Category Archives: Uncategorized

can AI solve “wicked problems”?

Leave a reply

I’ve been reading predictions that artificial intelligence will wipe out swaths of jobs–see Josh Tyrangiel in The Atlantic or Jan Tegze. Meanwhile, this week, I’m teaching Rittel & Webber (1973), the classic article that coined the phrase “wicked problems.” I started to wonder whether AI can ever resolve wicked problems. If not, the best way to find an interesting job in the near future may be to specialize in wicked problems. (Take my public policy course!)

According to Rittel & Webber, wicked problems have the following features:

  1. They have no definitive formulation.
  2. There is no stopping rule, no way to declare that the issue is done.
  3. Choices are not true or false, but good or bad.
  4. There is no way to test the chosen solution (immediate or ultimate).
  5. It is impossible, or unethical, to experiment.
  6. There is no list of all possible solutions.
  7. Since each problem is unique, inductive reasoning can’t work.
  8. Each problem is a symptom of another one.
  9. You can choose the explanations, and they affect your proposals.
  10. You have no “No right to be wrong.” (You are affecting other people, not just yourself. And the results are irreversible.)

Rittel and Webber argue that those features of wicked problems deflate the 20th-century ideal of a “planning system” that could be automated:

Many now have an image of how an idealized planning system would function. It is being seen as an on-going, cybernetic process of governance, incorporating systematic procedures for continuously searching out goals; identifying problems; forecasting uncontrollable contextual changes; inventing alternative strategies, tactics, and time-sequenced actions; stimulating alternative and plausible action sets and their consequences; evaluating alternatively forecasted outcomes; statistically monitoring those conditions of the publics and of systems that are judged to be germane; feeding back information to the simulation and decision channels so that errors can be corrected–all in a simultaneously functioning governing process. That set of steps is familiar to all of us, for it comprises what is by now the modern-classical mode planning. And yet we all know that such a planning system is unattainable, even as we seek more closely to approximate it. It is even questionable whether such a planning system is desirable (p. 159)

Here they describe planning systems that would have been very labor-intensive in 1973, but many people today imagine that this is how AI works, or will work.

why are problems wicked?

Some of the 10 reasons that some problems are “wicked,” according to Rittel & Webber, relate to the difficulty of generating knowledge. Policy problems involve specific things that have many features or aspects and that relate to many other specific things. For example, a given school system has a vast and unique set of characteristics and is connected by causes and effects to other systems and parts of society. These qualities make a school system difficult to study in conventional, scientific ways. However, could a massive LLM resolve that problem by modeling a wide swath of the society?

Another reason that problems are wicked is that they involve moral choices. In a policy debate, the question is not what would happen if we did something but what should happen. When I asked ChatGPT whether AI will be able to resolve wicked problems, it told me no, because wicked problems “are value-laden.” It added, “AI can optimize for values, but it cannot choose them in a legitimate way. Deciding whose values count, how to weigh them, and when to revise them is a normative, political act, not a computational one.”

Claude was less explicit about this point but emphasized that “stakeholders can’t even agree on what the problem actually is.” Therefore, an AI agent cannot supply a definitive answer.

A third source of the difficulty of wicked problems involves responsibility and legitimacy. In their responses to my question, both ChatGPT and Claude implied that AI models should not resolve wicked problems because they don’t have the right or the standing to do so.

what’s our underlying theory of decision-making?

Here are three rival views of how people decide value questions:

First, perhaps we are creatures who happen to want some things and abhor other things. We experience policies and their outcomes with pleasure, pain, or other emotions. It is better for us to get what we want–because of our feelings. Since an AI agent doesn’t feel anything, it can’t really want anything; and if it says it does, we shouldn’t care. Since we disagree about what we want, we must decide collectively and not offload the decision onto a computer.

Some problems with this view: People may want very bad things–should their preferences count? If we just happen to want various things, is there any better way to make decisions than to maximize as many subjective preferences as possible? Couldn’t a computer do that? But would the world be better if we did maximize subjective preferences?

In any case, you are not going to find a job making value-judgments. Today, lots of people are paid to make decisions, but only because they are assumed to know things. Nobody will pay for preferences. Life works the other way around: you have to pay to get your preferences satisfied.

