The Garbage Can

Do you ever wonder whether your brain is like a garbage can? I’m going to argue that mine is.

Let me backup a little. The question is obviously a little silly. It’s inspired by a classic 1972 paper, “A Garbage Can Model of Organizational Change.” That paper asks a really interesting question: How does an organization make decisions when its participants have ambiguous and shifting goals, are organized according to rules that they don’t fully understand, and are variable? These three criteria — “problematic preferences,” “unclear technology,” and “fluid participation” — characterize what Cohen, March, and Olsen call “organized anarchies,” which they suggest that many governmental institutions can be approximated as.

One way to approach their argument is to compare it with what we might assume a rational actor does when making choices. Suppose you’re the President and you hear that a deadly virus has begun spreading throughout the world. You might first assess the problem — how dangerous does it seem, when is it likely to affect the United States and its interests, and so on. Then, you might turn to potential solutions: Do I need to ask the governors to shut down their economies? Should I impose a travel ban? Should I ask the populace to wear masks? You’d weigh the different solutions — what are their costs and benefits? — and then decide which action you want to take. In theory, a good organization would do the same. If all members can agree on the goal, if everyone had the same knowledge of the potential solutions, and if the organization had stable membership, you could theoretically follow a very similar model of decision making, with the caveat that you’d need certain “rules of the game” to guide that decision process.

The garbage can model says that there’s no linearity to decision making under conditions of organized anarchy. Problems and solutions are continually being tossed by participants (who act as advocates and who have certain kinds of knowledge) into the metaphorical garbage can, which represents a choice opportunity.

Why think of problems, solutions, and participants as interdependent? First, problem definition isn’t unambiguous — think about the debates, especially early on in the pandemic, about how dangerous the coronavirus was (Is it just the flu? Is it ebola?). The way you conceive the problem (rightly) will be linked to what you think is the right solution (more dangerous -> stricter solution, and vice versa). But the direction isn’t always so rational. People have stakes in particular solutions. The restaurateur probably wants to down-play the dangers of the virus just as much as the Zoom executive wants to up-play them, because the former might think of themselves as benefiting from an open economy and the latter from a shut down one. Any given solution, Cohen, March, and Olsen write, is really just “an answer actively looking for a question.” So problems are linked to solutions. Who gets to diagnose problems and design solutions — who is the metaphorical salesperson, if a solution is really just, as they put it, “somebody’s product”? It’s the participant. And in conditions of organized anarchy, the participants are fairly random, and they churn in and out — when one person leaves, another joins. Each holds their own specialty, interests, and ways of thinking that will make them more or less receptive to certain problem-solution pairs. So with the garbage can model, problems, solutions, and participants are all interlocked rather than independent factors in decision making.

There’s two main implications of this line of thinking. First, decision making is really chaotic — it’s not nearly as linear as we’d like it to be, at least in conditions of organized anarchy. Second, chance and timing really matter — a given solution needs the right audience (the right participant in the organization) and the right environment (i.e., the right problem must be on the agenda) in order to become reality. Ultimately, that’s just a matter of having the right people (participants) throwing in the right combination of solutions and problems into the garbage can (the decision opportunity) before some part of the mass gets whisked away as garbage is collected and removed.

Let me try to apply the analogy to how I make decisions, then. Remember that to be a fair analogy to an organized anarchy, there must be three analogous conditions: unclear preferences, uncertain technology, and fluid participants. If the analogy is about how I “make up my mind” about something, then the “unclear preferences” apply in a fairly straightforward way: I hold a multitude of values, and it’s never really clear which is most important. Importantly, in the garbage can model, it’s not just that preferences are unclear — it’s that there are violations of rationality conditions.

Take the “axiom of irrelevant alternatives.” this basic criterion for rationality says that if you have two options (A and B), and you prefer A to B, then adding a third option (C) shouldn’t make you rank B over A. That doesn’t hold if you have variable values, though. If I’m offered a hot dog or a cheese pizza for lunch, I might want the hot dog if I’m thinking only about what will satisfy my cravings most. But if the menu suddenly includes a veggie burger and primes me to think about how I ought to avoid eating meat, then my preferences might suddenly flip — maybe I’d prefer the veggie burger over the cheese pizza over the hot dog.

Remember the garbage can idea that solutions are linked to problems. In the same way, at least for me, options are linked to values. Being presented with certain opportunities will make me reformulate how I evaluate the situation if only because what values I am trying to maximize will depend on the moment.

Next, “uncertain technology.” The analogy here, I think, are my tools for making decisions — mental shortcuts (heuristics) or algorithms (something more like a philosophical method). Both are uncertain, but perhaps in different ways. With heuristics, I think they’re uncertain because I don’t have a strong understanding of how they operate. Studies about implicit bias, for example, suggest that we have all sorts of perceptions about people or situations that we can’t consciously control but that will shape how we relate to the world around us. (Think, too, of the “sixth sense” that you sometimes have.) With algorithms, on the other hand, I think the uncertainty concerns their application — how do you apply a certain set of principles to a fact pattern, and are you comfortable with those principles in different situations?

