Some of Pearce’s ideas goes against certain established notions in EA: e.g. he thinks sentience of classical digital computers is impossible under the known laws of physics, that minimising suffering should take priority over increasing happiness of the already well-off, that environmental interventions alone, w/o raising individuals’ hedonic setpoints and making these individuals invincible to severe suffering, cannot solve the problem of suffering and achieve sustainable high wellbeing for all.
I also should mention that I’m biased in proposing this tag, as Pearce’s work played a major role in my becoming an EA.
Arguments for:
For over 25 years David Pearce has been researching and writing about addressing the root cause of suffering on the planet using bio/nano/info/robo technology.
Pearce has been raising awareness about, and proposing solutions for, wild-animal suffering at least since 1995.
Several relatively prominent EAs cite Pearce’s work as having major influence on their values, including Brain Tomasik, the Qualia Research Institute’s Andrés Gómez Emilsson, and the Center for Reducing Suffering’s Magnus Vinding. Another recognition of Pearce’s work is his being invited as a speaker for EA Global: Melbourne 2015.
Unlike most other transhumanists, Pearce is antispeciesist and advocates using technology to benefit all sentient life.
Michael is correct that the inclusion criteria for entries of individual people hasn’t been made explicit. In deciding whether a person was a fit subject for an article, I haven’t followed any conscious procedure, but merely relied on my subjective sense of whether the person deserved a dedicated article. Looking at the list of people I ended up including, a few clusters emerge:
people who have had an extraordinary positive impact, and that are often discussed in EA circles (Arkhipov, Zhdanov, etc.)
people who have attained eminence in their fields and who are connected to EA to a significant degree (Pinker, Hassabis, Boeree, etc.)
academics who have conducted research of clear EA relevance (Ng, Duflo, Parfit, Tetlock, etc.)
historical figures that may be regarded as proto-EAs or that are seen as having inspired the EA movement (Bentham, Mill, Russell, etc.)
“core figures” in the EA community (Shulman, Christiano, Tomasik, etc.)
Some people, such Bostrom, MacAskill, Ord, fit into more than one of these clusters. My sense is that David Pearce doesn’t fit into any of the clusters. It seems relatively uncontroversial that he doesn’t fit into clusters 1-4, so the relevant question—at least if one broadly agrees with the approach I’ve taken—is whether he is sufficiently close to the “core” to merit inclusion as part of cluster 5.
As someone who has been involved with EA since its inception and who has (I believe) a reasonably good sense of how central to the movement different people have been, my impression is that Pearce isn’t central enough. If others have different (or similar) impressions, I would encourage them to post them here. We could, alternatively, try to go beyond impressionistic evidence and look at more objective measures, such as citation counts (broadly construed to include not just academic citations but links from the EA Forum and EA Blogs), though conducting that kind of analysis might be time consuming and may not be fully conclusive. Do others have thoughts on how to operationalize the relevant criteria?
Do others have thoughts on how to operationalize the relevant criteria?
FWIW, I think your comment is already a good step! I think I broadly agree that those people who fit into at least one of those clusters should typically have entries, and those who don’t shouldn’t. And this already makes me feel more of a sense of clarity about this.
I still think substantial fuzziness remains. This is mostly just because words like “eminence” could be applied more or less strictly. I think that that’s hard to avoid and maybe not necessary to avoid—people will probably generally agree, and then we can politely squabble about the borderline cases and thereby get a clearer sense of what we collectively think the “line” is.
But I think “people who have had an extraordinary positive impact, and that are often discussed in EA circles (Arkhipov, Zhdanov, etc.)” may require further operationalisation, since what counts as extraordinary positive impact can differ a lot based on one’s empirical, moral, epistemological, etc. views. E.g., I suspect that nil might think Pearce has been more impactful than most people who do have an entry, since Pearce’s impacts are more targeted at suffering reduction. (nil can of course correct me if I’m wrong about their views.)
So maybe we should say something like “people who are widely discussed in EA and who a significant fraction of EAs see as having had an extraordinary positive impact (Arkhipov, Zhdanov, etc.)”? (That leaves the fuzziness of “significant fraction”, but it seems a step in the right direction by not just relying on a given individual’s view of who has been extraordinarily impactful.)
Then, turning back to the original example, there’s the question: Would a significant fraction of EAs see Pearce as having had an extraordinary positive impact? I think I’d lean towards “no”, though I’m unsure, both because I don’t have a survey and because of the vagueness of the term “significant fraction”.
I think there’s a relatively clear sense in which Arkhipov, Borlaug, and similar figures (e.g. winners of the Future of Life Award, names included in Scientists Greater than Einstein, and related characters profiled in Doing Good Better or the 80,000 Hours blog) count as having had an extraordinary positive impact and Pearce does not, namely, the sense in which also Ord, MacAskill, Tomasik, etc. don’t count. I think it’s probably unnecessary to try to specify in great detail what the criterion is, but the core element seems to be that the former are all examples of do-gooding that is extraordinary from both an EA and a common-sense perspective, whereas if you wanted to claim that e.g. Shulman or Christiano are among humanity’s greatest benefactors, you’d probably need to make some arguments that a typical person would not find very persuasive. (The arguments for that conclusion would also likely be very brittle and fail to persuade most EAs, but that doesn’t seem to be so central.)
So I think it really boils down to the question of how core a figure Pearce is in the EA movement, and as noted, my impression is that he just isn’t a core enough figure. I say this, incidentally, as someone who admires him greatly and who has been profoundly influenced by his writings (some of which I translated into Spanish a long time ago), although I have also developed serious reservations about various aspects of his work over the years.