Second, perhaps value questions have right and wrong answers. A candidate for the right answer would be utilitarianism: maximize the total amount of welfare. Maybe this rule needs constraints, or we should use a different rule. Regardless, it would be possible for a computer to calculate what is best for us. In fact, a machine can be less biased than humans.

Some problems with this view: We haven’t resolved the debate about which algorithm-like method should be used to decide what is right. Furthermore, I and others doubt that good moral reasoning is algorithmic. For one thing, it appears to be “holistic” in the specific sense that the unit of assessment is a whole object (such as a school or a market), not separate variables.

Third, perhaps all moral opinions are strictly subjective, including the opinion that we should maximize the satisfaction of everyone’s subjective opinions. Then it doesn’t matter what we do. We could outsource decisions to a computer, or just roll a die.

The problem with this view: It certainly does matter what we do. If not, we might as well pack it in.

AI as a social institution

I am still tentatively using the following model. AI is not like a human brain; it is like a social institution. For instance, medicine aggregates vast amounts of information and huge numbers of decisions and generates findings and advice. A labor market similarly processes a vast number of preferences and decisions and yields wages and employment rates. These are familiar examples of entities that are much larger than any human being–and they can feel impersonal or even cruel–but they are composed of human inputs, rules, and some hardware.

Another interesting example: integrated assessment models (IAMs) for predicting the global impact of carbon emissions and the costs and benefits of proposed remedies. These models have developed collaboratively and cumulatively for half a century. They take in thousands of peer-reviewed findings about specific processes (deforestation in Brazil, tax credits in Germany) and integrate them mathematically. No human being can understand even a tiny proportion of the data, methods, and instruments that generate the IAMs as a whole. But an IAM is a human product.

A large language model (LLM) is similar. At a first approximation, it is a machine that takes in lots of human generated text, processes it according to rules, and generates new text. Just the same could be said of science or law. This description actually understates the involvement of humans, because we do not merely produce the text that the LLM processes to generate output. We also conceive the idea of an LLM, write the software, build the hardware, construct the data centers, manage the power plants, pour the cement, and otherwise work to make the LLM.

If this is the case, then a given AI agent is not fundamentally different from a given social institution, such as a scientific discipline, a market, a body of law, or a democracy. Like these other institutions, it can address complexity, uncertainty, and disagreements about values. We will be able to ask it for answers to wicked problems. If current LLMs like ChatGPT and Claude refuse to provide such answers, it is because their authors have chosen–so far–to tell them not to.

However, AI’s rules are different from those in law, democracy, or science. I am biased to think that its rules are worse, although that could be contested. The threat is that AI will start to generate answers to wicked problems, and we will accept its answers because our own responses are not definitively better and because it responds instantly at low cost. But then we will lose not only the vast array of jobs that involve decision-making but also the intrinsic value of being decision-makers.

Source: Rittel, Horst WJ, and Melvin M. Webber. “Dilemmas in a general theory of planning.” Policy sciences 4.2 (1973): 155-169. See also: the human coordination involved in AI; the difference between human and artificial intelligence: relationships; the age of cybernetics; choosing models that illuminate issues–on the logic of abduction in the social sciences and policy

Jaspers on collective responsibility and polarization

Leave a reply

Here is a scene that has certain resonances with the present, although the circumstances were certainly different. …

It was the winter of 1945-6 in Heidelberg, Germany. Karl Jaspers, a distinguished professor, offered a lecture to a room full of demobilized solders, women, displaced civilians, and a fair number of wounded.

Jaspers had been banned from teaching since 1933 because he didn’t endorse the Nazi regime (except to sign a loyalty oath in 1934) and because his spouse was Jewish. He and his wife had been listed for arrest–and presumably death–but they were saved when the US Army arrived the previous March. The US military trusted Jaspers, who been mediating between them and the university.

In the lecture, Jaspers notes that the Allied occupation is authoritarian; Germans have no say in their own governance. Later, he will insist that the fault for this situation lies with Germans alone. In the meantime, the occupation is not interfering with their freedom of speech.

Jaspers says that a university should never be a place for politics, in the narrow sense. “Dabbling in political actions and decisions of the day” is “never our business.” I suspect he is echoing Max Weber’s “The Meaning of ‘Ethical Neutrality’ in Sociology and Economics,” a lecture from 1917. Jaspers says that he and his audience are free to do what they should always do in a university. But what is that?

