Some Thoughts on Public Discourse

Thanks to Ben Hoff­man and sev­eral of my cowork­ers for re­view­ing a draft of this.

It seems to me that there have been some dis­agree­ments lately in the effec­tive al­tru­ism com­mu­nity re­gard­ing the proper role and con­duct for pub­lic dis­course (in par­tic­u­lar, dis­cus­sions on the pub­lic Web). I de­cided to share some thoughts on this topic, be­cause (a) my thoughts on the mat­ter have evolved a lot over time; (b) some of the dis­agree­ment and frus­tra­tion I’ve seen has been speci­fi­cally over the way Open Philan­thropy ap­proaches pub­lic dis­course, and rather than re­spond­ing to com­ments piece­meal I thought it would be more pro­duc­tive to lay out my views at a high level.

First I’ll dis­cuss my past and pre­sent views on the role of pub­lic dis­course, and why they’ve changed. (In brief, I see sig­nifi­cantly fewer benefits and greater costs to pub­lic dis­course than I used to, but I still value it.) Then I’ll list some guidelines I fol­low in pub­lic dis­course, and some ob­ser­va­tions about what kinds of re­sponses peo­ple are likely to get from the Open Philan­thropy Pro­ject de­pend­ing on how they ap­proach it.

By “pub­lic dis­course,” I mean com­mu­ni­ca­tions that are available to the pub­lic and that are pri­mar­ily aimed at clearly de­scribing one’s think­ing, ex­plor­ing differ­ences with oth­ers, etc. with a fo­cus on truth-seek­ing rather than on fundrais­ing, ad­vo­cacy, pro­mo­tion, etc.

My past and pre­sent views on the role of pub­lic discourse

Vipul Naik re­cently quoted a 2007 blog post of mine as say­ing, “When I look at large foun­da­tions mak­ing mul­ti­mil­lion-dol­lar de­ci­sions while keep­ing their data and rea­son­ing ‘con­fi­den­tial’ – all I see is a gi­gan­tic pile of the most un­be­liev­ably mind-blow­ing ar­ro­gance of all time. I’m se­ri­ous.” (It con­tinues, “De­cid­ing where to give is too hard and too com­plex – with all the judg­ment calls and all the differ­ent kinds of think­ing it in­volves, there is just no way Bill Gates wouldn’t benefit from hav­ing more out­side per­spec­tives. I don’t care how smart he is.”)

I’d guess that there are many other quotes in a similar vein. My old writ­ing style tended to­ward hy­per­bole rather than care­ful state­ment of the strength of my views, but over­all, I think this quote cap­tures some­thing I be­lieved. It’s hard to say ex­actly what I thought more than nine years ago, but I think some key parts of my model were:

  1. On any given topic, knowl­edge and in­sight are broadly dis­tributed. It’s hard to pre­dict what sort of per­son will have helpful in­put, and hard to as­sess an idea with­out sub­ject­ing it to a broad “mar­ket­place of ideas.” Thus, the ideal way to ar­rive at truth would be to broad­cast one’s views in as much de­tail to as many peo­ple as pos­si­ble and in­vite max­i­mal in­put.

  2. Foun­da­tions face lit­tle down­side to pub­lic dis­course, be­cause they are not ac­countable to the pub­lic at large. Their hes­i­ta­tion to en­gage in pub­lic dis­course can most eas­ily be ex­plained by want­ing to avoid em­bar­rass­ment, bad press, etc. - and/​or by fol­low­ing habits de­rived from other kinds of in­sti­tu­tions (com­pa­nies, gov­ern­ment agen­cies) that face more sub­stan­tive down­sides. Be­cause of their lack of ac­countabil­ity to the pub­lic at large, foun­da­tions are uniquely po­si­tioned to raise the level of pub­lic dis­course and set ex­am­ples for other in­sti­tu­tions, so it’s a shame that they don’t.


Over time, I’ve come to es­ti­mate both less benefit and more cost to pub­lic dis­course. The de­tails of this evolu­tion are laid out in Challenges of Trans­parency (2014) and Up­date on How We’re Think­ing about Open­ness and In­for­ma­tion Shar­ing (2016).