If you mean that the vast majority of EAs would agree that Arkhipov, Borlaug, Zhdanov, and similar figures count as having had an extraordinary positive impact, or that that’s the only reasonable position one could hold, I disagree, for reasons I’ll discuss below.
But if you just mean that a significant fraction of EAs would agree that those figures count as having had an extraordinary impact, I agree. And, as noted in my previous comment, I think that using a phrasing like “people who are widely discussed in EA and who a significant fraction of EAs see as having had an extraordinary positive impact (Arkhipov, Zhdanov, etc.)” would probably work.
And that phrasing also seems fine if I’m wrong about (1), so maybe there’s no real need to debate (1)?
(Relatedly, I also do ultimately agree that Arkhipov etc. should have entries.)
Expanding on (1):
This is mostly due to crucial considerations that could change the sign or (relative) magnitude of the moral value of the near-term effects that these people are often seen as having had. For example:
It’s not obvious that a US-Russia nuclear war during the Cold War would’ve caused a negative long-term future trajectory change.
I expect it would, and, for related reasons, am currently focused on nuclear risk research myself.
But I think one could reasonably argue that the case for this view is brittle and the case for e.g. the extraordinary positive impact of some people focused on AI is stronger (conditioning on strong longtermism).
Some EAs think extinction risk reduction is or plausibly is net negative.
It’s plausible that expected moral impact is dominated by effects on the long-term future, farm animals, wild animals, invertebrates, or similar, in which case it may be both less clear that e.g. Borlaug and Zhdanov had a net positive impact and less clear that it is “extraordinary” relative to the impact of people whose actions were more targeted to helping those populations.
But it’s also because of uncertainties about whether they really had those near-term effects, whether similar things would’ve happened without them, and—at least in Zhdanov’s case—whether they had other near-term effects that may have been very negative. For example:
My understanding is that it’s not actually very clear whether Arkhipov played a crucial role in preventing a launch.
E.g., Baum, de Neufville, and Barrett write “The second captain, Vassily Arkhipov, has been credited with having vetoed the decision to launch the torpedo over the objections of the two other officers (Lloyd 2002). Sources conflict on whether the submarine crew had the authority to launch the torpedo without direct orders from Moscow. The submarine’s communications officer later said in an interview that Arkhipov did play an important role in calming the captain down, but that while there was a danger of an accident or equipment malfunction, they were never close to intentionally launching the nuclear torpedo (Savranskaya 2007).”
Zhdanov also “chaired the Soviet Union’s Interagency Science and Technology Council on Molecular Biology and Genetics, which among its many functions directed the Soviet biological weapons program” (Wikipedia), which I think makes it plausible that his expected impact (evaluated during the Cold War) on the long-term future was very negative.
My more basic point is just that it seems very hard to say with high confidence what actions had net positive vs net negative impacts and how to rank them, and there’s room for reasonable disagreement.
Again, though, I think we can probably sidestep all of this by just saying “people who are widely discussed in EA and who a significant fraction of EAs see as having had an extraordinary positive impact (Arkhipov, Zhdanov, etc.)”.
For those who may want to see the deleted entry, I’m posting it below:
David Pearce is a philosopher and writer best known for his 1995 manifesto The Hedonistic Imperative and the associated ideas about abolishing suffering for all sentient life using biotechnology and other technologies.
Pearce argues that it is “technically feasible” and ethically rational to abolish suffering on the planet by replacing Darwinian suffering-based motivational systems with minds animated by “information-sensitive gradients of intelligent bliss” (as opposed to indiscriminate maxed-out bliss). He stresses that this “abolitionist project” is compatible with a diverse set of values and “intentional objects” (i.e. what one is happy “about”).
In 1998 together with Nick Bostrom, Pearce co-founded the World Transhumanist Association, today known as Humanity+.
Thanks again, nil, for taking the time to create this entry and outline your reasoning. After reviewing the discussion, and seeing that no new comments have been posted in the past five days, I’ve decided to delete the article, for the reasons I outlined previously.
Please do not let this dissuade you from posting further content to the Wiki, and if you have any feedback, feel free to leave it below or to message me privately.
I’m sorry to hear this, Pablo, as I haven’t been convinced that Pearce isn’t relevant enough for effective altruism.
Also, I really don’t see how the persons below have contributed more or are more relevant to effective altruism than Pearce (that is not necessarily to say that their entities aren’t warranted!). May it be correct to infer that at least some of these entries received less scrutiny than Pearce’s nomination?
After reviewing the discussion, and seeing that no new comments have been posted in the past five days, I’ve decided to delete the article, for the reasons I outlined previously.
May I ask why five days since the last comment were deemed enough for proceeding to the deletion? Is this part of the wiki’s rules? (If so, it must be my fault that I didn’t have time to reply in time.)
I also wanted to say that despite the disagreement, I appreacite that the wiki has a team commiteed to it.
>Also, I really don’t see how the persons below have contributed more or are more relevant to effective altruism than Pearce
I tried to outline some criteria in an earlier comment. Chalmers and Hassabis fall under the category of “people who have attained eminence in their fields and who are connected to EA to a significant degree”. Drexler, and perhaps also Chalmers, fall under the category of “academics who have conducted research of clear EA relevance”. Matthews doesn’t fall under any of the categories listed, though he strikes me as someone worth including given his leading role at Future Perfect—the only explicitly EA project in mainstream journalism—and his long-standing involvement with the EA movement.
As the example of Matthews shows, the categories I identified aren’t exhaustive. That was just my attempt to retroactively make sense of the tactic criterion I had followed in selecting these particular people. Despite still not having a super clear sense of the underlying categories, I felt reasonably confident that Pearce didn’t qualify because (1) it seemed that there was no other potential category he could fall under besides that of “EA core figure” and (2) he is not, in my opinion, a core figure in EA. Perhaps the closest situation is that of Aubrey de Grey, who is also a leading figure in an adjacent movement, has had some involvement with the EA movement, but isn’t really central to EA (and is, for that reason, also excluded).
