Epistemic Status: Proposing new terminology to clarify discourse surrounding potential new equilibria.

TL;DR: We should use the term summit-seeking to refer to the evaluation of equilibria different from the current one, and the term mountaineering to refer to the discussion of the practical considerations of moving from one equilibrium to another. Posts and conversations can avoid confusion by specifying which kind of discussion they are or wish to have.


I. An Object-Level Example of The Problem

I greatly enjoyed the series of guest posts on AstralCodexTen by Lars Doucet regarding the empirical claims of Georgism (1st, 2nd, 3rd).

The comment section of those posts, on the other hand, confused me.  This post is about trying to identify that confusion, and providing a shorthand terminology for resolving it in future discussions so as to help us communicate better.

Briefly, so we may use concrete examples, Georgism posits that land, that which is already there without human labor or capital (think a literal plot of dirt), is importantly distinct from both, in that the value of a given piece of land - a location - is mostly due to what's around it, not what's on it.  A house in a really good school district is worth more than that same house in a really bad school district; an empty plot of dirt in the middle of New York is worth far more than an empty plot of dirt in Siberia.  Georgism then makes claims about how land ought to be taxed (as distinct from how property taxes currently work), for various moral and financial reasons.

For a complete (and much better) explanation, I encourage you to read Lars' original book review.

Lars' posts attracted a large and healthy debate, with the highlights I generally expect from Scott Alexander's commentariat.  I also saw, however, a large number of comments with the nature of 

I just bought a house, and if we switched to a Georgist Land Value Tax I'd be underwater on my mortgage.

Given that the nature of the post was about whether or not the claims of Georgism hold up empirically (or are at least plausible), this strikes me as a category error.

Replies to those comments also pointed this out, but a sizable percentage of the discourse remained concerned with the consequences of a switch from our current tax law to a more Georgist tax law.  I've no wish to single anyone out, and I think the discourse remained reasonably productive, but what it did not do was remain on topic.

The topic of Lars' posts were about whether or not an economic theory - Georgism, in this case - held water, after reviewing the empirical evidence for and against to the best of his ability.

The topic of much of the comment section revolved around whether a Georgist tax system should be phased in over twenty years to avoid disrupting the housing market too much, or whether tax credits should be given to existing homeowners to compensate for the changes a Georgist tax regime would impose.

These two discussions are both valuable.  They're both important to have.

But they're not the same, and I think it made the discourse more confusing than it needed to be.


II. Discussing Equilibria

I recently reread Yudkowsky's Inadequate Equilibria.  Among other things, I think the book provided a valuable way of talking about the state and status of systems and institutions to the general discourse.  After reading it, I now have a convenient label for a-system-that-is-sub-optimal-but-stuck-in-place-because-no-individual-can-benefit-from-moving-to-a-different-system.

I can just say that the system is in an inadequate equilibrium, and be done with it.

What I'd like to do here is follow on Yudkowsky's work by adding to the zeitgeist terms for the type of discussion one is having, when one's discussion involves these inadequate equilibria.

The need for new terminology stems from observing the discourse mentioned above.  Let's see if I can lay out the case in general terms, and then bring it back to the above concrete example.

There seems to be confusion when new equilibria are discussed. A description of the new equilibrium is offered, and the response deals with the issues of getting from the current equilibria to the new one.

While discussion of how to get to a new equilibrium is important and valuable in practice, it isn't the same thing as discussing which equilibrium we should be aiming towards.

We need a term for comparing equilibria on various merits without muddling the discussion with the practical considerations involved in traveling there. Again, such considerations are invaluable, but they are separate and distinct from deciding which equilibria we ought to aim for.

Likewise, we need a term to describe seeking to move ourselves from one equilibrium to another without worrying about whether or not the new equilibrium is the one we ought to be moving to.

The goal is to enforce the norm that these are two different conversations, and that we can avoid a lot of confusion and disagreement by figuring out which one we're currently in.

A part of the intellectual background of Inadequate Equilibria, as I understand it, centers around the idea of different sorts of equilibria, the technical names for which are things like Nash Equilibrium and Pareto-optimal.  For our purposes, however, I'm going to use a two-dimensional graph and some arrows, because that makes sense to me.[1]

Imagine that the system that we're talking about - education, healthcare, taxation, whatever - is represented by a point on this graph:

A graph of successfulness versus current state, with peaks and summits, and a red dot on a peak (but not the highest peak).

