Seeking Cause X

Note: Found this in drafts from July 2022, publishing it because I quite like it.

---

"Whether you can observe a thing or not depends on the theory which you use. It is the theory which decides what can be observed.“ Einstein"

Intro

When we are evaluating anything, we need some means by which to evaluate the thing. So long as we wish for some Thing to become some different Thing, we will incorporate the notion of Values into it. That is, we value Thing more to be in State Y rather than State X, and thus we expend energy we deem worthwhile to influence it so.

A lion values a gazelle to be in an "eat-able" state, while the gazelle values its "continuing-to-live" state, and insofar as these two are incompatible in the specific context of an engagement between a hungry lion and an available gazelle, the two will respectively expend energy towards its valued state until a multitude of factors in the world collapses the query into the valued state of the lion or the gazelle.

When we wish to do good, our definitions of good (state Y is preferable over state X), consciously or not, communicable or not, will bleed through our actions that seek to influence the world. Our ability to communicate with each other, beyond the local value systems we each have, will be the basis for creating the parallax on which new, previously imperceivable, ways of doing good emerge.

An idea I have great affinity for is the notion of "Cause X", an idea representing epistemic humility – keeping in mind that there may be unknown ways of doing good, or Cause Areas, in the world missed by our current understanding. If we hope to have a chance to discover any Cause Xs or even just any information about the possibility of there being any Cause Xs, it seems we must get better at navigating different systems of value and knowing.

A Scenario on Navigating Differences in Education:

Suppose my father works at a United Nations education initiative, aiming to keep children in school. This initiative is backed by enormous financial investments to incentivize young children, particularly girls, to go to school in developing countries. It is supported by ample research pointing to this type of intervention as one of the most reliable lead measures for a nation's growth towards a "developed country". My father knows well that his work of helping children stay in school will lead to greater gender equality, income equality, and greater scientific and economic output, all foundational values of modern democracies.

Meanwhile, I work at the Thiel Fellowship, with the opposite aim of keeping youths out of school. I operate with the model, increasingly recognized, that modern education is one of the more ghastly experiments of humankind. Born out of a dark history of religious dogmatism and imperial servitude, and refined by industrial-style management, our education raises each generation with basic wounds around self-worth, an inability to deal with the unknown, and a broken sense of interdependence with those around them. I work with the knowledge that every person liberated from modern schooling has the potential to be re-empowered to become a creative citizen in society, able to tackle modernity's most complex challenges.

One can probably imagine what my family life will look like here: dinner arguments over who is correct and hurt feelings around why my father/son seems to be personally disrupting what is "obviously good".

What one can perhaps imagine, but may feel is closer to idealistic fantasy, is what I believe must be strived for. In my not-so-distant-fantasy, my father and I could start any conversation with an implicit or explicit agreement and mutual appreciation of our shared desire to support future generations. How incredibly fortunate! We could subsequently be curious about how and why our model of doing good (in education) appears opposite, exploring the differences in our respective lives growing up and fostering the intimacy between a Father and a Son. We might arrive at the remarkable observation that there's a notable socio-economic divide in that: those of higher social mobility tend to explore alternative schooling options while the global poor are encouraged to stay in school, and that this may be indicative of an uneven distribution of innovation and knowledge around the impact of education. We could exclaim together how this disconnect leaves the global poor behind with more of the same trauma, and doesn't support the global rich in recognizing the abandonment/colonial pattern, which undermines their own effectiveness in innovation. From our parallax, my father and I may discover that I have always wanted to scale our alternative education experiments beyond eccentric young founder-types, while my father has just the right local partners to start pilot programs around the world. We are now learning together, with more effectiveness and intimacy as a team and as a family, and we are seeing more about the nature of the complex problem, in this case education, than either of us could before.

Briefly on Objectivity as an Incomplete Lens for Doing Good Together

One could consider that, in examples like the above, were we to evaluate from the almighty lens of Objectivity, we could enlighten more systematically the "clearly superior/most effective" way to do good. Examining the world “objectively” has revealed many hidden cause areas that were once neglected. It does appear invaluable when we discover that it takes $30,000 to train a guide dog, while it takes $3 to deworm someone who might develop blindness, and that the latter is meaningfully underfunded than the former. This unveils a meaningful way of doing good through correcting for the funding gap.

But this too is partial, as illustrated by the school example above. Robust research suggests both directions from "me" and "my father" as valid interventions for “better education”, and the underpinning cruxes are undecidable purely from the perspective of which is “objectively better”. This is because it pertains to the impossibility of perfectly predicting the future and its associated counterfactuals, as well as when comparing judgements around value, Objectivity is no longer the relevant realm of consideration, Subjectivity is. [footnote: The best we often seem to achieve in these scenarios is an “agree to disagree” situation that leaves both sides with slight-to-siginificant bitter tones.]

Assuming that we are within the shared context of a common definition of good, though, we can then imagine creating a market around that definition, and smooth out inefficiencies, like in the guide dog and deworming example. Markets are extraordinarily good at this! This is obviously an important mechanism in the entire ecology of "cultivating a better world". However, a market is only as good as creating equilibria within the definition embedded in said market. If the market's "Good" is about the maximization of profit for participating agents over time, then desirable and undesirable behaviors will occur. If a market's "Good" is defined as discovering the best arbitrage opportunities within globally legible altruistic projects, under some model of utility as a proxy for "Good", then we have something approximating Effective Altruism, which also has desirable and undesirable behaviors. [*footnote: Modifying this definition of utility without future discounting seems to yield something approximating EA Longtermism, and with the discounting seems to yield something approximating classic EA. Granted, this is a relatively strawman version of EA and its strands, but what it also does suggest is that there are likely many more strands of EA yet to find maturity.]