With that said, I’m open to criticism for my selection, and in retrospect I am not confident in all the choices I made. For instance, it now seems to me debatable whether we should have an article on Bryan Caplan. We also don’t have articles on some individuals who some may argue, probably with justification, that deserve one, such as Robin Hanson or Nick Beckstead. I’d be happy to reconsider my decisions in these and other cases. In the case of Pearce (or de Grey), I trust my judgment more, though of course there’s still room for reasonable disagreement.
>May I ask why five days since the last comment were deemed enough for proceeding to the deletion? Is this part of the wiki’s rules?
I wasn’t following an explicit rule: I just got the impression that the discussion had come to an end. Pearce’s entry was online for almost two weeks, so after a five-day period of inactivity I thought it was appropriate to make a decision. I am considering ways to make the process of resolving disputes more structured and less opaque (feedback welcome). If there are any further comments that you would like to make, I am happy to reopen the discussion and consider your arguments.
FWIW, I agree that Hassabis and Drexler meet your proposed criteria and warrant entries, and that Chalmers and Caplan probably do (along with Hanson and Beckstead). But Matthews does seem roughly on par with Pearce to me. (Though I don’t know that much about either of their work.)
I also agree that Pearce seems to be a similar case to de Grey, so we might apply a similar principle to both.
Maybe it’d be useful to try switching briefly from the discussion of specific entries and criteria to instead consider: What are the pros and cons of having more or much more entries (and especially entries on people)? And roughly how many entries on people do we ultimately want? This would be similar to the inclusionism debate on Wikipedia, I believe. If we have reason to want to avoid going beyond like 50 or 100 or 200 or whatever entries on people, or we have reason to be quite careful about adding less prominent or central people to the wiki, or if we don’t, then that could inform how high a “bar” we set.
Chalmers and Hassabis fall under the category of “people who have attained eminence in their fields and who are connected to EA to a significant degree”. Drexler, and perhaps also Chalmers, fall under the category of “academics who have conducted research of clear EA relevance”.
First, I want to make it clear that I don’t question that any of the persons I listed in my previous comment should be removed from the wiki. I just disagree that not including Pearce is justified.
Again, I honestly don’t think that it is true that Chalmers and Drexler are “connected to EA to a significant degree” while Pearce isn’t. Especially Chalmers: from what I know, he isn’t engaged w/ effective altruism, besides once agreeing for being interviewed at the 80,000 Hours podcast.
As for the “attained eminence in their fields” condition, I do see that it may be harder to resolve for Pearce’s case since he isn’t an academic but rather an independent philosopher, writer, and advocate. But if Pearce’s field as suffering abolitionism, then the “attained eminence in their fields” condition does hold, in my view: he both is the founder of the “abolitionist project” and has written extensively on why’s and how’s of the project.
Also, as I mentioned in the original comment proposing the entry, Pearce’s work has inspired many EAs, including Brain Tomasik, the Qualia Research Institute’s Andrés Gómez Emilsson, and the Center for Reducing Suffering’s Magnus Vinding, and the nascent field of welfare/compassionate biology. Also, Invincible Wellbeing research group has been inspired by Pearce’s work as well.
I don’t have any new arguments to make, and I don’t expect anyone involved to change their minds anyway. I only hope it may be worth time of others to contribute their perspectives on the dispute.
And as Michael suggests above, it may be more productive at this point to consider how many entries on EA-relevant persons are desirable in the first place.
Chalmers was involved with EA in various ways over the years, e.g. by publishing a paper on the intelligence explosion and then discussing it at one of the Singularity Summits, briefly participating in LessWrong discussions, writing about mind uploading, interacting (I believe) with Luke Muehlhauser and Buck Shlegeris about their illusionist account of consciousness, etc.
In any case, I agree with you (and Michael) that it may be more productive to consider the underlying reasons for restricting the number of entries on individual people. I generally favor an inclusionist stance, and the main reason for taking an exclusionist line with entries for individuals is that I fear things will get out of control if we adopt a more relaxed approach. I’m happy, for instance, with having entries for basically any proposed organization, as long as there is some reasonable link to EA, but it would look kind of weird if we allowed any EA to have their own entry.
An alternative is to take an intermediate position where we require a certain degree of notability, but the bar is set lower, so as to include people like Pearce, de Grey, and others. We could, for instance, automatically accept anyone who already has their own Wikipedia entry, as long as they have a meaningful connection to EA (of roughly the same strength as we currently demand for EA orgs). Pearce would definitely meet this bar.
I personally feel that the proposal would allow for the inclusion of a number of people (not Pearce) who intuitively should not have their own Wiki entry, so I’m somewhat reluctant to adopt it. More generally, an advantage of having a more exclusionist approach for individuals is that the class of borderline cases is narrower, and so is therefore the expected number of discussions concerning whether a particular person should or should not be included. Other things equal, I would prefer to have few of these discussions, since it can be tricky to explicitly address whether someone deserves an entry (and the unpleasantness associated with having to justify an exclusionist position specifically—which may be perceived as expressing a negative opinion of the person whose entry is being considered—may unduly bias the discussion in an inclusionist direction).
Perhaps voting on cases where there is a disagreement could achieve a wider inclusiveness or at least less controversy? Voters would be e.g. the moderators (w/ an option to abstain) and several persons who are familiar w/ the work of a proposed person.
It may also help if inclusion criteria are more specific and are not hidden until a dispute arises.
I think discussion will probably usually be sufficient. Using upvotes and downvotes as info seems useful, but probably not letting them be decisive.
