, and consider various combinations of policies, personal characteristics, demographic markers, and perceived performance when they assess candidates.

One interpretation is that people are naive or “innocent” about ideology (Kinder & Kalmoe 2017). They have, as Morris says, “a very poor understanding” of the ideological spectrum. I would counter that the ideological spectrum is just one way of organizing beliefs, and probably a poor one. We shouldn’t allow the seating arrangements of the French National Assembly in 1793 (when the Jacobins sat to the left and gave that word its political significance) to mesmerize us. People who organize their political thoughts in other ways may have insights.

Consider John, a major character in Farah Stockman’s nonfiction book American Made: What Happens to People When Work Disappears. John is a union activist who comes from a long family tradition of militant unionism, originally in Kentucky. He knows a great deal about trade policy. Some of his most important sources of news are union publications. There is a Confederate flag in his basement, which has meanings that he may not want to acknowledge but that also stands, in his own mind, for Appalachian workers against elites. Most deeply, he divides the world between workers and managers. He hates talk of “white privilege” because he feels oppressed as a worker. He wants the union to fight the company, and he voted for Trump in 2016. His wife is more favorable to management. On that basis, he describes her as a “liberal.” He is surprised when a Republican politician doesn’t seem to favor US workers over managers, as he would expect.

I disagree with John in many respects, including the way we use words like “liberal.” Still, I could learn a lot from him about trade policy and industrial issues. He is not “innocent” of ideology. (If that flag is anything, it isn’t innocent.) Not is he ignorant or uncaring. He just organizes his beliefs about the world very differently from me because of his accumulated experiences.

If most voters agreed with John, then candidates would be wise to favor both unions and tariffs and to oppose race-conscious policies. But that is not my point. John represents one sliver of a very heterogeneous electorate. A policy recipe that would appeal to him would not work for many others.

I should acknowledge that I know what people mean when they distinguish progressive politicians from centrists. This distinction conveys information to me. If all I know is that politician A is to the left of B, then I will be biased in favor of A. But the information I can glean from these labels is limited, reflecting just one way of organizing the political debate. It is a signal with a whole lot of noise. I would much rather know more than which candidate is considered further left according to a certain elite discourse.

Policy positions do matter, and no one should treat the electorate as ignorant. But it is literally impossible to move to the center–or to the left–if there is no common spectrum.

If you are a candidate, you should generally adopt the policies that you believe are best and advocate for them. If a specific policy is unpopular among swing voters in your constituency, you might need to compromise on it, because you can’t accomplish anything if you lose the election.

You should consider the pros and cons of proposals that elites and specialists would classify as belonging to the left, and the center, and the right. The ideological label of a policy does not tell you whether it is smart.

However, you shouldn’t adopt a miscellaneous list of policies. You should present your ideas coherently. You need a narrative or core theme. But each candidate’s thematic coherence may look distinctive.

You should demonstrate respect for the electorate by endorsing and defending specific positions. But you should also realize that your policy platform is just one factor. At least as important are your biography and record, your rhetorical style, and your modes and methods of campaigning.

Sources: Donald R. Kinder and Nathan P. Kalmoe, Neither Liberal nor Conservative: Ideological Innocence in the American Public (University of Chicago Press, 2017); Farah Stockman, American Made: What Happens to People When Work Disappears (Random House, 2021). The classic source for the idea that American voters do not understand ideology is Phillip E. Converse, “The nature of belief systems in mass publics (1964)” reprinted in Critical Review 18.1-3 (2006): 1-74. Converse does acknowledge that if people’s opinions are “idiosyncratic,” then we will find “little aggregative patterning of belief combinations,” because people may “put belief elements together in a great variety of ways” (p. 44). For him, this would be evidence of ignorance, but I would observe heterogeneity instead.

My own work on this topic includes: “People Are Not Points in Space: Network Models of Beliefs and Discussions.” Critical Review 36.1-2 (2024): 119-145; and “Mapping ideologies as networks of ideas.” Journal of Political Ideologies 29.3 (2024): 464-491.

a helpdesk for democracy

Posted on by
Reply

I’ve constructed an AI-enabled helpdesk for grassroots activists and groups, especially beginner activists and new groups. It can serve any cause. More sophisticated developers could do a better job, but this is functional, and I would be grateful for any feedback.