The biggest sur­prise for me, over time, has been on the “benefits” side of the ledger. This point is noted in the above blog posts, but it’s worth go­ing into some de­tail here.

For nearly a decade now, we’ve been putting a huge amount of work into putting the de­tails of our rea­son­ing out in pub­lic, and yet I am hard-pressed to think of cases (es­pe­cially in more re­cent years) where a pub­lic com­ment from an un­ex­pected source raised novel im­por­tant con­sid­er­a­tions, lead­ing to a change in views. This isn’t be­cause no­body has raised novel im­por­tant con­sid­er­a­tions, and it cer­tainly isn’t be­cause we haven’t changed our views. Rather, it seems to be the case that we get a large amount of valuable and im­por­tant crit­i­cism from a rel­a­tively small num­ber of highly en­gaged, highly in­formed peo­ple. Such peo­ple tend to spend a lot of time read­ing, think­ing and writ­ing about rele­vant top­ics, to fol­low our work closely, and to have a great deal of con­text. They also tend to be peo­ple who form re­la­tion­ships of some sort with us be­yond pub­lic dis­course.

The feed­back and ques­tions we get from out­side of this set of peo­ple are of­ten rea­son­able but fa­mil­iar, seem­ingly un­rea­son­able, or difficult for us to make sense of. In many cases, it may be that we’re wrong and our ex­ter­nal crit­ics are right; our lack of learn­ing from these ex­ter­nal crit­ics may re­flect our own flaws, or difficul­ties in­her­ent to a situ­a­tion where peo­ple who have thought about a topic at length, form­ing their own in­tel­lec­tual frame­works and pre­sup­po­si­tions, try to learn from peo­ple who bring very differ­ent com­mu­ni­ca­tion styles and pre­sup­po­si­tions.

The dy­namic seems quite similar to that of academia: aca­demics tend to get very deep into their top­ics and in­tel­lec­tual frame­works, and it is quite un­usual for them to be moved by the ar­gu­ments of those un­fa­mil­iar with their field. I think it is some­times jus­tified and some­times un­jus­tified to be so un­moved by ar­gu­ments from out­siders.

Re­gard­less of the un­der­ly­ing rea­sons, we have put a lot of effort over a long pe­riod of time into pub­lic dis­course, and have reaped very lit­tle of this par­tic­u­lar kind of benefit (though we have reaped other benefits—more be­low). I’m aware that this claim may strike some as un­likely and/​or dis­ap­point­ing, but it is my lived ex­pe­rience, and I think at this point it would be hard to ar­gue that it is sim­ply ex­plained by a lack of effort or in­ter­est in pub­lic dis­course.

I have also come to have a bet­ter un­der­stand­ing of the costs of pub­lic dis­course. Th­ese costs are enu­mer­ated in some de­tail in the posts linked above. A cou­ple as­pects that seem worth high­light­ing here are:

  • I’ve come to ap­pre­ci­ate the tan­gible benefits of hav­ing a good rep­u­ta­tion, in terms of hiring, re­ten­tion, ac­cess to ex­perts, etc. I’ve also raised my es­ti­mate of how risky pub­lic dis­course can be to our rep­u­ta­tion; I think there are a lot of peo­ple who ac­tively seek out op­por­tu­ni­ties to draw at­ten­tion by quot­ing things in bad faith, and a lot of other peo­ple who never cor­rect the first im­pres­sion they get from en­coun­ter­ing these quotes.

  • Be­cause of how much valuable feed­back we’ve got­ten from “in­sid­ers” who know the topic at hand well, I’ve come to feel that care­less pub­lic dis­course would do more harm than good to our abil­ity to learn from feed­back, via dam­ag­ing re­la­tion­ships with the peo­ple most likely to give good feed­back.

I rec­og­nize that an out­sider might be skep­ti­cal of this nar­ra­tive, be­cause there is a sim­ple al­ter­na­tive one: that we val­ued pub­lic dis­course when we were “out­siders” des­per­ate for more in­for­ma­tion, and we value it less now that we are “in­sid­ers” who usu­ally are able to get the in­for­ma­tion we want. I think this is, in fact, part of why my at­ti­tude has changed; but I think the above fac­tors are more im­por­tant.

