Ranking The Sacred

Consider four possible acts:

  1. Eating Twinkies
  2. Watching Gilligan’s Island
  3. Fighting cancer
  4. Working for racial justice

Now consider pairwise comparisons of value between these acts. You might say which you prefer, or which matters more, or is more important or admirable.

It seems to me that we don’t mind ranking #1 vs #2. We might think the exercise silly, but we’d still be comfortable expressing an opinion. It also seems to me that we don’t mind puffing up our chest and intoning very seriously that either of #3,4 are more noble and admirable than either of #1,2, and looking sadly down on those who might say otherwise. But if asked to rank #3 vs #4, we are much less comfortable. In this case we could be seen as saying something against an act many find important and admirable. That isn’t the sort of thing we like to be quoted on. We don’t like to speak against the sacred.

Because of this, we end up sharing less info about relative rankings among the acts we most admire. Which, alas, are the acts most valuable to rank. We learn what others think of the relative ranking of silly tv shows and minor foods, but not about our most important choices. Silly humans.

I’m fond of this classic question pair: “What is the most important research question in your discipline?” followed by “Why aren’t you working on it?”

GD Star Rating
Tagged as:
Trackback URL:
  • #1 is clearly the most important, except that it’s spelled “Twinkies”. One of the great things about it is that it can be easily combined with #2. I suppose you could combine it with the others as well if you like that sort of thing.


    Yes, indeed people often are unwilling to set political priorities, almost acting as if they intuitively do not grasp the finiteness of resources. This is probably because it feels good to support idealized causes without having to make hard, and possibly depressing choices. And yes, there’s a lot of “folks” just going for the signalling effect.

    Having said that it has to be pointed out that people have been “trained” to mistrust politicians and pundits who speak of priorities. Too often the question has been “we have to choose between this or that: we can’t do anything about racism because then all these people you (the majority race audience) know will die of cancer”. Instead of “if you had to divide $100 billion of govt budget, and given such and such info about impacts on societal utility and such and such info on the marginal utility curves of throwing money at the problems, how much would you give to fight cancer and how much to fight racism?”

    Inspired by another commenter on another one of your posts I’m inclined to say that it’s part ignorance/hypocrisy and part rational use of bayesian filters applied to life experiences.


    Of course one can’t do 3 and 4 without the occasional indulgence in things like 1 and 2 to reduce stress levels. Also, what’s the point of completing 3 and 4 if it means you can never have things like 1 and 2 anymore: if there were no joy in life I wouldn’t want to be cured from cancer. The goal of 3 and 4 IS to ensure broader and longer access to things like 1 and 2 (well except for those who enjoy the struggle itself). Shaming people for attaching even some slither of value to things like 1 and 2 is a cheap political trick/hypocritical sigmalling tactic.

    • IMASBA

      * sliver of value

  • Josh Morrison

    The problem with ranking is that it reduces multidimensional comparisons to a single dimension, warping and simplifying our understanding of the separate concepts (in the same way that occurs when you reduce a globe to a map)

  • abmodality

    I don’t think walks-for-diseases are a good way of allocating money between diseases or even a good way of getting resources to particular diseases. I suppose I could tell that to people who participate in those walks, but I don’t think that would be appropriate. People generally participate in those not to fight diseases but because they have some personal connection. They’re getting something out of it. They may not be totally self-aware about that motivation, but who among us is totally self-aware regarding all of our motivations?

    • Ronfar

      My father does long bicycle rides for MS because he likes to go on long bicycle rides with people.

  • Charlene Cobleigh Soreff

    “But if asked to rank #3 vs #4, we are much less comfortable.”

    Huh? I gain from #3 (if successful), while #4 is neutral. No contest!

  • The first two are more homogeneous. One Twinkie is exactly like another; one episode of Gilligan, almost the same.

    But the latter two have myriad expressions, some which one may think highly of and others one may despise. It’s unsurprising that we would have trouble comparing them globally.

    A more precise opposition would be what would be a greater achievement, eliminating all cancer or eliminating all racial injustice. I don’t think most people would have trouble commenting about their prioritization. If I’m right, this shows sacredness isn’t relevant. [In fact, I think we have no problem prioritizing the sacred, and doing so isn’t typically perceived as denigration. (cf. Dante’s Inferno and Paradisio.)]