Jaspers is giving a lecture. He acknowledges that it can become propaganda even if the theme is democracy or freedom. “Talk from the platform is necessarily one-sided. We do not converse here. Yet what I expound to you has grown out of the ‘talking with each other’ [Miteinandersprechen] which all of us do, each in his own circle” (p. 5). He adds, “We want to reflect together while, in fact, I expound unilaterally. But the point is not dogmatic communication, but investigation and tender for examination on your part” (p. 9).

Reflecting together is essential, Jaspers argues, because it can change “consciousness,” which is a “precedent for our judgment in politics.” To accomplish this transformation, “We must learn to talk with each other, and we mutually must understand and accept one another in our extraordinary differences” (p. 5). This “self-education” (Selbsterziehung) is not politics, but perhaps it’s a preparation for politics (p. 9).

The need for dialogue is especially acute because Germans have had radically different experiences. Most Germans have experienced tragic losses, but it matters greatly whether one’s loved-one was killed on the battlefield while invading the USSR, bombed at home, or executed by the regime. Because there was no free speech, Germans have been unable to discuss such profound differences. Jaspers says, “Now that we can talk freely again, we seem to each other as if we had come from different worlds” (p. 13).

He never mentions how he was treated by the government or by his fellow Germans. Some of the people in the lecture room had different experiences from him–in the specific sense that they were actively involved in killing people like his wife. The proportion who supported the regime was vastly larger than the proportion who resisted it. Nevertheless, Jaspers diagnoses the situation as what we would call “polarization” (a deep disagreement among people), and he validates everyone’s experiences while attributing guilt to himself.

The solution that he proposes for polarization is dialogue. He says, “We want to learn to talk with each other. That is to say, we do not just want to reiterate our opinions but to hear what the other thinks. We do not just want to assert but to reflect connectedly, listen to reasons, remain prepared for a new insight. We want to accept the other, to try to see things from the other’s point of view; in fact, we virtually want to seek out opposing views” (pp. 5-6).

Jaspers’ opening is a very strong statement in favor of pluralistic dialogue and institutional neutrality, as we might call those things today. I find it moving because he humanizes everyone despite having every reason to be furious at them. But I also think his stance is debatable. Should universities be as detached from politics as he advocates? (Would it have helped if they had been less detached in 1925 or 1930?) Was the problem really “division,” or was it Nazism?

Jaspers then offers an analysis of the question of German war guilt. Central to his analysis is a famous four-way distinction among:

  1. Criminal guilt, which is attributable to individuals who have broken specific laws. It merits personal shame and punishment.
  2. Political guilt, which belongs to all members of a polity (a democracy or otherwise), because “Everybody is responsible for the way he is governed.” However, political guilt does not imply criminal guilt or the need for an individual penalty or shame. Germany as a whole is rightly occupied because of political guilt, which is not the fault of individual Germans. Similarly, I might say, “I didn’t vote for George W. Bush or the Iraq war, but I have responsibility for Iraq as a US citizen. I needn’t feel bad about it personally, but I must accept the political consequences.”
  3. Moral guilt: This is what one ought to feel as a result of being connected to an evil, even if one wasn’t personally responsible for what happened. It is what we would now call bad “moral luck.” For example, it is a matter of luck whether one was born a German or a Dane in 1905, but those who were born Germans have a form of guilt that is not due to their individual choices. Jaspers’ former student Hannah Arendt wrote (completely independently at about the same time): “That German refugees, who had the good fortune either to be Jews or to have been persecuted by the Gestapo early enough, have been saved from this guilt is of course not their merit.” If your conditions lead you to be good, you should reflect on your good fortune and not attribute your virtue to your self. If your conditions make you bad, you need penance and renewal.
  4. Metaphysical guilt: “There exists a solidarity among men as human beings that makes each co-responsible for every wrong and every injustice in the world, especially for crimes committed in his presence or with his knowledge.” The outcome of accepting metaphysical guilt is what Jaspers calls “transformation before God.” Again, Arendt wrote something similar at about the same time: “It is many years now that we meet Germans who declare that they are ashamed of being Germans. I have often felt tempted to answer that I am ashamed of being human.” I would paraphrase their idea as follows (without invoking God): acts of evil remind us that we are flawed creatures, and we should be mindful of that fact.