Why I still value pub­lic discourse

De­spite all of the above con­sid­er­a­tions, we still en­gage in a large amount of pub­lic dis­course by the stan­dards of a fun­der, and I’m glad we do. Some rea­sons:

  • I think our pub­lic con­tent helps oth­ers un­der­stand where we’re com­ing from and why we do what we do. I think that this has, over time, been a ma­jor net pos­i­tive for our rep­u­ta­tion, and in par­tic­u­lar for our abil­ity to con­nect with peo­ple who deeply res­onate with our val­ues and ap­proach. Such peo­ple can later be­come highly in­formed, en­gaged crit­ics who in­fluence our views.

  • I still em­pathize with my ear­lier self, and my frus­tra­tion at not be­ing able to learn about top­ics I cared about, un­der­stand the think­ing of key in­sti­tu­tions, and gen­er­ally get “up to speed” in key ar­eas. I worry about what I per­ceive as a lack of men­tor­ship in the effec­tive al­tru­ism com­mu­nity, and I won­der how the next gen­er­a­tion of highly in­formed, en­gaged crit­ics (al­luded to above) is sup­posed to de­velop if all sub­stan­tive con­ver­sa­tions are hap­pen­ing offline.

  • Writ­ing pub­lic con­tent of­ten forces me to clar­ify my own think­ing, helps (via oth­ers’ re­ac­tions) high­light the most con­tro­ver­sial parts of it (which then leads to fur­ther re­flec­tion), and of­ten leads to bet­ter feed­back than we would’ve oth­er­wise got­ten from highly in­formed and en­gaged peo­ple.

How­ever, it is much more costly for me to par­ti­ci­pate in pub­lic dis­course than it used to be, both be­cause the stakes are higher (call­ing for more care in com­mu­ni­ca­tions) and be­cause I have less time.

I’ll add that I don’t see it as a cost to us when some­one pub­li­cly crit­i­cizes our work, and I’d gen­er­ally like to see more of this rather than less—pro­vided that such crit­i­cism does not mis­rep­re­sent our views and ac­tions. And as dis­cussed be­low, I think mis­rep­re­sen­ta­tion is fairly easy to avoid. It is still the case that the peo­ple we most want to reach are peo­ple we ex­pect to fairly con­sider differ­ent ar­gu­ments and reach rea­son­able con­clu­sions; if pub­lic crit­i­cism hurt our rep­u­ta­tion among such peo­ple (with­out mis­rep­re­sent­ing our views), I would by de­fault con­sider this de­served and good.

How I ap­proach pub­lic dis­course today

Prin­ci­ples I gen­er­ally fol­low in pub­lic discourse

  • I am very se­lec­tive about where I en­gage. In gen­eral, if I write some­thing pub­li­cly, it ei­ther (a) lays out a fun­da­men­tal set of ideas and ar­gu­ments that I ex­pect to link to re­peat­edly in or­der to help peo­ple un­der­stand some­thing im­por­tant about my think­ing; (b) ad­dresses a crit­i­cism/​con­cern that is im­por­tant to one or more spe­cific peo­ple whose re­la­tion­ships I value; or (c) ad­dresses a ques­tion posed di­rectly to me, while not say­ing more than needed to ac­com­plish this.

  • I strive to con­vey the nu­ances of my think­ing, and gen­er­ally pri­ori­tize avoid­ing harm over get­ting at­ten­tion. My com­mu­ni­ca­tions of­ten take pa­tience to read through, but have rel­a­tively low risk of leav­ing peo­ple with prob­le­matic im­pres­sions.

  • I always seek at least one other pair of eyes to look over what I’ve writ­ten be­fore I post it pub­li­cly. Usu­ally much more than one.