It may also help if inclusion criteria are more specific and are not hidden until a dispute arises.
This might just be a case where written communication on the internet makes the tone seem off, but “hidden” sounds to me unfair and harsh. That seems to imply Pablo already knew what the inclusion criteria should be, and was set on them, but deliberately withheld them. This seems extremely unlikely.
I think it’s more like the wiki is only a few months old, and there’s (I think) only one person paid to put substantial time into it, so we’re still figuring out a lot of policies as we go—I think Pablo just had fuzzier ideas, and then was prompted by this conversation to make them more explicit, and then was still clearly open to feedback on those criteria themselves (rather than them already being set).
I do agree that it will help now that we have possible inclusion criteria written up, and it would be even better to have them shown more prominently somewhere (though with it still being clear that they’re tentative and open to revision). Maybe this is all you meant?
I didn’t have in mind to sound harsh. Thanks for pointing this out: it now seems obvious to me that that part sounds uncharitable. I do appologise, belatedly :(
What I meant is that currently these new, evolving inclusion criteria are difficult to find. And if they are used in dispute resolutions (from this case onwards), perhaps they should be referenced for contributors as part of the introduction text, for example.
Thanks for the feedback. I have made a note to update the Wiki FAQ, or if necessary create a new document. Feel free to ping me if you don’t see any updates within the next week or so.
Thanks, Pablo. The criteria will help to avoid some future long disputes (and thus save time for more important things), although it wouldn’t have prevented my creating the entry for David Pearce, for he does fit the second condition, I think. (We disagree, I know.)
[Just responding to one specific thing, which isn’t central to what you’re saying anyway. No need to respond to this.]
Again, I honestly don’t think that it is true that Chalmers and Drexler are “connected to EA to a significant degree” while Pearce isn’t. Especially Chalmers [...]
For what it’s worth, I think I agree with you re Chalmers (I think Pearce may be more connected to EA than Chalmers is), but not Drexler. E.g., Drexler has worked at FHI for a while, and the FHI office is also shared by GovAI (part of FHI, but worth listing separately), GPI, CEA, and I think Forethought. So that’s pretty EA-y.
Plus he originated some ideas that are quite important for a lot of EAs, e.g. related to nanotech, CAIS, and Paretotopia.
(I’m writing quickly and thus leaning on acronyms and jargon, sorry.)
I should have been more clear about Drexler: I don’t dispute that he is “connected to EA to a significant degree”. But so is Pearce, in my view, for the reasons outlined in this thread.
(I think it’s weird and probably bad that this comment of nil’s has negative karma. nil is just clarifying what they were saying, and what they’re saying is within the realm of reason, and this was said politely.)
I’m roughly neutral on this, since I don’t have a very clear sense of what the criteria and “bars” are for deciding whether to make an entry about a given person. I think it would be good to have a discussion/policy regarding that.
I think some people like Nick Bostrom and Will MacAskill clearly warrant and entry, and some people like me clearly don’t, and there’s a big space in between—with Pearce included in it—where I could be convinced either way. (This has to do with relevance and notability in the context of the EA Forum Wiki, not like an overall judgement of these people or a popularity contest.)
Some other people who are perhaps in that ambiguous space:
Nick Beckstead (no entry atm)
Elie Hassenfeld (no entry atm, but an entry for GiveWell)
Max Tegmark (no entry atm, but an entry for FLI)
Brian Tomasik (has an entry)
Stuart Russell (has an entry)
Hilary Greaves (has an entry)
(I think I’d lean towards each of them having an entry except Hassenfeld and maybe Tegmark. I think the reason for The Hassenfeld Exception is that, as far as I’m aware, the vast majority of his work has been very connected with GiveWell. So it’s very important and notable, but doesn’t need a distinct entry. Somewhat similar with Tegmark inasmuch as he relates to EA, though he’s of course notable in the physics community for non-FLI-related reasons. But I’m very tentative with all those views.)
… I think the reason for The Hassenfeld Exception is that, as far as I’m aware, the vast majority of his work has been very connected with GiveWell. So it’s very important and notable, but doesn’t need a distinct entry. Somewhat similar with Tegmark inasmuch as he relates to EA, though he’s of course notable in the physics community for non-FLI-related reasons. …
This makes sense to me, although one who is more familiar w/ their work may find their exclusion unwarranted. Thanks for clarifying!
In this light I still think an entry for Pearce is justified, to a degree scientifically grounded proposals for abolishing suffering is an EA topic (and this is the main theme of Pearce’s work). But I’m just one input of course.
Regarding Tomasik, we have different intuitions here: if an entry for Tomasik may not be justified, then I would say this sets a high bar which only original EA founders could reach. (For Tomasik himself is a founder of an EA charity—the Foundational Research Institute / Center on Long-Term Risk—has written extensively on many topics highly relevant to EA, and an advisor at the Center for Reducing Suffering, another EA org.) Anyway, this difference doesn’t probably matter in practice since you added that you lean towards Tomasik’s having an entry.
I agree with you that a Tomasik entry is clearly warranted. I would say that his entry is as justified as one on Ord or MacAskill; he is one of half a dozen or so people who have made the most important contributions to EA, in my opinion.
I will respond to your main comment later, or tomorrow.
As noted, I do lean towards Tomasik having an entry, but “co-founder of an EA org” + “written extensively on many topics highly relevant to EA” + “is an advisor for another EA org”, or 1 or 2 of those things plus 1 or 2 similar things, includes a fair few people, including probably like 5 people I know personally and who probably shouldn’t have their own entries.
I do think Tomasik has been especially prolific and his writings especially well-regarded and influential, which is a big part of why I lean towards an entry for him, but the criteria and cut offs do seem fuzzy at this stage.