This is not to say our decisions can not be informed by an "Objective" Lens. In the education example above, we can likely agree that both interventions from my father and me are "good" interventions, and we can better understand how and why the respective intervention is good by examining the data/research that supports each side. What the "Objective" lens does poorly on though, is that it does not support my father and I collaborating on a synthesis that emerges fusion solutions that are likely to be "better" than both interventions done independently of each other. To achieve this, we must get past the objective lens' tendency to evaluate something to be “better” than the other, rather that one is likely informative of the other. [footnote, Which, if anything, is more "objective" in the end. my personal gripes with objectivity is a lot more in how we seem to relate to the idea of objectivity than objectivity itself.]

Ultimately, doing good together effectively seems to require crossing the bridge of being able to navigate different understandings of value. This means moving away from "destroying each other with facts and logic" to creating parallax and new possibilities that multiple parties would want to collaborate and contribute to.

"Seeking Cause X"

In the late 1800s, there was a popular sentiment among physicists and other scientists claiming that we are near the end of Science, where we will soon discover everything there is to know about the cosmos. With the advent of relativistic and quantum physics – models that opened up a greater realm of knowing while accounting for the old Newtonian model – we discovered that we are nowhere near the “end of Science”. It seems instead it was more apt to describe the 1800s sentiment as: we are hitting significantly diminishing returns on “the theories by which we are understanding the universe/doing science”. As we broke past the limits of our own theories, we instead found more mysteries than we could have imagined.

Metaphorically, if a given system of values (and its associated ways of knowing) is like a tree that can bear fruits (i.e. good things that we want), then as we become really familiar with the particular system of knowing, any low-hanging fruits will eventually be plucked. A methodology that isn’t bound to any particular system of value, then, would be like searching for the forest to find new fruitful trees, without being bound by any fruiting tree. [Footnote: For example, upon the invention of the combustion engine, the value of hydrocarbons dramatically shifted, creating an explosion of advancements in the movement of things, until the diminishing returns of the present, where we are now looking for new models that can replace and exceed the dominant technology.]

Similarly, any given theory around doing good would have its own associated low-hanging fruits and would over time hit diminishing returns. In order to reliably find new worthwhile ways of doing good, we have to be flexible enough to fluidly adopt multiple theories of good. In the metaphor of the fruiting trees in a forest, it means we would need some means to actually compare apples and oranges.

So, if we were to assume any Cause X, defined as a cause area, or a way of doing good, that is previously unrecognizable, then any meaningful method of discovery must account for what’s beyond our local epistemology. Thus, such a methodology has to be transcendent of any particular system of value – a meta-systematic method of evaluation.

Any development in research or training on such a meta-methodology that can evaluate multiple kinds of systems of value could potentially yield an abundance of Cause Xs. This creates a plethora of new research conundrums, taking us from “what Cause Xs might be out there?” to “which of all these Cause Xs should I or we pursue/prioritize?”. While it creates more mysteries than answers, what it can provide is a platform that accepts multiple ways of doing good in radically different yet complementary systems of knowing/value. It permits collaboration amongst altruists that departs from the argumentative “what kind of good is ‘best’” and towards complementary discoveries of new ways of doing good that more effectively achieve multiple parties’ aims, despite their apparent initial contradictions.

The dilemma of navigating differences of value is pervasive on all levels of human organizing. It’s when parents discuss which is more important, acquiring a math tutor for their daughter or letting her continue drawing. It’s in an NGO deciding whether direct cash transfers would be better or radio broadcasted education programs. It’s in the political left and the political right and those that aren’t covered by that single axis. It’s in the indigenous elder saying we need to learn to slow down as well as the environmentalist saying we must act now. A method for navigating these differences can unlock the latent complementary aspects of different interventions, forming more effective synthesis and a more collaborative culture amongst altruists, a virtuous cycle that begets more effectiveness.

If we assume our theories of doing good are always partial, then we can start from a place that recognizes differences as valuable perspectives that might yield information on what’s being missed. This transforms the nature of conflict. We can disentangle feedback from praise and blame and recognize feedback for what it is – information. From this information, we can see previously invisible aspects of what we care about and thereby be empowered to affect the world in previously inconceivable ways – discovering Cause X.

Every naysayer becomes our most valued collaborator. Every contradiction becomes an opportunity for wonder.

Summary

Whether there exists an Ultimate Good is beyond this post, but conveniently, we don't need to go there. So long as we are subjective beings using proxies or representations to navigate the universe, then any single dominant representation will create distortions.

Our theories are our models, or maps, from which we examine the territory. Whether it's the biological range of what we can see and hear, or our model of physics, the map is what give us clues in the first place as to what landmarks we can even recognize. If we are not cognizant of the models we use, we are liable to miss the territory. If the territory is in fact infinitely more complex than our maps, then we must treat maps that are different from ours as, at minimum, deeply informative for our goals of influencing the world to be "better".

Starting from the epistemic humility of recognizing the possible existence of Cause X, we can invoke our curiosity and wonder to seek out the perspectives yet unavailable to us. Fortunately, those perspectives are readily available around us, in other subjective entities from other people to the nation state to the forests. We have to learn the meta-methodology that permits the listening that can integrate the differences in perspective to form new understandings.

More illustratively, if I'm looking for herbs to heal a wound, rather than looking at google maps and lamenting that herbs don't seem to exist, one might want to consult the weird lady telling us to look at the bush right outside our door.