    • Vitalik Buterin

      Thomas Sowell’s observation that private markets tend to be primarily about incremental judgements (eg. “should I spend this next $1000 on going out to restaurants more or upgrading my cell phone plan or getting a new computer?”), whereas politics tends to deal with categorical judgements (eg. “Those who give up liberty for security deserve neither.”) is pretty apt here [note that the original quote from Benjamic Franklin was quite a bit more incremental with its use of qualifiers, but the way it tends to be interpreted today is quite categorical].

      Twinkies and Gilligans are quite different from each other in terms of experience, but I think it’s the highly incremental and chunk-based nature of both experiences that makes the decision easier.

  • Nancy Lebovitz

    I *think* the whole point of sacredness is that you don’t do that sort of conscious comparison, but I admit this is my fast reaction, and I don’t have a theoretical structure to fit it into.

  • Michael.vassar

    Phillip Tetlock researches his.

    • Tetlock looked at intolerance for trading a bit of sacred for any amount of unsecured. I don’t recall him looking at trading sacred for sacred.

  • Auroch

    Neither of those registers as sacred to me. Any suggestions for other things to insert so that I could get a better handle on the idea in question?

    • Any suggestions for other things to insert

      Liberty and efficiency.

  • Lord

    Why would anyone want to rank them? The most important criteria is what you have to contribute and how effective it can be and that is a question we can only answer for ourselves. Are you asking what should we, either individually or as a society, spend more money on, potentially at the expense of reducing spending on the other? While money can help, it is rarely the most important factor and what it can accomplish is severely limited. It is not obvious we shouldn’t be spending more on both or less on both. Our preferences have less to do with this than our expectations of progress, so whether they are important or not is not itself that important.

  • dat_bro06

    A politician asked to prioritize #3 versus #4, even if he is thoughtful in his response (and even if he acknowledges that both are “NPV-generating” pursuits), can expect his words to be taken completely out of context in the popular press for the purpose of generating buzz/controversy, not to mention used against him by the lobby for the lower ranked group.

    Feels to me like this is as much a phenomenon related to media incentives; as well as ‘audiences’ who can’t think beyond topics for which they have a personal affinity or a built-in political / socio-political allegiance —- than it is about the inherent act of ranking sacred menu items.

    E.g., the question, “Which issues would you like the new politician to spend his political capital on in the upcoming legislative session?”, seems like one I could easily ask among friends or like-minded acquaintances over drinks.

    • This isn’t just about official politicians. We are all politicians to some extent, and this applies to most of us.

    • John_Maxwell_IV

      I wonder if what makes sacred values sacred is that when thinking about them, we’re more concerned with staying defensible against accusations than gaining something of value to us. (Maybe sacredness isn’t so much an intrinsic property of a value, but a fact about how some people in our society regard the value. If I make fun of Zeus, no one is likely to be especially upset with me because no one believes in him anymore. Spitting on someone is a trivial act that’s likely to get you in trouble because it’s regarded as offensive.) So then the question would be how to change the discussion environment so that the sort of accusations leveled against people intelligently trading off one value against another don’t fly.

      Robin says “We are all politicians to some extent, and this applies to most of us.” But if I’m right and being a politician is the rational response to a difficult set of circumstances, it might not make sense to blame people for behaving rationally… it might be more sensible to change the incentive structure. Maybe if we can determine the set of accusations that seem counterproductive, we can find a new brief way to accuse people of making counterproductive accusations?

      • I wonder if what makes sacred values sacred is that when thinking about them, we’re more concerned with staying defensible against accusations than gaining something of value to us.

        Consider, then, the analogy to things we can be literally accused of to our detriment: crime. The threat of accusation doesn’t render crimes unrankable; rather it intensifies the incentive to rank because you need to know which crime to defend against more avidly. Thus, in criminal law there is a clear hierarchy of sins; at a broad level, life trumps property.

  • bellisaurius

    “What is the most important research question in your discipline?” followed by “Why aren’t you working on it?”

    Why wouldn’t the person claim comparative advantage here? Yes, curing heart disease will save my life years, but Dr M is much better at that. I’ll keep plugging away at the acne problem.