As the head of the Forum, I’ll second Pablo in thanking you for creating the entry. While I defer to Pablo on deciding what articles belong in the wiki, I thought Pearce was a reasonable candidate. I appreciate the time you took to write out your reasoning (and to acknowledge arguments against including him).
David Pearce (the tag will be removed if others think it’s not warranted)
Arguments against:
One may see David Pearce much more related to transhumanism (even if to the most altruistic “school” of transhumanism) than to EA (see e.g. Pablo’s comment).
Some of Pearce’s ideas goes against certain established notions in EA: e.g. he thinks sentience of classical digital computers is impossible under the known laws of physics, that minimising suffering should take priority over increasing happiness of the already well-off, that environmental interventions alone, w/o raising individuals’ hedonic setpoints and making these individuals invincible to severe suffering, cannot solve the problem of suffering and achieve sustainable high wellbeing for all.
I also should mention that I’m biased in proposing this tag, as Pearce’s work played a major role in my becoming an EA.
Arguments for:
For over 25 years David Pearce has been researching and writing about addressing the root cause of suffering on the planet using bio/nano/info/robo technology.
Pearce has been raising awareness about, and proposing solutions for, wild-animal suffering at least since 1995.
Several relatively prominent EAs cite Pearce’s work as having major influence on their values, including Brain Tomasik, the Qualia Research Institute’s Andrés Gómez Emilsson, and the Center for Reducing Suffering’s Magnus Vinding. Another recognition of Pearce’s work is his being invited as a speaker for EA Global: Melbourne 2015.
Unlike most other transhumanists, Pearce is antispeciesist and advocates using technology to benefit all sentient life.
Michael is correct that the inclusion criteria for entries of individual people hasn’t been made explicit. In deciding whether a person was a fit subject for an article, I haven’t followed any conscious procedure, but merely relied on my subjective sense of whether the person deserved a dedicated article. Looking at the list of people I ended up including, a few clusters emerge:
people who have had an extraordinary positive impact, and that are often discussed in EA circles (Arkhipov, Zhdanov, etc.)
people who have attained eminence in their fields and who are connected to EA to a significant degree (Pinker, Hassabis, Boeree, etc.)
academics who have conducted research of clear EA relevance (Ng, Duflo, Parfit, Tetlock, etc.)
historical figures that may be regarded as proto-EAs or that are seen as having inspired the EA movement (Bentham, Mill, Russell, etc.)
“core figures” in the EA community (Shulman, Christiano, Tomasik, etc.)
Some people, such Bostrom, MacAskill, Ord, fit into more than one of these clusters. My sense is that David Pearce doesn’t fit into any of the clusters. It seems relatively uncontroversial that he doesn’t fit into clusters 1-4, so the relevant question—at least if one broadly agrees with the approach I’ve taken—is whether he is sufficiently close to the “core” to merit inclusion as part of cluster 5.
As someone who has been involved with EA since its inception and who has (I believe) a reasonably good sense of how central to the movement different people have been, my impression is that Pearce isn’t central enough. If others have different (or similar) impressions, I would encourage them to post them here. We could, alternatively, try to go beyond impressionistic evidence and look at more objective measures, such as citation counts (broadly construed to include not just academic citations but links from the EA Forum and EA Blogs), though conducting that kind of analysis might be time consuming and may not be fully conclusive. Do others have thoughts on how to operationalize the relevant criteria?
FWIW, I think your comment is already a good step! I think I broadly agree that those people who fit into at least one of those clusters should typically have entries, and those who don’t shouldn’t. And this already makes me feel more of a sense of clarity about this.
I still think substantial fuzziness remains. This is mostly just because words like “eminence” could be applied more or less strictly. I think that that’s hard to avoid and maybe not necessary to avoid—people will probably generally agree, and then we can politely squabble about the borderline cases and thereby get a clearer sense of what we collectively think the “line” is.
But I think “people who have had an extraordinary positive impact, and that are often discussed in EA circles (Arkhipov, Zhdanov, etc.)” may require further operationalisation, since what counts as extraordinary positive impact can differ a lot based on one’s empirical, moral, epistemological, etc. views. E.g., I suspect that nil might think Pearce has been more impactful than most people who do have an entry, since Pearce’s impacts are more targeted at suffering reduction. (nil can of course correct me if I’m wrong about their views.)
So maybe we should say something like “people who are widely discussed in EA and who a significant fraction of EAs see as having had an extraordinary positive impact (Arkhipov, Zhdanov, etc.)”? (That leaves the fuzziness of “significant fraction”, but it seems a step in the right direction by not just relying on a given individual’s view of who has been extraordinarily impactful.)
Then, turning back to the original example, there’s the question: Would a significant fraction of EAs see Pearce as having had an extraordinary positive impact? I think I’d lean towards “no”, though I’m unsure, both because I don’t have a survey and because of the vagueness of the term “significant fraction”.
I think there’s a relatively clear sense in which Arkhipov, Borlaug, and similar figures (e.g. winners of the Future of Life Award, names included in Scientists Greater than Einstein, and related characters profiled in Doing Good Better or the 80,000 Hours blog) count as having had an extraordinary positive impact and Pearce does not, namely, the sense in which also Ord, MacAskill, Tomasik, etc. don’t count. I think it’s probably unnecessary to try to specify in great detail what the criterion is, but the core element seems to be that the former are all examples of do-gooding that is extraordinary from both an EA and a common-sense perspective, whereas if you wanted to claim that e.g. Shulman or Christiano are among humanity’s greatest benefactors, you’d probably need to make some arguments that a typical person would not find very persuasive. (The arguments for that conclusion would also likely be very brittle and fail to persuade most EAs, but that doesn’t seem to be so central.)
