Fillers Neglect Framers

We can usefully divide intellectual tasks into two sets: filling and framing.   Fillers add more useful detail or content within some framework, while framers explore possible new frameworks.  While both tasks are essential, framing has higher variance; most frame attempts fail, but a few produce great value. 

We tend to be best at evaluating people with skills like our skills.   So good fillers can best evaluate other fillers, and good framers can best evaluate other framers.   You might think then that these groups would separate, so that fillers choose the next generation of fillers and framers choose the next generation of framers.   

What I see in the intellectual world around me, however, is very different.    When people choose new subordinates and assistants, who will eventually take over their institutional roles in the next generation, I see both fillers and framers choosing fillers.   Each person tends to prefer successors who will increase his own fame and reputation by build directly on his own work.    The more people who have followed up on your work, the more important you must be. 

Since fillers tend to more directly build on the work of others, both fillers and framers tend to choose fillers as successors.   For fillers it is a no-brainer; fillers are more like them.  For framers there is tension; fillers are less like them, and harder to evaluate.  But fillers are more likely to add to the glory of framers’ great new frames, and so framers choose fillers as successors nonetheless.

As a result, our institutions do not do very well at selecting promising framers; while we do a decent job of selecting and encouraging promising fillers from the next generation, successful framers tend to be accidents, people we chose as fillers who turned out to be good framers.   Presumably our production of new frames suffers as a result.

Addendum: I think I notice a similar trend in referee reports, which tend to reject framing-style papers, even those with high expected values compared to typical filling-style publications. 

GD Star Rating
Tagged as:
Trackback URL:
  • Gordon Worley

    Thanks for this interesting post. I never put it into words before, but I’ve been feeling this tension for a long time. I’m in my third year of working on a PhD and I always find I’m in an awkward position because I’m not closely tied with an advisor. My advisor does his work and I do mine, and he helps me along when he can, but I’m pushing into new mathematical territory in a way that may or may not produce anything interesting. Meanwhile my fellow students are publishing lots of papers and finishing their degrees because they work on someone’s problem and eventually have enough results to write a thesis and graduate. Not to say that their work is invalid or even of lesser quality or importance, since many useful results come from this kind of work, but it’s not what I want to do and I pay a price for it.

    So I agree. As a framer, in your language, I’m at a disadvantage because the system doesn’t want me until after I graduate. But thanks to this post, I hope that after I graduate and have students of my own I can try to help my fellow framers.

  • Bruce G Charlton

    Nicely put. This could be seen as a new slant on Thomas Kuhn’s distinction between ‘revolutionary’ and ‘normal’ science.

    I’ve just finished a mini analysis of trends in Nobel science prizes as a measure of revolutionary science.

    Only about 15 institutions in the world have three or more Nobels in the past 20 years, all but one of them are in the USA. Top is MIT, then Stanford, then Columbia and Chicago – only then comes Harvard. Quite a few new institutions have emerged in the past 20 years – all in the Western states (eg. U Cal branches, Boulder Colorado, U Washington) The rest appear to be sporadic occurences in the Westernized nations.

    So – perhaps only the US system actively promotes revoutionary science (‘framers’), and increasingly in the West.

  • Gustavo Lacerda

    Thanks for this post. The two comments above already say most of what I had to say.

    It feels terribly unfair: not only do fillers (conformists?) get industry jobs more easily, but this is the case *even* in academia.

    I can imagine more explanations for why admissions committees select “fillers”:
    * risk-aversity (although we shouldn’t expect this from large enough departments)
    * framer students are harder to assign to an advisor
    * if you’re a framer, it’s harder to have track record of good work with which to signal your quality (i.e. I would argue that even framers have a hard time evaluating other framers)

    What would you propose as a solution?

  • Robin Hanson

    Gustavo, the main fix I could imagine is to give framers more publicity and credit for mentoring new framers.

  • Matthew

    It’s always much more politically expedient to bash the newest framers and laugh at them. . . Until it turns out that they are right.

    The problem is, like other institutions before it, academia has become a sinecure for fillers. Being a framer is bad for your career, and the bigger your frame, the more likely that you will be dismissed by the fillers of the previous frames (who are not at all wont to admit that there might be something amiss with the frames they are busy filling).

    The existence of this blog is an encouraging step in the right direction, though. . .

  • Gustavo Lacerda

    “The main fix I could imagine is to give framers more publicity and credit for mentoring new framers.”