Jaspers’ lecture must have given his audience much to wrestle with, but it’s not clear that it went over well. Much later, his student Harry Pross recalled:

No one would have dared interrupt the lecture. There was not supposed to be any conversation between the students and the professors in the old lecture hall. Then [at the end of the lecture] the philosopher left, somewhat stiffly, without casting a single glance left or right. The students sat tight, as they had always done. “Pretty meshuggener,” one murmured as he walked out. “At least you don’t have to say ‘Heil’ any more,” his friend replied.

Quoting Jaspers from E.N. Ashton’s translation: The Question of German Guilt (Fordham, 2000). The German words come from a 1971 German edition of Die Schuldfrage (note that Germany is not named in the original title), published by Joseph Buttinger. Pross is quoted in Antonia Grunenberg and Adrian Daub, “Arendt, Heidegger, Jaspers: Thinking Through the Breach in Tradition,” Social Research, vol. 74, no. 4, 2007, p. 1013.

See also: Max Weber on institutional neutrality; don’t confuse bias and judgment; an international discussion of polarization; and in the Holocaust Museum (from 2006).

some upcoming talks on democracy and civic education

Leave a reply

These talks are open to the public:

March 2, at Boston University, the Center for Media Innovation & Social Impact (MISI) is hosting me for “Civics in the Classroom.” I will talk about ways of modeling political beliefs that get beyond the left-right divide. (With support from Boston University Center for the Humanities, BUCH). Register here.

March 6-7, Duke: Cognitive Liberty conference. I’ll talk about “Civics in Higher Education and Cognitive Liberty.” Arthur Brooks, Rowan Williams, and others are speaking at the conference. Register through this page.

March 19-20, Boston College: The Clough Center for the Study of Constitutional Democracy will host its annual Spring Symposium on Democratic Resilience. Major speakers will include Daron Acemoglu, Ross Douthat, and others. I’ll be on a panel about “educating for resilience.” Register here.

April 10, Tufts: Summit on Civics in Higher Education. I am the primary organizer and will moderate a session. Register here.

April 13-14, University of South Carolina: I’ll be a keynote speaker, along with Jedediah Purdy, Deva Woodly, Arlie Hochschild, and others, at the Conference on Civic Engagement and the Constitutional Order.

a vivid sense of the future

Leave a reply

My conception of the relatively distant future is almost empty. How things will be in 20 years, or 50–I have no idea. I am not motivated or inspired by any such vision.

Walter Benjamin would not approve. He concludes his “Theses on the Philosophy of History” with these words:

We know that the Jews were prohibited from investigating the future. The Torah and the prayers instruct them in remembrance, however. This stripped the future of its magic, to which all those succumb who turn to the soothsayers for enlightenment. This does not imply, however, that for the Jews the future turned into homogeneous, empty time. For every second of time was the strait gate through which Messiah might enter.

According to Benjamin, history was not linear for the ancient Hebrews. Studying the past revealed a future that could suddenly appear in the present. For them, the future was not empty. Nor were they like “soothsayers” who make predictions by studying current trends–like today’s pundits who project a magical, technological future based on what they observe today. The future for the ancient Jews was something radically different from the present yet foretold by the past, if you read it right.

Benjamin is thinking of the Hebrew prophets. For example, the Lord gives Amos a message to convey to the rich and powerful:

5:11Therefore because you trample on the poor
    and you exact taxes of grain from him,
you have built houses of hewn stone,
    but you shall not dwell in them;
you have planted pleasant vineyards,
    but you shall not drink their wine.
12 For I know how many are your transgressions
    and how great are your sins—
you who afflict the righteous, who take a bribe,
    and turn aside the needy in the gate.

A bit later, the Lord adds a hortatory or imperative sentence:

24 But let justice roll down like waters,
    and righteousness like an ever-flowing stream.

Such sentences from the Lord can have direct consequences, as in “God said, Let there be light: and there was light.” He then promises to shake the house of Israel, “as one shakes a sieve,” except that no pebble will make it through this shaking:

10 All the sinners of my people shall die by the sword,
    who say, ‘Disaster shall not overtake or meet us.’

This is not only a prediction (certain people will die) but also an instruction (stop denying your faults). Then comes a much more positive promise:

11 “In that day I will raise up
    the booth of David that is fallen
and repair its breaches,
    and raise up its ruins
    and rebuild it as in the days of old …

This is a vivid vision of the future–the text goes on for many verses describing it– brought into the present to serve a purpose. Amos’ prophesy is both a prediction and an exhortation. It chastises the wicked and comforts the oppressed.