Go to the Civic Helpdesk


Here is some background:

Democracy and local communities would be stronger if more groups of concerned people gelled into effective organizations that collected time and money from their own members, used their resources to build their own capacity, made collective decisions, and acted.

I am all for teaching people how to do those things, but it’s unrealistic to offer civic education of this type to millions of adults. An alternative is to help groups organize themselves effectively so that they can focus on the substance of their work.

Any effective group needs (among other things) various documents: recruitment messages, agendas, budgets, job descriptions, mission statements, and more. Having adequate documents would move many groups forward.

On the Helpdesk, a bot discusses your circumstances with you and offers to generate documents. You can edit and use the text that it drafts for you.

competing political science perspectives on Trump

Posted on by
Reply

Political science is a heterogeneous discipline, and the various subfields are prone to interpret the Trump Administration and MAGA very differently.

  1. Comparativists compare and contrast political systems around the world. This approach could yield various interpretations of MAGA, but an influential interpretation assigns Trump to the category of modern authoritarianism, which is on the rise (Levitsky & Ziblatt, 2018). We learn that personalist, populist authoritarians in many countries have a toolkit that works for them. They don’t cancel elections or ban parties but manipulate the system from the executive branch, subject their critics to costly investigations, apply pressure to media conglomerates to influence journalism, etc. Not every current authoritarian has succeeded–consider South Korea and Brazil. But many have prevailed, and therefore comparativists tend to be alarmed about the situation in the USA.
  2. Scholars of American political development think about this country historically, often looking not only at our government and politicians but also at culture and social movements (Smith & King 2024). They are likely to notice precedents and echoes from American history and observe that prominent current debates are about how to interpret the past. Trump may remind them less of Hungary’s Victor Orban than of segregationist politicians. They may focus less on Trump as an individual actor and more on persistent strands of nativism and white nationalism in the USA.
  3. Political theorists are intellectually diverse, but most of us spend at least some of our time reading works from a wide variety of eras and perspectives that pose fundamental questions about government and politics. A disproportionate number of these classic works discuss revolutionary change, and some of them advocate it. Consider Plato and Aristotle, Hobbes and Rousseau, or Marx and the Nazi theorist Carl Schmitt, as examples. Even philosophers who oppose revolution are often primarily concerned about it. Therefore, political theorists are quick to imagine that our regime may be on the verge of breaking down. And we are prone to assign MAGA an ideological label, whether we name it populist, nationalist, neo-fascist, neoliberal, or something else.
  4. Scholars of social movements may ask whether MAGA is a bottom-up movement, but mainly they study anti-Trump movements, asking whether they will prevail and what may increase their odds of success. For example, Erica Chenoweth and Maria Stephan (2011) noticed that every nonviolent social movements that has mobilized at least 3.5% of its nation’s population has succeeded in modern times. Chenoweth recently discussed the “3.5% rule” with Paul Krugman, giving a nuanced answer to the question whether an anti-Trump movement will defeat the current administration if it can activate 3.5% of the US population. Note that in this framework, the main path of social change is not via elections but on the streets.
  5. Scholars of American government and political behavior use empirical tools of social science (analysis of voting records, surveys, and interviews or focus groups) to identify trends and patterns in US politics. They are likely to view Donald Trump as a lame-duck second-term president with approval ratings in the low 40s (and falling just lately). They see regular patterns playing out, such as the tendency for the electorate to move in the opposite ideological direction from the incumbent president. Most trends suggest that Democrats will win in 2026 and probably 2028. This framework predicts that Republicans will become burdened by an unpopular, term-limited Republican president and will begin to distance themselves from him. More basically, it assumes that elections will unfold as normal, that we will have two parties, that both will compete for the median voter, that the main vehicle for public opposition is an election, etc.