  • I always run con­tent by (a sam­ple of) the peo­ple whose views I am ad­dress­ing and the peo­ple I am di­rectly nam­ing/​com­ment­ing on, as­sum­ing they are not out­right ad­ver­saries (e.g., poli­ti­cal op­po­nents of ideas the con­tent is ar­gu­ing for). I con­sider this a uni­ver­sally, un­ques­tion­ably good prac­tice. Al­most always, I learn some nu­ance of their views or ac­tions that leads to my im­prov­ing the con­tent, from both my per­spec­tive and theirs. Al­most always, they ap­pre­ci­ate the chance to com­ment. Some­times, they make small re­quests (e.g. about timing) that I can eas­ily ac­com­mo­date. I think this prac­tice is good for both the ac­cu­racy of the con­tent and my re­la­tion­ships with the peo­ple af­fected. And it does not make crit­i­cism more costly for me—quite the con­trary, it makes crit­i­cism less costly (in terms of re­la­tion­ships, and in terms of time due to im­proved ac­cu­racy), and in­creases the quan­tity of crit­i­cism I’m will­ing to make. I see es­sen­tially no case against this prac­tice. Note that run­ning con­tent by peo­ple is not the same as giv­ing ed­i­to­rial con­trol or seek­ing their per­mis­sion (see next point).

  • When run­ning con­tent by oth­ers, I com­mu­ni­cate ex­plic­itly about my ex­pec­ta­tions for their feed­back. In par­tic­u­lar, I am clear about when I ex­pect to pub­lish by de­fault, and clear that I am not offer­ing them ed­i­to­rial con­trol. I am usu­ally happy to de­lay pub­li­ca­tion for an agreed-upon, non-ex­ces­sive amount of time. I will make cor­rec­tions to my con­tent if it im­proves ac­cu­racy, and some­times if it offers a ma­jor re­la­tion­ship benefit for a neg­ligible sub­stance cost. But I do not wait in­definitely if there’s no re­sponse, and I do not ac­cept sug­ges­tions that re­sult in my writ­ing some­thing in some­one else’s voice, or stat­ing some­thing I don’t be­lieve to be true and fair.

  • I don’t try to write for ev­ery­one. I try to write for our most thought­ful crit­ics and for our most val­ued pre­sent and fu­ture re­la­tion­ships. I do not try to ad­dress ev­ery de­tail of crit­i­cisms and claims peo­ple make, or to ad­dress ev­ery mis­con­cep­tion some­one might have.

  • When writ­ing at length, I provide a sum­mary and a roadmap, and I gen­er­ally try to make it easy for peo­ple to quickly un­der­stand my ma­jor claims and how to find the sup­port­ing ar­gu­ments be­hind them.

  • I try to avoid straw-man­ning, steel-man­ning, and nit­pick­ing. I strive for an ac­cu­rate un­der­stand­ing of the most im­por­tant premises be­hind some­one’s most im­por­tant de­ci­sions, and ad­dress those. (As a side note, I find it very un­satis­fy­ing to en­gage with “steel-man” ver­sions of my ar­gu­ments, which rarely re­sem­ble my ac­tual views.)

  • I try to bear in mind how limited my un­der­stand­ing of oth­ers’ views is. I be­lieve it is of­ten pro­hibitively difficult and time-con­sum­ing to com­mu­ni­cate com­pre­hen­sively about the rea­son­ing be­hind one’s think­ing. I of­ten ob­serve oth­ers who have ex­tremely in­ac­cu­rate pic­tures of my think­ing but are quite con­fi­dent in their anal­y­sis, and I don’t want to make that mis­take. So I have a high bar for mak­ing judg­ments and as­ser­tions about the ra­tio­nal­ity, char­ac­ter, and val­ues of peo­ple based on pub­lic dis­course, and I gen­er­ally con­fine my writ­ing to top­ics that don’t rely on views about these things. More on this in the note at the bot­tom.