So I think it really boils down to the question of how core a figure Pearce is in the EA movement, and as noted, my impression is that he just isn’t a core enough figure. I say this, incidentally, as someone who admires him greatly and who has been profoundly influenced by his writings (some of which I translated into Spanish a long time ago), although I have also developed serious reservations about various aspects of his work over the years.
If you mean that the vast majority of EAs would agree that Arkhipov, Borlaug, Zhdanov, and similar figures count as having had an extraordinary positive impact, or that that’s the only reasonable position one could hold, I disagree, for reasons I’ll discuss below.
But if you just mean that a significant fraction of EAs would agree that those figures count as having had an extraordinary impact, I agree. And, as noted in my previous comment, I think that using a phrasing like “people who are widely discussed in EA and who a significant fraction of EAs see as having had an extraordinary positive impact (Arkhipov, Zhdanov, etc.)” would probably work.
And that phrasing also seems fine if I’m wrong about (1), so maybe there’s no real need to debate (1)?
(Relatedly, I also do ultimately agree that Arkhipov etc. should have entries.)
Expanding on (1):
This is mostly due to crucial considerations that could change the sign or (relative) magnitude of the moral value of the near-term effects that these people are often seen as having had. For example:
It’s not obvious that a US-Russia nuclear war during the Cold War would’ve caused a negative long-term future trajectory change.
I expect it would, and, for related reasons, am currently focused on nuclear risk research myself.
But I think one could reasonably argue that the case for this view is brittle and the case for e.g. the extraordinary positive impact of some people focused on AI is stronger (conditioning on strong longtermism).
Some EAs think extinction risk reduction is or plausibly is net negative.
Some EAs think population growth is or plausibly is net negative, e.g. for reasons related to the meat-eater problem or to differential progress.
It’s plausible that expected moral impact is dominated by effects on the long-term future, farm animals, wild animals, invertebrates, or similar, in which case it may be both less clear that e.g. Borlaug and Zhdanov had a net positive impact and less clear that it is “extraordinary” relative to the impact of people whose actions were more targeted to helping those populations.
But it’s also because of uncertainties about whether they really had those near-term effects, whether similar things would’ve happened without them, and—at least in Zhdanov’s case—whether they had other near-term effects that may have been very negative. For example:
My understanding is that it’s not actually very clear whether Arkhipov played a crucial role in preventing a launch.
E.g., Baum, de Neufville, and Barrett write “The second captain, Vassily Arkhipov, has been credited with having vetoed the decision to launch the torpedo over the objections of the two other officers (Lloyd 2002). Sources conflict on whether the submarine crew had the authority to launch the torpedo without direct orders from Moscow. The submarine’s communications officer later said in an interview that Arkhipov did play an important role in calming the captain down, but that while there was a danger of an accident or equipment malfunction, they were never close to intentionally launching the nuclear torpedo (Savranskaya 2007).”
Zhdanov also “chaired the Soviet Union’s Interagency Science and Technology Council on Molecular Biology and Genetics, which among its many functions directed the Soviet biological weapons program” (Wikipedia), which I think makes it plausible that his expected impact (evaluated during the Cold War) on the long-term future was very negative.
My more basic point is just that it seems very hard to say with high confidence what actions had net positive vs net negative impacts and how to rank them, and there’s room for reasonable disagreement.
Again, though, I think we can probably sidestep all of this by just saying “people who are widely discussed in EA and who a significant fraction of EAs see as having had an extraordinary positive impact (Arkhipov, Zhdanov, etc.)”.
For those who may want to see the deleted entry, I’m posting it below:
Thanks again, nil, for taking the time to create this entry and outline your reasoning. After reviewing the discussion, and seeing that no new comments have been posted in the past five days, I’ve decided to delete the article, for the reasons I outlined previously.
Please do not let this dissuade you from posting further content to the Wiki, and if you have any feedback, feel free to leave it below or to message me privately.
I’m sorry to hear this, Pablo, as I haven’t been convinced that Pearce isn’t relevant enough for effective altruism.
Also, I really don’t see how the persons below have contributed more or are more relevant to effective altruism than Pearce (that is not necessarily to say that their entities aren’t warranted!). May it be correct to infer that at least some of these entries received less scrutiny than Pearce’s nomination?
Dylan Matthews
David Chalmers
And perhaps:
Demis Hassabis
K. Eric Drexler
May I ask why five days since the last comment were deemed enough for proceeding to the deletion? Is this part of the wiki’s rules? (If so, it must be my fault that I didn’t have time to reply in time.)
I also wanted to say that despite the disagreement, I appreacite that the wiki has a team commiteed to it.
>Also, I really don’t see how the persons below have contributed more or are more relevant to effective altruism than Pearce
I tried to outline some criteria in an earlier comment. Chalmers and Hassabis fall under the category of “people who have attained eminence in their fields and who are connected to EA to a significant degree”. Drexler, and perhaps also Chalmers, fall under the category of “academics who have conducted research of clear EA relevance”. Matthews doesn’t fall under any of the categories listed, though he strikes me as someone worth including given his leading role at Future Perfect—the only explicitly EA project in mainstream journalism—and his long-standing involvement with the EA movement.
As the example of Matthews shows, the categories I identified aren’t exhaustive. That was just my attempt to retroactively make sense of the tactic criterion I had followed in selecting these particular people. Despite still not having a super clear sense of the underlying categories, I felt reasonably confident that Pearce didn’t qualify because (1) it seemed that there was no other potential category he could fall under besides that of “EA core figure” and (2) he is not, in my opinion, a core figure in EA. Perhaps the closest situation is that of Aubrey de Grey, who is also a leading figure in an adjacent movement, has had some involvement with the EA movement, but isn’t really central to EA (and is, for that reason, also excluded).