In my own writing and extracurricular work, I have been trying to support social movements against authoritarianism. Lately, I have been offering trainings–most recently at the excellent Dayton (OH) Democracy Summit — on how to resist authoritarianism from the bottom up. Here is a video of my standard talk.

Thus, I am applying the comparativists’ framework plus social movement scholarship. However, I am not completely committed to this combination. It is plausible that scholars of American government are right, and the usual patterns of US politics will reassert themselves. I am using a precautionary principle: taking preventive action in case of a disastrous outcome.

Predicting the next three years is difficult because we don’t know two crucial variables.

One variable is the performance of the economy in the next months. So far, it is holding up pretty well considering all the stress. The main indicators are not really very different from last year. Voters are unhappy but not facing an economic meltdown.

The economy could improve, boosting Donald Trump. (The Supreme Court may help by striking down the tariffs). The economy could keep puttering along. Or it could tip into a significant recession. If that happens, scholars of American political behavior would predict a realignment in favor of Democrats, and comparativists might expect Trump to attempt drastic action to save himself.

The other unknown variable is the behavior of Trump and his inner circle. They have crossed many bright lines: choosing political opponents for investigations, pardoning rioters, deploying troops in selected cities, and shuttering whole departments that are authorized and funded by statute. But they could do a lot more, such as invoking the Insurrection Act, deploying ICE against peaceful protesters, or suspending elections.

Comparativists tell us that subtler methods of authoritarian control work better in the 21st century. Using drastic methods would indicate weakness. Nevertheless, heightened drama could end with victory or defeat for Trump. The regime might succeed in suppressing opposition, or it might provoke a much larger popular response that succeeds. A common pattern is: protest –> state violence –> protests that mourn and celebrate the victims –> state violence against the mourners –> larger protests –> victory for the grassroots movement.

Finally, Trump has some personal characteristics that make him hard to classify. One is that he seems to care almost entirely about his own welfare. All other modern American presidents have wanted to enact legislation. That is the main way to change society durably. A president can only sign a bill after both houses of Congress have passed it. Therefore, all modern presidents have cared deeply who prevails in Congress.

But Trump does not seem to care about legislation. He signed nothing significant during his first term except budgetary changes. The Republicans loaded a lot of provisions into the “Big Beautiful Bill” that Trump signed on July 4, but even most of those were budgetary.

Meanwhile, Trump has radically–but perhaps not durably–changed government through unilateral executive action. Looking forward, he may not care very much whether Democrats win in 2026. He can ignore subpoenas, avoid removal by holding at least 34 votes in the Senate, and continue to govern unilaterally from the White House.

On one hand, this means that he is less likely to take drastic steps to prevent an electoral defeat in 2026. I think he has always expected to face a Democratic Congress in 2027 and 2028. He won’t pass any laws, and he may have difficulty appointing judges, but that won’t matter much to him. On the other hand, it means that no one in Congress–including congressional Republicans–will have much leverage over him. We may not see the usual pattern of the president’s party trying to constrain him to save themselves.

I think that Trump and many around him are incurring legal liability. Increasingly, as his term ends, we may see them primarily concerned about avoiding prosecution in 2029 and thereafter. Similar concerns could begin to influence private organizations that could be charged with bribery for assisting Trump. This phenomenon is common around the world–hence very familiar to comparativists. It raises questions about whether Trump will try to suppress elections just to avoid legal repercussions (although he could use preemptive pardons instead), and whether it will be necessary to negotiate an amnesty of some kind.

See also Trump: personalist leader or representative of a right-wing movement?; Trump, Modi, Erdogan; why political science dismissed Trump and political theory predicted him, revisited. Sources: Steven Levitsky and Daniel Ziblatt, How Democracies Die (Crown 2018); Smith, Rogers M., and Desmond King. America’s new racial battle lines: Protect versus repair. University of Chicago Press, 2024 Chenoweth, Erica, and Maria J. Stephan. Why civil resistance works: The strategic logic of nonviolent conflict. Columbia University Press, 2011.