  • Espe­cially when deal­ing with or­ga­ni­za­tions, I re­strict my defi­ni­tion of “im­por­tant dis­agree­ments” to “be­liefs un­der­ly­ing im­por­tant ac­tions I dis­agree with.” I think there is some­times a prac­tice in the effec­tive al­tru­ist and ra­tio­nal­ist com­mu­ni­ties of pay­ing a lot of at­ten­tion to in­con­sis­ten­cies, or to ac­tions that seem know­ably non-op­ti­mal, or to other dis­agree­ments, even when they don’t per­tain to im­por­tant dis­agree­ments on ac­tions, on the grounds that such things demon­strate a lack of good epistemic stan­dards or a lack of value al­ign­ment. I think this is mis­guided. As an or­ga­ni­za­tion leader, I am con­stantly mak­ing trade­offs about when to think care­fully about a dilemma and reach a great an­swer, vs. when to go with a hacky “mid­dle ground” ap­proach that is know­ably non-op­ti­mal but also min­i­mizes the risk of any par­tic­u­lar dis­aster, vs. when to sim­ply defer to oth­ers or stick with in­er­tia and ac­cept the risk of do­ing some­thing clearly flawed. I strongly feel that one can­not get a read on the val­ues and episte­mol­ogy of an or­ga­ni­za­tion’s lead­er­ship by fo­cus­ing on the de­ci­sions that seem sim­plest to an­a­lyze; one must fo­cus on the de­ci­sions that are im­por­tant and that one feels could have been done in a spe­cific bet­ter way. Even then, one will of­ten lack a great deal of con­text.

What to ex­pect in terms of re­sponses from Open Philanthropy

If you have ques­tions or crit­i­cisms of Open Philan­thropy and are hop­ing for a di­rect re­sponse, here are some gen­eral guidelines for what to ex­pect:

  • If some­one com­ments on the Open Philan­thropy Blog (in­clud­ing one of our reg­u­lar open threads, and even if the open thread is old) and asks a di­rect ques­tion, I or some­one else from Open Philan­thropy will an­swer it. (Tag­ging me on Face­book or men­tion­ing Open Philan­thropy in a fo­rum com­ment does not have the same effect, at least not con­sis­tently. I feel I owe a re­sponse to peo­ple who speci­fi­cally “ap­proach” Open Philan­thropy with a ques­tion, which in­cludes peo­ple who email, peo­ple who com­ment on our blog, and peo­ple who come to our events; I don’t feel a similar obli­ga­tion to peo­ple who ex­press in­ter­est/​cu­ri­os­ity in our views but are ul­ti­mately hav­ing their own dis­cus­sion.)

  • If some­one whose re­la­tion­ship I value speci­fi­cally tells me they are cu­ri­ous about my an­swer to a ques­tion or crit­i­cism, and that they think it’s worth my time to en­gage, I gen­er­ally do.

  • I gen­er­ally ad­dress spe­cific claims that seem cru­cial to an ar­gu­ment im­ply­ing Open Philan­thropy’s ac­tions are sub­op­ti­mal. I of­ten do not re­spond at all to claims that seem tan­gen­tial, or to vague ex­pres­sions of dis­agree­ment/​dis­satis­fac­tion that I can’t pin down to par­tic­u­lar claims.

  • I have limited time for read­ing as well as writ­ing. When some­one writes a long cri­tique of Open Philan­thropy, I look to the sum­mary to de­ter­mine whether there’s any­thing worth ad­dress­ing, then drill down to see the sup­port­ing ar­gu­ments be­hind key points. When the sum­mary is ab­sent or in­effec­tive, this usu­ally means I will not re­spond in a satis­fy­ing way; I do not con­sider my­self obli­gated to read long (>5pg) pieces just be­cause they ad­dress Open Philan­thropy.

  • I do not feel offended when peo­ple crit­i­cize us, and I also do not gen­er­ally feel taxed by it (I of­ten feel no obli­ga­tion no re­spond, and when re­spond­ing out of obli­ga­tion, I of­ten re­spond quite briefly). I lower my opinion of some­one when I feel they are mis­rep­re­sent­ing us (or oth­ers), but I gen­er­ally do not lower my opinion of some­one sim­ply be­cause they ex­press crit­i­cism or dis­agree­ment. (I elab­o­rate on this point in a note be­low, since peo­ple who re­viewed a draft of this piece gen­er­ally found this state­ment sur­pris­ing.) As noted above, I be­lieve that mis­rep­re­sen­ta­tions can wrongfully dam­age our rep­u­ta­tion, but I do not worry about the rep­u­ta­tional effects of crit­i­cism based on ac­cu­rate rep­re­sen­ta­tions. I don’t have a de­sire for peo­ple to crit­i­cize us less; I do have a de­sire for peo­ple to be more un­der­stand­ing about the fact that we some­times do not re­spond at length or at all.