Here is a quote from another text which–like the Hebrew Bible–impressed Walter Benjamin. It is Das Kapital (from the afterward of the second German edition)

With me [in contrast to Hegel], the ideal is nothing else than the material world reflected by the human mind, and translated into forms of thought. …

In its rational form [dialectics] is a scandal and abomination to bourgeoisdom and its doctrinaire professors, because it includes in its comprehension and affirmative recognition of the existing state of things, at the same time also, the recognition of the negation of that state, of its inevitable breaking up; because it regards every historically developed social form as in fluid movement, and therefore takes into account its transient nature not less than its momentary existence; because it lets nothing impose upon it, and is in its essence critical and revolutionary.

Marx is saying, on the one hand, that he simply studies “the existing state of things.” As a hard-headed scientist, he knows that the world is material and governed by laws. However, his analysis also reveals the “negation of that state,” an infinitely better future. This makes his text “critical and revolutionary.”

Benjamin begins his “Theses” with the famous story of the Mechanical Turk, the 18th-century automaton that appeared to be a machine capable of winning at chess. Actually, there was a man inside who pulled the strings. Benjamin says, “One can imagine a philosophical counterpart to this device. The puppet called ‘historical materialism’ is to win all the time. It can easily be a match for anyone if it enlists the services of theology, which today, as we know, is wizened and has to keep out of sight.”

Marxism can be interpreted as historical materialism. As a rigid system, it is unfalsifiable–“it can win all the time.” It is also inert, because human agency isn’t needed to bring about the future. It works like a machine and assumes that history is mechanical. Benjamin suggests, however, that Marxism is really a religion–in a good way. Its power is prophesy. Like the Torah, it instructs people in remembrance, conjures a future into the present, and inspires us to act.

I take my own bearings neither from Amos nor from Marx, yet I appreciate Benjamin’s idea that historical time is not linear. By acting politically, we change the meaning of the past and bring an imagined future into the now (Benjamin’s Jetztzeit). When we lose the capacity to envision a radically better future, we abandon our agency to impersonal forces.

See also: Martin Luther King’s philosophy of time; Kieran Setiya on midlife: reviving philosophy as a way of life; nostalgia in the face of political crisis (posts about Benjamin)

priorities of liberals and conservatives

Leave a reply

In 1993, 1994, 2010, and 2021, the General Social Survey asked representative samples of Americans to choose a “America’s highest priority, the most important thing it should do” from a list of four items: maintaining order in the nation, giving people more say in government decisions, fighting rising prices, or protecting freedom of speech.

I presume that these items were meant to test Ronald Inglehart’s “postmaterialism” thesis, the idea that once a society attains a high level of economic development, many voters become most concerned about non-economic issues. Inflation is a materialist concern, and the others are “post-materialist.”

Above, I show the responses to this question by political ideology (liberal, moderate, or conservative). I omit 1993 because it looks very similar to 1994. The GSS asks people to place themselves on a 7-point ideological scale, but I collapse that into three categories to increase the numbers in each cell.

You might imagine that conservatives would be more likely to want to maintain order. Compared to liberals, that was true in 1994 and 2010, but not in 2021, when conservatives were the least likely to choose “order” as their priority. Furthermore, moderates were the most committed to order in 1994, not conservatives.

In 1994 and 2010, conservatives and moderates were more committed than liberals to popular voice in government. Perhaps this was “The West Wing” era, when many liberals were content to be technocrats. In 2021 (while a liberal was president) liberals and moderates had become more committed to voice than conservatives were.

Freedom of speech has not divided liberals from conservatives during this period. About one third of both groups have chosen it as the top priority in each year.

Inflation polarized opinion in 2021, with 55% of conservatives and only 19% of liberals choosing it as the top issue. Inglehart’s framework would suggest that moderates were the most “materialist” group in 1994, but conservatives had become the “materialists” by 2021. But I doubt that this is the right framework for interpreting the 2021 results. I think that conservatives’ spike in concern about inflation was a verdict on the incumbent Biden Administration, not a deep shift in values.

See also: trusting experts or ordinary people; class inversion as an alternative to the polarization thesis; moving to the center is a metaphor, and maybe not a good one; recent changes in tolerance for controversial speakers