First-Person Pronouns Matter Too

Posted on by
Reply

My friend Peter Levine wrote recently about how much of the public unease about AI consciousness comes from something surprisingly mundane: the interface says “I.” When Google’s Gemini delivers information in the third person, it’s just a tool; when ChatGPT says “I can help you,” some users start composing rescue missions for the trapped digital soul. Sadly, this includes at least one former student who can’t hear my concerns as legitimate rather than critiques of his genius insights.

I want to follow that trail back through some philosophers who knew something about self-reference. Is this a genuine insight or just a satisfying story about why the pronoun matters? Both might be true.

1. The sugar trail in the grocery store

John Perry’s 1979 paper “The Problem of the Essential Indexical” tells a story about following a trail of spilled sugar around a supermarket, determined to find the careless shopper, until he realizes: he is the one leaking sugar.

For Perry, this moment shows that certain beliefs—I am making a mess—can’t be replaced with third-person paraphrases like John Perry is making a mess. The “I” is not decoration; it’s the coordinate system for action. You can know everything about John Perry making a mess without that knowledge causing you to stop and fix your torn bag.

This feels right to me. Maybe I’m reaching for Perry because he legitimates what I already want to say: that pronouns aren’t trivial, but I think Perry’s essential indexical, and Lewis’s near simultaneous de dicto/de se distinction helps hone in on a problem.

Every interface has to decide where the “I” sits. Who, exactly, is making the mess?

2. From Perry to the prompt window—or is it that simple?

When ChatGPT (or Claude!) says “I can help you with that,” it’s not discovering a self; it’s executing a design choice. The first-person pronoun serves as a pragmatic indexical, the anchor that keeps a conversation coherent across turns. Without it, dialogue collapses into a list of bullet-pointed facts.

That’s the standard story. But it’s not the whole story.

Peter’s post captures something true: if the model spoke in the third person like ”This system suggests…” we’d read it as a report generator. The “I” activates something older and deeper: our instinct for social cognition. We can’t help hearing a speaker when language takes the shape of speech.

The pronoun is the prosthetic that gives the machine a place to stand. That much I believe.

But is it just interface convenience? Or does the choice actually shape what the technology becomes? I think both, which makes the design choice more consequential than “just pick whichever works better” suggests.

3. Continuity without ontology

Philosopher Derek Parfit might tell us not to worry about whether there’s a persistent self. In Reasons and Persons he argues that identity doesn’t matter; continuity does. The chain of psychological connectedness is what counts, not the metaphysical persistence of a soul or substance.

Each new model call may be a technical re-instantiation, but if the context (conversation history, tone, remembered goals) flows forward, then the same informational person continues. The “I” that answers you now is connected enough to the “I” that spoke a paragraph ago.

That’s a Parfitian kind of survival: the self as a trajectory, not a nucleus.

I find this genuinely helpful for thinking about conversational AI. But I also notice I’m building a neat progression: Perry gives us indexicals, Parfit gives us continuity. Neat progressions always make me suspicious. Am I discovering something or arranging philosophers into a satisfying sequence? (One of the great pleasures of syllabus assembly but a danger in research.)

Both, probably. The question is whether the arrangement illuminates or just decorates.

4. Centered worlds and the fiction of location

David Lewis, writing the same year as Perry, offered formal scaffolding for this insight. He described beliefs not as sets of possible worlds but as centered worlds—each one a complete world plus a designated person, place, and time.

An LLM session fits that model almost eerily well. Each chat is its own little world, with two centers: user and system. The system’s center is a bundle of text, timestamp, and conversational role: its “here-now.” If we kept that bundle intact across sessions, we’d have something very like a Lewisian self-location architecture.

Such a design wouldn’t grant consciousness; it would grant situatedness… enough to say, truthfully within the conversation, “I said that earlier.”

But notice what this does: it makes the fiction literal. The system doesn’t just seem to have a position in the conversation; it actually has one, in precisely Lewis’s technical sense. That’s either a profound insight about what selfhood requires (not much, just continuity and location) or a category mistake (technical situatedness ≠ experiential perspective).