    Robin, can you translate this into a specific recommendation? i.e. if you were in a position of power, what would you do about it? Create awards / professorships for those who mentor framers?

  • Robin Hanson

    Gustavo, this is as far as I have thought about this subject.

  • Eliezer Yudkowsky

    Framers work from within metaframes – larger worldviews – that are both hard to communicate, and hard to submit directly to the cleansing fire of experiment. Einstein’s SR and GR were both falsifiable, but stemmed from a larger philosophy of invariance that is much more ineffable (the only really good exposition I’ve ever seen of it is Julian Barbour’s).

    When framers meet, they tend to clash because their background metaframes are dissimilar. Einstein’s metaframe was substantially more dissimilar from, say, Lorentz’s or Minkowski’s metaframes, than their actual contributions to physics.

    Metaframes tend to contain not-easily-testable elements and lofty-sounding methodological prescriptions. Faced with such temptation, it would take a great deal of mental discipline to prevent some nonsense from working its way in. And different nonsenses are likely to disagree, since they have no mutual territory to be consistent with.

    When framers meet and sniff each other suspiciously, they have sensitive noses and plenty to take offense at – even if their actual theories look pretty similar. Or to put it another way, I’m a lot more likely to take offense at some would-be framer’s side beliefs about neural networks, compared to a neural network algorithm as innocent symbols on paper. (Though of course this is because *other* people have accumulated nonsensical but hard-to-dispel side ideas about neural networks – my own metaframe is naturally free of such debris.)

    No, framers can’t just get rid of the metaframes and work without them. Thank you for asking.

  • Bruce G Charlton

    The recommendation I have made (for medical science) is that ‘framers’/ pure scientists/ revolutionary scientists should form a separate, distinct scientific specialty – with different institutions, funding, journals, conferences etc – linked to different evaluation criteria than would be used for ‘fillers’/ applied scientists/ normal scientists.

    The key is different evaluation criteria, with credit given to quality and potential importance of work; because fillers can be more productive and reliable over the short term.

  • Bruce G Charlton

    Reference to my comment above is ‘The future of pure medical science’ at:

  • Anders Sandberg

    In Europe it is common that after a postdoc researchers return to their home group, continuing working with their advisor. This is less common in the US. The European system seems to encourage fillers more, both because they have an easier time working in the tiny academic empire of their advisor and that framers might become increasingly fillers out of practical necessity. Meanwhile in the US framers have an easier time forming their own groups, which at least means that their particular frame gets a chance to be tested.

    One way of dealing with this might be to make sure grant money (especially for younger researchers) go to the researchers and not to advisors or institutions. They should be able to set up shop at a university that wants them, perhaps cooperatively with other new researchers.

  • Robin Hanson

    Anders, excellent comment and suggestion.

  • Adrian Tschoegl

    I am reminded of a colleague who used to remark “Pioneers get scalped by the Indians”, and “You can always tell a pioneer, they are the ones with arrows in their backs.” I suspect that short tenure clocks select against framers. Getting an article accepted is a process with a large stochastic element and attempting to get a new frame out into the literature would increase the risk of failure to hurdle tenure. Post tenure, the researcher has more freedom, but the persons who are most likely to pass the process are the ones for whom it is least constraining. This line of argument would suggest that extending the tenure clock substantially would increase the number of framers.

  • Rue Des Quatre Vents

    I can’t quite claim to be in the framer mold, but I left grad school because I saw no value in adding the 10th decimal place to someone else’s research. At any rate, I believe there are major institutional problems, at least in humanities, that occlude the creation of bold new ideas. One of the main problems, as you’ve pointed out, has to do with the patronage and nepotism endemic to the transistion from undergrad to graduate to professorial studies. To my thinking, one proposal might be to prevent members of a department from choosing the entering grad student class. In philosophy for example, I think it terribly quaint to let Kantians cultivate Kantians, Rawlsians Rawlsians and so on. Recommendations, while important, are overrated, largely because they’re prone to confirmation bias: professors will tend to like recommendations that exemplify their own perceived virtues. (Well that’s best case. Worst case, someone will respect a recommendation because they respect or like or are friends with the recommender.) In general, to foster genuine creativity in graduate school and further, academia needs greater fragmenting, separating research interests from professional power.

  • http://e Cedrick Shulse

    Love rules without rules