  • When peo­ple run things by me be­fore post­ing them, I try to be helpful by cor­rect­ing any in­ac­cu­ra­cies I no­tice, though I of­ten do not en­gage much be­yond that, and will usu­ally save sub­stan­tive dis­agree­ments for pub­lic dis­course (or de­cline to get into them at all).

  • I have a strong bias to en­gage peo­ple who seem like they are think­ing hard, rea­son­ing care­fully, en­gag­ing re­spect­fully, and do­ing their best to un­der­stand the pub­lic con­tent that is already available, even if this doesn’t fit my other crite­ria.

A note on eval­u­at­ing peo­ple based on pub­lic discourse

This sec­tion is tan­gen­tial to the rest of the piece, but I in­clude it as an elab­o­ra­tion on a cou­ple of com­ments above that stem from a some­what un­usual at­ti­tude to­ward eval­u­at­ing peo­ple.

I think it’s good and im­por­tant to form views about peo­ple’s strengths, weak­nesses, val­ues and char­ac­ter. How­ever, I am gen­er­ally against form­ing nega­tive views of peo­ple (on any of these di­men­sions) based on seem­ingly in­cor­rect, poorly rea­soned, or seem­ingly bad-val­ues-driven pub­lic state­ments. When a pub­lic state­ment is not mis­lead­ing or tan­gibly harm­ful, I gen­er­ally am open to treat­ing it as a pos­i­tive up­date on the per­son mak­ing the state­ment, but not to treat­ing it as worse news about them than if they had sim­ply said noth­ing.

The ba­sic rea­sons for this at­ti­tude are:

  • I think it is very easy to be wrong about the im­pli­ca­tions of some­one’s pub­lic state­ment. It could be that their state­ment was poorly ex­pressed, or aimed at an­other au­di­ence; that the reader is failing to un­der­stand sub­tleties of it; or that the state­ment is in fact wrong, but that it merely re­flects that the per­son who made it hasn’t been suffi­ciently re­flec­tive or knowl­edge­able on the topic yet (and could be­come so later).

  • I think pub­lic dis­course would be less costly and more pro­duc­tive for ev­ery­one if the at­ti­tude I take were more com­mon. I think that one of the best ways to learn is to share one’s im­pres­sions, even (es­pe­cially) when they might be badly wrong. I wish that pub­lic dis­course could in­clude more low-cau­tion ex­plo­ra­tion, with­out the risks that cur­rently come with such things.

  • I gen­er­ally be­lieve in eval­u­at­ing peo­ple based on what they’ve ac­com­plished and what they’ve had the op­por­tu­nity to ac­com­plish, plus any tan­gible harm (in­clud­ing mis­in­for­ma­tion) they’ve caused. I think this ap­proach works well for iden­ti­fy­ing peo­ple who are promis­ing and peo­ple whom I should steer clear of; I think other meth­ods add lit­tle of value and mostly add noise.

I up­date nega­tively on peo­ple who mis­lead (in­clud­ing ex­press­ing great con­fi­dence while be­ing wrong, and es­pe­cially in­clud­ing avoid­able mischar­ac­ter­i­za­tions of oth­ers’ views); peo­ple who do tan­gible dam­age (usu­ally by mis­lead­ing); and peo­ple who cre­ate lit­tle of value de­spite large amounts of op­por­tu­nity and time in­vest­ment. But if some­one is sim­ply ex­press­ing a view and be­ing open about their rea­sons for hold­ing it, I try (largely suc­cess­fully, I think) not to make any nega­tive up­dates sim­ply based on the sub­stance.