The “technology,” in this analogy, is also linked to the last requirement of an organized anarchy — “fluid participants.” In my decision making process, I think the analogy to participants are my imaginations of advocates of certain ways of thinking — whether it’s Brandeis, or Gandhi, or David Runciman, or my professors, or my friends. Part of what I value in reading about different people’s ideas is that it gives you a (perhaps) coherent framework for making judgments. It’s obviously not always clear whose way of thinking is right in a given circumstance. But more importantly for the analogy, who’s at the front of my mind often changes. And when making decisions, who’s at the front of my mind has a big effect (potentially) on what I end up deciding because their ease of recall shapes the probability that I invoke their logic. I’m not sure I ever invoked Lincoln until a few weeks ago, after I read a book about him. But now that he’s on the top of my mind, his way of thinking certainly influences how I think about politics today.

If the analogy holds that my brain is like organized anarchy, then it follows that my decisions can be characterized by the garbage can model. Let me end with a few reasons why I like this analogy.

First, the model is clearly, intentionally iconoclastic. It’s meant to provoke. (After all, the authors are likening the decision making process of institutions like Congress and Pomona to a garbage can. You have to admire the irreverence.) Thinking about my mind as a garbage can is a kind of grounding — it helps me not take myself too seriously. And here’s my permission for you all to hold me accountable — do feel free, if you wish, to call me a garbage can :)

Second, thinking of solutions as answers looking for problems, and of ideas as fluid participants, helps remind me that what I choose to read and think about has downstream ramifications on what I will end up valuing and how I will end up thinking. If I were perfectly rational, had unambiguous preferences, and could hold unlimited information, then it wouldn’t really matter if I read Mein Kampf versus Hind Swaraj. For neither would impact how I view the world; preferences precede the information, so I would only read the books to reference them as confirming or rejecting my way of thinking. But what I read, and what I think about, definitely shapes how I think moving forward. The garbage can analogy helps remind me to be conscientious about that fact. To put it in a less flowery way, the garbage can analogy tells me to get off Twitter (truly an echo chamber) and go back to reading books on the list I wrote down for New Years (and the many recommendations I’ve received from friends).

Third, I like that the garbage can analogy emphasizes how chaotic and dynamic decision making is. Writing stuff down has helped me recognize how scattered my thinking tends to be — and how often I want to contradict myself! (An example of this: I literally contradicted my last post just a few days ago when talking to friends — not about the overarching argument, but about how to think about a small aspect of it.) Thinking of myself as a mobile garbage can, with ideas and beliefs cycling in and out of my head in a chaotic, somewhat random way, helps me understand why I might expect my judgment to be inconsistent at times, especially by drawing my attention to the role of imagined advocates and what primes me to want to think about them.

* * * * *

Rose: A few this week!

Professor Hollis-Brusky and I submitted our paper on the Office of Legal Counsel for review. I’m getting more and more excited about this project (we’re in the early stages of a second paper), and it felt good to get that one off our Dropbox and into the review void.

I got to meet (through Zoom) a random Marshall in my year, and I learned so much about agriculture. It was fascinating. (I’m not sure she enjoyed it all that much — I suspect I asked her too many questions, in a way that was off putting. But it was just so interesting.)

I went for a long walk with another Marshall in London. It was a cold but beautiful day, and he also studies political science, so it was fun to chat and get his perspective (he studied comparative politics, so his thinking is a bit broader than mine tends to be). Here are some pictures from that walk.

Banksy at the Barbican

"Probable Site" of Methodism

Roman Wall!

The Bard

City Hall, the Shard, Sunset

The City and Tower

Tower Bridge

I went cycling with my cycling buddy, and we went to another one of the Magnificent Seven cemeteries — this one was Tower Hamlets. I’ve also formed a support bubble with her and her partner, so we’ve cooked together a few times, which was very fun. I’ve tried making paneer from scratch three times, including for when they last came over, and each time was a complete failure. She tried it once, and it was perfect. I think it was the milk — I bought cheap milk, whereas she shelled out (in her words: “I got the best I could find, given the stakes”). Oh well. Here are pictures from Tower Hamlets.

Bud: I’ll be starting research for my competition law professor soon — I’m hoping to finalize the paperwork on Monday, so that’s exciting!

Thorn: I have been writing an essay for my public management class for the last few days, and it has not been very fun. Alas, it will be done soon.

Gratitude: I’ve been doing Zoom workouts with two of my friends (they’re both doing Masters at Cambridge), and that’s been really nice — and motivating. I’m finding it harder and harder to get myself to move around absent some jolt, like a walk with a friend or a scheduled workout.

Future topics:

* Mensa: a friend sent me this article about a comedian who joined Mensa (a self-described high IQ society), and I’ve been listening to her podcast about her experiences. I think it’s valuable to consider how, and why, a society focused on bringing together certified “smart” people might lead to all sorts of terribleness (lots of far-right, white supremacist members; intense cyberbullying, including death threats) — and whether many of our institutions of higher education might be similar.

* Nationalism and partisanship: I finished Benedict Anderson’s Imagined Communities somewhat recently, and I’ve been wondering about a possible analogy to partisanship. I’m not sure how compelling the argument would be, but it might be worth trying to make the argument to test it out.