    • I’m not sure of the point of Robin’s last paragraph. Is it that we avoid ranking the importance of research problems, hence work on the wrong topics?

      • Philon

        Maybe, but do research problems belong to “the sacred”? (Maybe in theology . . . .) Another question is: “Which is the most important discipline?”; few researchers can say with a straight face that *theirs* is the most important.
        I share your doubts about the last paragraph; the rest of the post is very insightful.

    • Vitor

      You’re right, of course. Instead of saying “it would be nice if people worked on what is most important”, it would be more adequate to say something like “it would be nice if people tried to maximize the expected utility they will produce”, which takes into account the relative importance of topics *and* personal factors like you suggested.

    • They could say that, but they rarely do.

  • zarzuelazen

    I came up with an objective method of ranking some abstract
    values based on a purely ontological argument.

    If you consider an ontology capable of full reality
    modelling (including full reflection, ability to model itself), you need 3 different levels of description – object, meta and meta-meta.

    Consider the teological domain. A careful search for natural categories that mirror the 3-level split yields 3 general classes of values which can then be ranked according to level of abstraction (the values with the highest level of abstraction have the most explanatory power, therefore the natural method is to give them higher rank – or greater worth).

    I come up with:

    BEAUTY – Meta-meta level – method or style of communicating values in general

    LIBERTY – Meta-level – ways of organizing social interactions

    PERFECTION – Object level – personal or individual virtues.

    So the natural hierarchy for terminal values is that BEAUTY is highest rank, followed by LIBERTY, then PERFECTION.

  • Cambias

    Fighting cancer is a concrete, achievable, measurable goal.

    Racial Justice is a subjective, elusive, and ever-shifting one.

    Obviously one should work for the second because that ensures job security.

    • Shawn

      Fighting cancer also involves similar subjective judgments (which type of cancer to fight, what methods to improve on…) and may not produce any meaningful results with a lifetime of work. They’re not as dissimilar as you suggest here, and cancer research shows no signs of ending any time soon.

  • JD

    I have absolutely no problem ranking #3 vs. #4. Because while I think they are equally noble, and perhaps equally important in some sense, one is at least more pressing. Specifically, I mean #4. (1) Not everyone gets cancer, but everyone is affected by racial injustice. Yes, everyone: some get unfair advantages, others unfair disadvantages; and whether your interests are considered important, not to mention how important they are considered to be, depends in part on your race. (2) My cancer cannot spread to you, but racial injustice can spread rapidly. One lynching goes unpunished and others are emboldened to follow suit. One corrupt officer gets away with something and others realize they can, too. Or on the flip side: a (completely justified, but still destructive) riot spreads and affects people who may not have been involved in the terrible incident that set it off. (3) Not every case of cancer is a tragedy (some people go into remission; some are at the stage of life where it’s just a matter of what gets them, not if and when; some people can afford superior treatment), but every case of racial injustice is a tragedy (even if the person who was treated unjustly could be said to deserve something bad happening to them, no one deserves racial injustice as punishment; mutatis mutandis for the cancer case, by the way). (4) Cancer and racial injustice often cut lives short, but racial injustice is more likely to cut lives early and its more tragic when it does (since it’s not just a harm, but also a wrong; we can’t control cancer, but we are in control of our actions, and thus of racial injustice; this goes back to the point about tragedy). So yeah, if I could snap my fingers and remove either cancer or racial injustice, I would choose racial injustice. No contest.

    • There actually are infectious cancers. Paul Ewald theorizes that most cancer is the result of infections, such as the link between HPV and cervical cancer.

      • JD

        First, your response engages in equivocation. “Most cancer is the result of infections” is much different from “cancer is infectious.” And even if there are some infectious cancers, they do not spread the way that racial injustice does. Therefore, my argument still stands.

      • Ewald is the one who theorizes that most cancers are the result of infections, his theory is not yet accepted by most of the field and so I myself do not state as a fact that most cancer is. You state that infectious cancers don’t spread the way racial injustice does, but I’m curious how specifically you think their transmission differs. Is one “more infectious” than the other? When I look at statistics for lynching, it doesn’t seem to track with what we’d normally consider to be racial justice (specifically, the decline occurs too early).