With that said, I’m open to criticism for my selection, and in retrospect I am not confident in all the choices I made. For instance, it now seems to me debatable whether we should have an article on Bryan Caplan. We also don’t have articles on some individuals who some may argue, probably with justification, that deserve one, such as Robin Hanson or Nick Beckstead. I’d be happy to reconsider my decisions in these and other cases. In the case of Pearce (or de Grey), I trust my judgment more, though of course there’s still room for reasonable disagreement.
>May I ask why five days since the last comment were deemed enough for proceeding to the deletion? Is this part of the wiki’s rules?
I wasn’t following an explicit rule: I just got the impression that the discussion had come to an end. Pearce’s entry was online for almost two weeks, so after a five-day period of inactivity I thought it was appropriate to make a decision. I am considering ways to make the process of resolving disputes more structured and less opaque (feedback welcome). If there are any further comments that you would like to make, I am happy to reopen the discussion and consider your arguments.
FWIW, I agree that Hassabis and Drexler meet your proposed criteria and warrant entries, and that Chalmers and Caplan probably do (along with Hanson and Beckstead). But Matthews does seem roughly on par with Pearce to me. (Though I don’t know that much about either of their work.)
I also agree that Pearce seems to be a similar case to de Grey, so we might apply a similar principle to both.
Maybe it’d be useful to try switching briefly from the discussion of specific entries and criteria to instead consider: What are the pros and cons of having more or much more entries (and especially entries on people)? And roughly how many entries on people do we ultimately want? This would be similar to the inclusionism debate on Wikipedia, I believe. If we have reason to want to avoid going beyond like 50 or 100 or 200 or whatever entries on people, or we have reason to be quite careful about adding less prominent or central people to the wiki, or if we don’t, then that could inform how high a “bar” we set.
First, I want to make it clear that I don’t question that any of the persons I listed in my previous comment should be removed from the wiki. I just disagree that not including Pearce is justified.
Again, I honestly don’t think that it is true that Chalmers and Drexler are “connected to EA to a significant degree” while Pearce isn’t. Especially Chalmers: from what I know, he isn’t engaged w/ effective altruism, besides once agreeing for being interviewed at the 80,000 Hours podcast.
As for the “attained eminence in their fields” condition, I do see that it may be harder to resolve for Pearce’s case since he isn’t an academic but rather an independent philosopher, writer, and advocate. But if Pearce’s field as suffering abolitionism, then the “attained eminence in their fields” condition does hold, in my view: he both is the founder of the “abolitionist project” and has written extensively on why’s and how’s of the project.
Also, as I mentioned in the original comment proposing the entry, Pearce’s work has inspired many EAs, including Brain Tomasik, the Qualia Research Institute’s Andrés Gómez Emilsson, and the Center for Reducing Suffering’s Magnus Vinding, and the nascent field of welfare/compassionate biology. Also, Invincible Wellbeing research group has been inspired by Pearce’s work as well.
I don’t have any new arguments to make, and I don’t expect anyone involved to change their minds anyway. I only hope it may be worth time of others to contribute their perspectives on the dispute.
And as Michael suggests above, it may be more productive at this point to consider how many entries on EA-relevant persons are desirable in the first place.
Best regards,
nil
Hey nil,
Chalmers was involved with EA in various ways over the years, e.g. by publishing a paper on the intelligence explosion and then discussing it at one of the Singularity Summits, briefly participating in LessWrong discussions, writing about mind uploading, interacting (I believe) with Luke Muehlhauser and Buck Shlegeris about their illusionist account of consciousness, etc.
In any case, I agree with you (and Michael) that it may be more productive to consider the underlying reasons for restricting the number of entries on individual people. I generally favor an inclusionist stance, and the main reason for taking an exclusionist line with entries for individuals is that I fear things will get out of control if we adopt a more relaxed approach. I’m happy, for instance, with having entries for basically any proposed organization, as long as there is some reasonable link to EA, but it would look kind of weird if we allowed any EA to have their own entry.
An alternative is to take an intermediate position where we require a certain degree of notability, but the bar is set lower, so as to include people like Pearce, de Grey, and others. We could, for instance, automatically accept anyone who already has their own Wikipedia entry, as long as they have a meaningful connection to EA (of roughly the same strength as we currently demand for EA orgs). Pearce would definitely meet this bar.
How do others feel about this proposal?
I personally feel that the proposal would allow for the inclusion of a number of people (not Pearce) who intuitively should not have their own Wiki entry, so I’m somewhat reluctant to adopt it. More generally, an advantage of having a more exclusionist approach for individuals is that the class of borderline cases is narrower, and so is therefore the expected number of discussions concerning whether a particular person should or should not be included. Other things equal, I would prefer to have few of these discussions, since it can be tricky to explicitly address whether someone deserves an entry (and the unpleasantness associated with having to justify an exclusionist position specifically—which may be perceived as expressing a negative opinion of the person whose entry is being considered—may unduly bias the discussion in an inclusionist direction).
Perhaps voting on cases where there is a disagreement could achieve a wider inclusiveness or at least less controversy? Voters would be e.g. the moderators (w/ an option to abstain) and several persons who are familiar w/ the work of a proposed person.
It may also help if inclusion criteria are more specific and are not hidden until a dispute arises.
I think discussion will probably usually be sufficient. Using upvotes and downvotes as info seems useful, but probably not letting them be decisive.
This might just be a case where written communication on the internet makes the tone seem off, but “hidden” sounds to me unfair and harsh. That seems to imply Pablo already knew what the inclusion criteria should be, and was set on them, but deliberately withheld them. This seems extremely unlikely.