We are considering changing our current state in order to reach a higher peak of successfulness, because more successfulness is better than less successfulness[2].  In practical terms this means we are looking at reforming healthcare or tax law or whatnot - a general change to the current system represented by the above graph.

Now, without using any economics jargon, we can see that we're currently at a peak of successfulness per our current state, but that in a different state over to the right on the x-axis, there's a higher peak, and the grass over there looks a lot greener than the dry, dusty weeds currently disintegrating under our feet.

In real life, of course, we don't get a handy graph promising a higher peak at a specific place, but we can argue that there ought to be a peak somewhere and theorize about how high it is, and we can do our best to back up our arguments and theories with evidence and experiments and generally just do some Scienceat it.

Coming back to our goal of distinguishing between two distinct conversations, we have: 

  1. Discussing the merits (or lack thereof) of a new equilibrium, and
  2. Discussing how to transition from the current equilibrium to the proposed new one.

In our picture, a. is represented by:

Now the peaks are colored with different color dots, and arrows stretch from the current (red) peak to the other peaks (lower, blue) and (higher, pink), comparing them.

And b. is represented by:

A path is drawn from the red (current) peak to the highest peak (pink), going through valleys in between.


III. Proposal of Terms: Summit-Seeking and Mountaineering

With these pictures in mind, I offer the metaphor of the difference between choosing which mountain to climb, and the practical considerations of climbing said mountain once chosen.

I hereby propose the term "summit-seeking" to describe discussing which equilibria (which "summit"), of those visible or theorized, is best to aim for.

Examples of summit-seeking include debates over whether a single-payer healthcare system would be better than our current system, or if college is worth attending.

I hereby propose the term "mountaineering" to describe discussing the practical considerations of reaching a different equilibrium from the current one.

Examples of mountaineering include debates over how to convince homeowners that building more housing could be good for them, or discussions of why nuclear power isn't the climate change panacea it could have been.

Coming back to the discussion on Lars' posts regarding Georgism:

The post was summit-seeking: Lars was evaluating whether or not Georgism is a viable equilibrium, and whether or not it would be superior to our current equilibrium.

The responses I was confused by were mountaineering: if we switch to a Georgist tax scheme (Land Value Tax), then suddenly I'm paying X amount more in taxes per year.  How is that fair?

While both discussions are valuable and absolutely worth having, they're different discussions. Summit-seeking discussions evaluate the worth and feasibility of distant equilibria. Mountaineering discussions revolve around how we could, in practice, get from the current equilibrium to a distant one.

Indeed, responses to the above response were often mountaineering themselves, indicating that Georgist policy could be phased in over twenty years, or could involve tax credits of one kind or another.

I don't want to rehash the Georgism debate here; that can be kept to its own thread. But I think that adoption of the terms "summit-seeking" and "mountaineering" could enhance the level of discourse surrounding alternative equilibria, and prevent confusion between groups of people arguing in good faith.  If anyone knows terms that already exist for this, or better terms, then I'm all ears.

  1. ^

    I also find the more technical terms from economics unnecessary so long as we're all looking at the same picture.

  2. ^

    Opinions may vary.


New Comment
2 comments, sorted by Click to highlight new comments since: Today at 6:25 PM

This seems a useful distinction to make. I think your terms also make sense, as this was indeed the kind-of-thing I expected them to distinguish between when reading the title.

I do want to note that discussions about summit-seeking might include mountaineering considerations. If the price to get to the new equilibrium is very high, this could lead one to decide not to aim for this equilibrium at all, even though the equilibrium itself (the summit) is good in other dimensions.

Maybe this is already what you're pointing at when you mention the "feasibility" of an equilibrium, but I think it's worth stating explicitly.

Thanks for the feedback!

I agree that the two conversations naturally want to coincide: any discussion of a distant equilibrium as compared to the current one seems likely to include, by default, an evaluation of the costs of the change, which is Mountaineering.  Likewise, wondering if a given equilibrium is a worthwhile goal (Summit-Seeking) seems to happen (at least to me) a lot in the middle of thinking about how to get there.

The two conversations are necessarily intertwined - but without the ability to distinguish clearly between them, I worry about people talking past one another, where one person thinks they're still Summit-Seeking while the other believes the conversation has moved to Mountaineering.

I guess I kind of think of it like dye in water - once properly dissolved, the two are indistinguishable, but it's still valuable to be able to talk about water-without-dye and dye-without-water.

New to LessWrong?