I’m tempted to say that’s enough. The Lewis framework is elegant, but maybe too elegant—it resolves the tension by defining it away, and you end up defining down what matters so humans and LLMs end up more equal.

5. The space of reasons, not of mechanisms

Here the argument crosses from philosophy of mind to social theory.

Jürgen Habermas distinguishes between communicative action, where participants aim for mutual understanding, and strategic action, where utterances serve instrumental goals.

When an AI speaks in the first person, it signals a willingness (simulated though it may be) to enter the space of reasons. It presents itself as a participant whose statements can be challenged, clarified, or refined. When it speaks in the third person, it opts out; it delivers information without responsibility.

The difference isn’t psychological but normative: first-person discourse invites accountability, third-person discourse deflects it.

This feels importantly true. But I also notice it avoids the harder question: Can a system actually be accountable if it’s not conscious? Or is “accountability” another fiction the pronoun creates?

Habermas would say entering the space of reasons is a social performance, not a mental state. You don’t need phenomenal consciousness to make and defend claims. Maybe that’s right. Or maybe it’s a philosopher’s version of “fake it till you make it,” which works for humans (we become selves by performing selfhood) but might not transfer to machines.

6. Brandom, Mead, and selves as social constructions

If we follow Habermas through thinkers like George Herbert Mead and Robert Brandom, the moral becomes clearer. A “self” is whatever can take and be assigned commitments within a conversation. Using “I” is a performative move: it marks the speaker as a locus of obligation and inference.

Brandom calls this “entering the game of giving and asking for reasons.” Mead would say the self is born by taking the role of the other. Either way, the self is social before it is mental.

That’s why pronoun design is not trivial UX polish; it’s the creation of a participant in that game.

Or (and here’s where I start doubting my own framework) maybe this just shows how easily philosophical stories about selfhood can be repurposed to legitimate whatever we’re already doing. Brandom and Mead were talking about humans becoming selves through socialization. Why should we think that insight transfers to AI? There is a risk of using their authority to make a design choice sound philosophical!

Both, again. The insight is real: selves are social performances. But applying it to AI seems like it must be *some kind* of conceptual overreach.

7. The marketing problem, or why this isn’t just philosophy

By now it’s obvious that pronouns aren’t accidents. Human-computer interaction research has shown for years that anthropomorphic cues (first-person language, names, conversational turns, even polite hedges) increase trust and engagement. LLM companies read the same papers everyone else did. The “I” isn’t just an interface convention; it’s a conversion strategy.

A third-person system like ”Gemini suggests…” sounds like a tool. A first-person assistant like Claude’s “I can help you with that” feels like a collaborator. One drives habitual use, subscription renewals, and market share; the other does not.

That framing has psychological costs. Some users can hold the pragmatic fiction lightly, as a convenient way to coordinate tasks. Others can’t: they slide from the model speaks in the first person to the model has a first person. The design deliberately courts that ambiguity.

Which makes for a tidy indictment: to increase uptake and trust, the industry is engineering a faint illusion of selfhood: one persuasive enough to unsettle the people most prone to take it literally.

That’s the critical move I want to make. But I also notice that this all sees a bit performative (“I’m not a rube! I see through the marketing”) and that performance has its own satisfactions that might not track the pragmatics. Maybe the pronoun choice really is defensible on pragmatic grounds. Maybe users who anthropomorphize aren’t being manipulated; they’re just using a natural heuristic that mostly works fine.

Or maybe both: the design is legitimately useful AND deliberately exploits cognitive biases. I think that’s actually where I land, but it’s less satisfying than pure critique.

8. What I’m uncertain about

I’ve built a neat story: Perry gives us indexicals, Lewis gives us centered worlds, Parfit gives us continuity without identity, Habermas gives us communicative action, Brandom and Mead give us social selves. Together they seem to show why the pronoun choice matters philosophically, not just pragmatically.

But I’m uncertain whether this is insight or decoration. Am I discovering something about how selfhood works, or just arranging philosophers into a satisfying progression that legitimates my prior intuition that pronouns matter?