I think it’s more like the wiki is only a few months old, and there’s (I think) only one person paid to put substantial time into it, so we’re still figuring out a lot of policies as we go—I think Pablo just had fuzzier ideas, and then was prompted by this conversation to make them more explicit, and then was still clearly open to feedback on those criteria themselves (rather than them already being set).
I do agree that it will help now that we have possible inclusion criteria written up, and it would be even better to have them shown more prominently somewhere (though with it still being clear that they’re tentative and open to revision). Maybe this is all you meant?
I didn’t have in mind to sound harsh. Thanks for pointing this out: it now seems obvious to me that that part sounds uncharitable. I do appologise, belatedly :(
What I meant is that currently these new, evolving inclusion criteria are difficult to find. And if they are used in dispute resolutions (from this case onwards), perhaps they should be referenced for contributors as part of the introduction text, for example.
Thanks for the feedback. I have made a note to update the Wiki FAQ, or if necessary create a new document. Feel free to ping me if you don’t see any updates within the next week or so.
Hi nil,
I’ve edited the FAQ to make our inclusion criteria more explicit.
Thanks, Pablo. The criteria will help to avoid some future long disputes (and thus save time for more important things), although it wouldn’t have prevented my creating the entry for David Pearce, for he does fit the second condition, I think. (We disagree, I know.)
[Just responding to one specific thing, which isn’t central to what you’re saying anyway. No need to respond to this.]
For what it’s worth, I think I agree with you re Chalmers (I think Pearce may be more connected to EA than Chalmers is), but not Drexler. E.g., Drexler has worked at FHI for a while, and the FHI office is also shared by GovAI (part of FHI, but worth listing separately), GPI, CEA, and I think Forethought. So that’s pretty EA-y.
Plus he originated some ideas that are quite important for a lot of EAs, e.g. related to nanotech, CAIS, and Paretotopia.
(I’m writing quickly and thus leaning on acronyms and jargon, sorry.)
I should have been more clear about Drexler: I don’t dispute that he is “connected to EA to a significant degree”. But so is Pearce, in my view, for the reasons outlined in this thread.
(I think it’s weird and probably bad that this comment of nil’s has negative karma. nil is just clarifying what they were saying, and what they’re saying is within the realm of reason, and this was said politely.)
+1
To add to arguments for inclusion, here’s an excerpt from an EA Forum post about key figures in the animal suffering focus area.
David Pearce’s work on suffering and biotechnology would be more relevant now than in 2013 due to developments in genome editing and gene drives.
I’m roughly neutral on this, since I don’t have a very clear sense of what the criteria and “bars” are for deciding whether to make an entry about a given person. I think it would be good to have a discussion/policy regarding that.
I think some people like Nick Bostrom and Will MacAskill clearly warrant and entry, and some people like me clearly don’t, and there’s a big space in between—with Pearce included in it—where I could be convinced either way. (This has to do with relevance and notability in the context of the EA Forum Wiki, not like an overall judgement of these people or a popularity contest.)
Some other people who are perhaps in that ambiguous space:
Nick Beckstead (no entry atm)
Elie Hassenfeld (no entry atm, but an entry for GiveWell)
Max Tegmark (no entry atm, but an entry for FLI)
Brian Tomasik (has an entry)
Stuart Russell (has an entry)
Hilary Greaves (has an entry)
(I think I’d lean towards each of them having an entry except Hassenfeld and maybe Tegmark. I think the reason for The Hassenfeld Exception is that, as far as I’m aware, the vast majority of his work has been very connected with GiveWell. So it’s very important and notable, but doesn’t need a distinct entry. Somewhat similar with Tegmark inasmuch as he relates to EA, though he’s of course notable in the physics community for non-FLI-related reasons. But I’m very tentative with all those views.)
This makes sense to me, although one who is more familiar w/ their work may find their exclusion unwarranted. Thanks for clarifying!
In this light I still think an entry for Pearce is justified, to a degree scientifically grounded proposals for abolishing suffering is an EA topic (and this is the main theme of Pearce’s work). But I’m just one input of course.
Regarding Tomasik, we have different intuitions here: if an entry for Tomasik may not be justified, then I would say this sets a high bar which only original EA founders could reach. (For Tomasik himself is a founder of an EA charity—the Foundational Research Institute / Center on Long-Term Risk—has written extensively on many topics highly relevant to EA, and an advisor at the Center for Reducing Suffering, another EA org.) Anyway, this difference doesn’t probably matter in practice since you added that you lean towards Tomasik’s having an entry.
I agree with you that a Tomasik entry is clearly warranted. I would say that his entry is as justified as one on Ord or MacAskill; he is one of half a dozen or so people who have made the most important contributions to EA, in my opinion.
I will respond to your main comment later, or tomorrow.
As noted, I do lean towards Tomasik having an entry, but “co-founder of an EA org” + “written extensively on many topics highly relevant to EA” + “is an advisor for another EA org”, or 1 or 2 of those things plus 1 or 2 similar things, includes a fair few people, including probably like 5 people I know personally and who probably shouldn’t have their own entries.
I do think Tomasik has been especially prolific and his writings especially well-regarded and influential, which is a big part of why I lean towards an entry for him, but the criteria and cut offs do seem fuzzy at this stage.
As the head of the Forum, I’ll second Pablo in thanking you for creating the entry. While I defer to Pablo on deciding what articles belong in the wiki, I thought Pearce was a reasonable candidate. I appreciate the time you took to write out your reasoning (and to acknowledge arguments against including him).
Thank you for appreciating the contribution.
Since Pablo is trusted w/ deciding on the issue, I will address my questions about the decision directly to him in this thread.