      • JD

        The decline in lynching has been well accounted for. Early lynching decreases the incentive for later lynching by creating a fear effect. The potential targets of future lynchings tend to leave or change their behavior. So lynching didn’t trail off because racism or racial injustice trailed off. It trailed off because it wasn’t needed as much to keep racial injustice in place (particularly once racists had regained political power). Focusing on lynching, which is only one facet of racial injustice, is literally missing the forest for the trees.

      • Lynching was the first example you brought up, and one we happen to have data on. You also mentioned riots, and there I will grant that some social scientists (like Timur Kuran or Alex Tabarrok) have written on how the chaos of it enables more chaos. However, it’s not clear to me that’s correlated with racial injustice. My impression of 1960s/1970s America is that there were relatively few riots in the most unjust places (the kind with a history of lynching) and more in places like California or northern cities.
        I should have made a point about cancer however: in Ewald’s theory, the cancer is downstream from infection, so focusing on cancer may be relatively ineffective at stopping the spread relative to focusing on the infection.

      • JD

        You seem to have lost the thread of this conversation. The original use of lynching in no way suggested that we should see a continuous increase in its usage. The point was that not punishing it emboldens those who would seek to expand racial injustice. So the statistical decline in lynching does not contravene this point in any way given that lynchings declined because racists were under the impression that they had successfully cowed their targets. Furthermore, my comments about riots were not limited to the 60s or 70s. I was thinking also about contemporary events like we are seeing in places like Missouri.

        Finally, if the cancer is downstream from the infection, that’s all the more reason for me to support #4 over #3. I’d rather fight a disease than a symptom.

      • Even today, I’m not sure about the correlation between riots and anyone’s measure of racial injustice. Ferguson seems rather average when it comes to the black/white arrest ratio, significantly less extreme than Santa Monica or Madison Wisconsin. The only respect in which Ferguson seems unusual is that the population is mostly black and the government isn’t, a result of its demographic change being relatively recent.

  • Sigivald

    I’m fond of this classic question pair: “What is the most important
    research question in your discipline?” followed by “Why aren’t you
    working on it?”

    “Because I’m not a rockstar and I can do more good by working on whatever it is I’m working on than that really hard problem – if it was easy, it’d already be fixed. And research is not strictly additive where if you throw more people at it you get linearly better results.”

    Also, “Because I have bills to pay and people are paying me to work on other things.”

    For that matter “because I really like working on this thing, even if it’s less important”.

    (Same thing about #3 and #4.

    People setting their own priorities and having their own preferences and directing their own lives is even more important that Maximizing Outcome Utility At All Times.

    Do something you think is good. Do something that pleases you and hurts no-one. Either. Both. But live, rather than worrying about utility maximization in every act or ever choice.

    Don’t let anyone shame you for “not doing enough”, as long as you’re not actively impeding others..

    They can go to outer darkness and wailing of teeth* if they want to be that way.

    * Worse than wailing, and gnashing of teeth, because, well, think about wailing teeth!)

  • stijn

    Hi Robin, this reminds me of what I call a moral gravity bias: https://stijnbruers.wordpress.com/2014/08/08/the-moral-gravity-bias/
    Ranking 3<4 seems like you lower the importance or value of 3 instead of increase the value of 4.

  • Shane Revis

    Does anyone think that person A’s suffering can accurately be stated to be objectively worse than person B’s suffering, if both of their suffering is of an intense kind – e.g., a female stuck in sexual slavery versus a male stuck in solitary confinement? If one affliction is objectively worse, can we say this it is also always subjectively worse? If we can, wouldn’t the effective altruist do the MOST good by only focusing on the objective worst suffering to befall man?
    The reason I ask these questions is because I am passionate about using a future career to help the objective worse off of humanity, but I don’t know how to rank the WORST types of suffering in order to choose what to major in. Anything anyone can say is appreciated. It’d also benefit me to hear about philosophical works on this issue if anyone has came across it. I’d love to merely ask Peter Singer, but it looks like he doesn’t accept emails from random members of the public. Instead of replying here, anyone with any info should email me. My email is shane33dr@gmail.com

    Thanks so much!