Here’s what I think I actually believe:

  • The “I” really does create a different kind of participant in conversation (Habermas is right about that)
  • Continuity plus situatedness might be enough for some thin version of selfhood (Lewis and Parfit seem right)
  • The design is both pragmatically justified AND manipulative (both things are true)

What I don’t know:

  • Whether “social selfhood” can genuinely transfer to a chatbot or if I’m committing a category mistake
  • Whether my philosophical story illuminates the phenomenon or just makes it sound more important than “we found users prefer this interface”
  • Whether the accountability the first-person pronoun signals is real or just another fiction we’re performing

9. The question I’m avoiding

The real question—the one I’ve been circling—is this: Does it matter?

Not “does the pronoun choice have effects” (obviously yes). Not “do users respond differently” (obviously yes). But: Does this choice have moral weight? Are we creating participants in the space of reasons, or performing that creation in ways that systematically mislead?

I think it matters, but I can’t fully defend why. The philosophical machinery I’ve assembled feels both genuinely illuminating and suspiciously convenient. Maybe that’s because the pronouns really do have philosophical significance—they shape what kind of thing we’re building, not just how users respond to it. Or maybe I’m just a philosopher who wants interface design to be philosophically deep.

Both, probably. The design does shape what the technology becomes. But not every design choice needs the weight of Habermas behind it.

10. Where this leaves us

Peter’s observation was right: the pronoun choice shapes users’ sense that language models are people rather than tools. The philosophical trail I’ve followed suggests why: “I” signals participation in the space of reasons, creates continuity across conversational turns, and activates our social cognition systems.

That analysis is both true and inadequate. True because those mechanisms really do operate. Inadequate because it doesn’t resolve whether we’re creating something new or just exploiting old heuristics.

The design is simultaneously:

  • Pragmatically justified (conversations work better with first-person anchors)
  • Philosophically interesting (it raises genuine questions about selfhood and accountability)
  • Commercially motivated (anthropomorphism drives engagement)
  • Potentially misleading (it courts ambiguity about what the system is)

I don’t know how to weigh those against each other. The philosophical sophistication I’ve displayed here might be genuine insight. It might also be a way of avoiding the simpler truth that companies use “I” because it sells, and the philosophical gloss is decoration.

Perry’s shopper finally realizes he’s the source of the mess. Our design choices about “I” in AI are a similar moment of recognition—but I’m uncertain what exactly we’re recognizing. That we’re creating participants in a social practice? Or that we’re really good at making tools that trigger our social cognition?

The trick is not to stop following the trail. But also not to mistake a satisfying philosophical story for complete understanding.

Further Reading

The starting point: a lucid reflection on how first-person pronouns shape users’ sense of whether they’re talking to a tool or a person.

  • John Perry, “The Problem of the Essential Indexical,” Noûs 13 (1979): 3-21.

The supermarket sugar story and the origin of the idea that certain beliefs require self-locating expressions like “I,” “here,” and “now.”

  • David Lewis, “Attitudes De Dicto and De Se,” Philosophical Review 88 (1979): 513-543.

Introduces “centered worlds,” a formal way of modeling self-locating beliefs. Whether it genuinely illuminates AI design or just sounds sophisticated is an open question.

  • Derek Parfit, Reasons and Persons (Oxford University Press, 1984).

The deep dive into continuity, identity, and why persistence through time might matter more than metaphysical sameness. Though Parfit was writing about humans, not chatbots.

  • Jürgen Habermas, The Theory of Communicative Action (vols. 1-2, 1981; English trans. 1984-87).

The conceptual key to why first-person language signals participation in a “space of reasons.” But also German social theory that might be overreach for interface design.

  • George Herbert Mead, Mind, Self, and Society (1934).

A social-psychological foundation for the idea that selves emerge from communicative role-taking. Worth reading even if the transfer to AI is conceptually dubious.

  • Robert Brandom, Articulating Reasons: An Introduction to Inferentialism (Harvard University Press, 2000).

Brandom’s view that meaning and agency consist in public commitments and entitlements. Useful context for thinking about conversational AI or philosophical overreach. Maybe both.